ECR 2024 Programme

AWA K E N I N G
V I E N N A / F E B R UA RY 2 8 – M A RC H 0 3
M Y E S R . O RG
THE ANNUAL MEE TING OF
FOREWORD BY THE ESR PRESIDENT

CARLO CATALANO
Dear Colleagues and Friends, Embodying our congress theme, this year’s In Focus
programme will be dedicated to cutting-edge technology and
It is with great pleasure and considerable anticipation that I practices in healthcare. Innovation in Focus will have its own
introduce to you the scientific and educational programme for series of stand-out sessions and an area at the congress
the European Congress of Radiology 2024. focussed on all things future.
This congress will be held at a time where we as radiologists As we continue to explore the evolving landscape of
find ourselves at the dawn of a new era – the era of 'Next radiology, we remain committed to preserving the most
Generation Radiology'. This congress slogan encapsulates our beloved session formats from previous congresses. At ECR
shared vision of not just staying ahead, but defining the future 2024, you'll find a rich variety of sessions, from State-of-the-Art
of our field. Symposia to Image Interpretation Quizzes, from Refresher
Courses to New Horizons sessions.
In the ever-evolving landscape of medical imaging,
radiologists continue to play a pivotal role in transforming Additionally, we will welcome international perspectives from
healthcare. Our journey at ECR 2024 will extend beyond the our three Meets countries this year, Argentina, Italy and
boundaries of the present, where innovation and discovery Singapore, sharing their unique insights and practices. The AI
converge to redefine patient care. We are no longer just Theatre & Exhibition, dedicated to the world of artificial
interpreters of images; we are pioneers of precision medicine, intelligence, and The Cube, dedicated to the world of
harnessing the power of data, technology, and collaboration Interventional Radiology, will also make a come-back, along
to chart a new course in medical care. Now more than ever, we with three outstanding Plenary Lectures from highly
stand at the nexus of information and innovation, serving as accomplished speakers.
the bridge between cutting-edge technology and improved
clinical outcomes – ECR 2024 will underline, embrace, and
celebrate this fact.
Our theme for this year, 'Next Generation Radiology,' reflects

not only our commitment to embracing the future, but also our
recognition of the transformative potential it holds. It signifies
our commitment to exploring the uncharted territories of
medical imaging, where artificial intelligence augments
human expertise, where radiomics unlocks the hidden patterns
within images, and where telemedicine connects us to patients
across borders. It's about embracing the future while
preserving the essence of compassionate patient care that has
always defined our profession.
In this congress, we will delve deep into the frontiers of

radiology, where quantum computing transforms image
processing, where 3D printing revolutionizes personalized
treatment, and where advanced molecular imaging unravels
the mysteries of disease. So, what can you expect at ECR 2024?
2
'Next Generation Radiology' demands collaborative thinking We understand that accessibility is paramount, and our aim is
and diverse perspectives. Thus, we will continue to feature to make 'Next Generation Radiology' accessible to all,
Open Forum sessions, encouraging dynamic discussions and regardless of geographical boundaries. In the spirit of
sharing of ideas among attendees. These sessions will adaptability that the ECR has demonstrated over the past few
spotlight the role of radiology trainees, radiographers, and the years, we remain committed to delivering a high-quality
numerous initiatives of the European Society of Radiology. experience, both onsite and online.
Furthermore, with a continued commitment to interactivity,
ECR 2024 will see a series of hands-on workshops on topics In closing, I would like to extend my heartfelt gratitude to all
ranging from Cardiac CT & MRI, imaging stroke patients, MRI those who have contributed and continue to contribute
of the prostate and MSK Ultrasound. tirelessly to bringing ECR 2024 to fruition. The collective effort
of countless individuals, often working behind the scenes, has
Another traditional and highly popular feature of our congress shaped this unique and memorable congress. Thank you for
will celebrate its tenth congress next year. The Clinical Trials in your dedication and passion.
Radiology session format will mark a decade at ECR with a
special commemorative session. I eagerly look forward to welcoming you to Vienna, where
together, we will explore the limitless possibilities of 'Next
I must also mention our Opening Ceremony at next year’s Generation Radiology'.
congress. Here, we have something very special for you
planned – an evening featuring technical innovations that Warm regards,
have never been seen at ECR before. I do not wish to give
anything away, so all I will say is that you will be truly shocked Prof. Carlo Catalano
by what we have in store for you. ECR 2024 President
Of course, the ECR is not just a meeting for radiologists; it's a

gathering of multidisciplinary professionals. We extend our
warm invitation to colleagues from allied sciences and other
medical specialties who will join us in this journey of discovery
and collaboration.
3
FOREWORD BY THE EFRS PRESIDENT
FOREWORD BY THE EFRS PRESIDENT

ANDREW ENGLAND
Dear colleagues and friends, Like previous congresses there will be a dedicated
Radiographers’ Lounge, EFRS Research Hub and Society
It is my honour as the President of the EFRS to invite you to Booths.
attend the ECR 2024 Congress in Vienna (February 28 to
March 3, 2024). The European Congress of Radiology (ECR) is As the European voice of radiographers, our aim is to make the
the official congress of the EFRS. Over the past months, the ECR 2024 congress the best yet! We look forward to
Radiographers' Scientific Subcommittee, under the leadership welcoming you to Vienna and continuing the journey to
of Dr Jonathan Portelli (Malta), has been working hard to improve professional practice and outcomes for our patients.
prepare a diverse and rich programme for all radiographers.
This will include over 30 dedicated radiographer sessions With best wishes,
containing over 100 presentations. Abstract submission
remains open, and we encourage all radiographers to consider Andrew England
disseminating their research at ECR 2024. EFRS President & ECR2024 Programme Planning Committee Member
4
ECR 2024 SESSION DEFINITIONS
ECR 2024
SESSION DEFINITIONS
Clinical Trials in Radiology Sessions E³ - The Beauty of Basic Knowledge Sessions
In Clinical Trials in Radiology (CTiR) sessions, scientific The E³ - Beauty of Basic Knowledge sessions offer participants
evidence is presented for imaging tests that are very likely to the chance to refresh their knowledge in fundamental topics of
impact a radiologist's clinical practice in the future. Each CTiR imaging. This year, the Beauty of Basic Knowledge sessions
presented during these sessions is accompanied by an focus on sports, hybrid and vascular imaging, allowing anyone
opinionated live discussion by experts in the respective field. from new residents to board-certified radiologists to
strengthen their understanding in these essentials of
E³ - Advanced Courses radiology.
The E³ - Advanced Courses are a series of sessions covering E³ - Young ECR Programme
subjects of great relevance. This year's courses include imaging
of the heart, sustainable technological innovations and image- The E³ - Young ECR Programme is tailored for residents,
guided therapy. Conducted by specialists in their discipline, students, radiographers, and trainee radiographers, designed
the various sessions within each course demonstrate the especially to offer a variety of essential knowledge. An
multiple facets of these topics while complementing one assortment of sessions is provided to ensure a thorough
another to provide an in-depth exploration of the respective examination of the topics, including Basic Sessions, Case-
field. Based Diagnosis Training Sessions, Student Sessions, and the
Radiology Trainees Forum Programme, including the RTF
E³ - ECR Master Classes Highlighted Lectures and the entertaining RTF Quiz.
The E³ - ECR Master Classes focus on continuous professional EFRS Meets Session
development and lifelong learning. The classes are designed
for subspecialised radiologists seeking cutting-edge The 'EFRS Meets' session is organised by the European
information in their particular fields of interest. They are held Federation of Radiographer Societies (EFRS) and provides a
by experts in the field and reflect state-of-the-art knowledge, as platform to introduce and highlight the accomplishments of
well as emerging trends. The contents mostly cover Level III their member societies in the world of radiography. This year's
European Training Curriculum (ETC) learning objectives and session features experts from Italy.
beyond.
ESR at Work Sessions
E³ - European Diploma Prep Sessions
The ESR at Work sessions are organised by bodies of the
The E³ – European Diploma Prep sessions aim to provide a European Society of Radiology to showcase their current
preparation for future European Diploma in Radiology (EDiR) activities.
candidates and are held in close cooperation with the
European Board of Radiology (EBR). They are also suitable for
residents wishing to obtain an overview of the various topics
relevant to imaging and for those preparing for their national
board examinations. The content of the programme reflects
Levels I+II European Training Curriculum for Radiology (ETC)
learning objectives across a two-year cycle.
5
ESR Meets Sessions Joint Sessions
The 'ESR Meets' sessions are an opportunity for the radiological The Joint Sessions are an opportunity to gain a deeper insight
community to gain a greater insight into the innovations and into multidisciplinary collaboration of the ESR with a multitude
perspectives of other nations and societies while also of other disciplines related to the world of medical imaging.
strengthening the bonds between the ESR and its guest The sessions highlight collaborative efforts in regard to science
societies. This year, the ESR is proud to host the countries of and professional issues.
Italy, Argentina and Singapore.
(Junior) Image Interpretation Quizzes
Hands-on Workshops
The Image Interpretation Quizzes, two traditional interactive
The ECR Hands-on Workshops are an excellent opportunity for highlights of every ECR, provide both education and
attendees to practice their skills in a "learning by doing" entertainment. Distinguished radiologists will share their
approach. Topics featured at ECR 2024 are cardiac CT and knowledge and diagnosis strategies with the audience.
MRI, imaging stroke patients, MRI of the prostate and
musculoskeletal ultrasound. Multidisciplinary Sessions
The programme will also include a radiology reporting The Multidisciplinary sessions (MS) aim to promote a
workshop series as well as the i-Violin workshop. multidisciplinary approach to detection and treatment,
integrating radiologists and other clinicians to share their
In Focus Programme expertise.
Amidst the dynamic landscape of medical imaging, New Horizons Sessions

radiologists persist as trailblazers, sculpting the future of
healthcare. But the future of this field is not always clear, and The New Horizons sessions (NH) provide an insight into recent
there are many discussions about the best ways to move developments within a specific area of practice, be that
forward as a strong and united medical discipline. innovations in technique, evolvements within a speciality, or
improvements in disease treatment. Exploring these topics
Through the Innovation in Focus programme, we will look at allows presenters to highlight advancements that may become
how radiology is currently transcending the boundaries of the routine in the near future, or that indicate a whole new area of
present, converging innovation and discovery to reshape the research and clinical application.
landscape of patient well-being. With topics ranging from
sustainability and data-management in radiology through to
Open Forum Sessions
the use of chat bots and integrated diagnostics, Innovation in
Focus will take attendees on a journey to the unexplored
The Open Forum (OF) sessions are an opportunity to discuss
realms of medical imaging and beyond.
important topics. This year's Open Forums are dedicated to
Trainees, Radiographers, and ESR activities.
Taking place in its own unique congress area dedicated to all
things future, the In Focus programme at ECRv 2024 is a must-
see series for all attendees who wish to understand and define Plenary Lectures
the next generation in radiology.
ECR 2024 will feature excellent plenary lectures, presented by
Rick Abramson, Annemiek Snoeckx and Geoffrey D. Rubin.
ISRRT Meets Session
The 'ISRRT Meets' session, hosted by the International Society Professional Challenges Sessions
of Radiographers and Radiological Technologists (ISRRT),
demonstrates the vital role that radiographers play in the The Professional Challenges sessions (PC) aim to provide a
medical imaging profession. This is accomplished through the platform for important issues experienced by those working
exploration of the practice, research, and unique challenges within the field of radiology. Subjects of current relevance,
faced by radiographers. This year's session features experts such as training, education, research, communication, and
from Australia. management, are presented.
6
Refresher Courses State of the Art Symposia
The Refresher Courses (RC) are comprised of a variety of key The State of the Art symposia (SA) are an opportunity for
radiological topics orchestrated by the scientific experts in their field to educate and inform the audience on
subcommittees. The aim of the sessions is not only to refresh essential topics such as anatomical regions, specific diseases,
the audience's knowledge, but also to review, inform, and or particular techniques. Presentations are backed by
offer updates of the various fields through engaging experience, evidence and data of the chosen subjects.
presentations suitable for new radiologists through to
experienced doctors. The Cube
Research Presentation Sessions The Cube, ECR's popular interventional radiology event,
focuses on the following four topics in an engaging and
This year's Research Presentation (RP) sessions will feature the hands-on environment: peripheral, central, oncological, and
best abstracts of each topic submitted for ECR 2024. neurological interventional radiology. It offers a balanced
combination of short lectures from experts in the field and a
Round Table Sessions large number of simulation-based activities at the popular
Simulator Adventure Park, where participants get the chance
The Round Table sessions (RT) feature a panel of experts who to work with devices and simulators under the guidance of
will engage and challenge each other in a live discussion on tutors.
the most relevant topics in radiology.
Transatlantic Course of ESR and RSNA
Special Focus Sessions
The Transatlantic Course (TC) combines the resources of the
The Special Focus sessions (SF) explore cutting edge ESR and RSNA (Radiological Society of North America) to
developments within the world of clinical radiology, tackling present a series of sessions concentrated around a central
these complex matters through in-depth analysis and debate. topic. This year’s focus is on the imaging of neurodegenerative
disorders.
7
Preliminary programme printed on February 16, 2024

subject to change without previous notice
AA 1 - Test session
Date: January 30, 2024 | 08:56 - 19:00
8
ST 1 - Morning Welcome with Carlo Catalano
Categories: Education, General Radiology, Multidisciplinary, Professional Issues

Date: February 28, 2024 | 07:50 - 08:00 CET
Grab your morning coffee and join our studio moderators as they discuss the most exciting highlights of the upcoming day with
Congress President Prof. Carlo Catalano. Make a list of what not to miss and hear his insights on some of the biggest trends currently
rocking the world of radiology.
Moderator:
Mélisande Rouger; Bilbao / Spain
Interview (30 min)

Carlo Catalano; Rome / Italy
9
E³ 24A - Hybrid imaging in oncology (part 1): where is it standard-of-care?
Categories: Hybrid Imaging, Nuclear Medicine, Oncologic Imaging

ETC Level: LEVEL I
CME Credits: 1
Moderator:
Clemens C. Cyran; München / Germany
Chairperson's introduction (5 min)

Hybrid imaging in haematological malignancies (15 min)

Marius E. Mayerhöfer; Vienna / Austria
1. To understand the role of imaging in haematological malignancies therapy guidance.

2. To define the advantages of hybrid imaging compared to conventional imaging.
3. To get acquainted with radiotracers beyond FDG.
Head and neck tumours: hybrid imaging for guiding therapy (15 min)
Andrew Scarsbrook; Leeds / United Kingdom
1. To identify key indications for hybrid imaging in head and neck cancer patients.
2. To list different semi-quantitative PET-based criteria which can be used for assessing treatment response.
3. To become familiar with methods for dealing with indeterminate findings that might cause interpretative difficulties.
PET/CT for lung cancer therapy management (15 min)

Helmut Prosch; Vienna / Austria
1. To become familiar with the strengths and limitations of FDG-PET-CT in lung cancer staging.
2. To understand the role of PET-CT in evaluating treatment response in lung cancer patients.
3. To develop a comprehensive understanding of the latest tracers utilised in lung cancer imaging.
Panel discussion: Hybrid imaging as standard-of-care: accessibility guaranteed? (10 min)
10
RPS 101 - Pre- and post-treatment imaging in cirrhosis and HCC
Categories: Abdominal Viscera, Imaging Methods, Oncologic Imaging

CME Credits: 1
Moderator:
Joanna Podgórska; Warszawa / Poland
Abbreviated liver MRI with second shot arterial phase image to assess the viability of treated HCC (7 min)
Jaewon Han; Busan / Korea, Republic of
Author Block: S. Kim, N. K. Lee, S-B. Hong, J. Han; Busan/KR

Purpose: The purpose of this study was to evaluate the feasibility of abbreviated liver MRI (AMRI) with second shot arterial phase
(SSAP) for the viability of treated HCC.
Methods or Background: We retrospectively identified 95 patients (70 men, 25 women; mean age, 68.7 years) between March
2021 and February 2022 meeting the following inclusion criteria; (1) patients with locoregional therapy for HCC who performed the
modified gadoxetic acid-enhanced liver MRI protocol including the routine dynamic imaging after the first injection of 6 mL and SSAP
imaging after the second injection of 4 mL and (2) available reference standard for tumour viability in treated observation. Two
radiologists independently reviewed two MRI sets: full-protocol MRI and AMRI with SSAP. In the full-protocol MRI set, both reviewers
assigned observations according to the liver imaging reporting and data system treatment response (LR-TR) algorithm (TR-viable, TR-
equivocal, or TR-nonviable). In AMRI with SSAP set, we assigned a abbLR-TR category (abbLR-TR viable, abbLR-TR equivocal, or
abbLR-TR nonviable) according to arterial hyperenhancement in SSAP and Hepatobiliary phase hypointensity, using the modified
version of LI-RADS treatment response algorithm. We compared the diagnostic performance between two MRI sets.
Results or Findings: In 95 patients, 42 patients (44.2%) had viable lesions, and 53 patients (55.8%) had nonviable lesions. There
was no significant difference in sensitivity for treated lesions between full-protocol MRI assigned according to the LR-TR algorithm and
AMRI with SSAP assigned according to abbLR-TR category (Full-protocol, 73.8 % (31/42); AMRI with SSAP, 71.4 % (30/42); p=1.00).
There was also no significant difference in specificity between two protocols (Full-protocol, 98.1 % (52/53); AMRI with SSAP, 96.2 %
(51/53); p=1.00).
Conclusion: The abbLR-TR score in AMRI with SSAP showed the non-inferior diagnostic performance to the full-protocol MRI in
evaluating the viability for the treated HCCs.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by PNUH - 2309-013-131.
The imaging features of LI-RADS nonviable or equivocal lesions in the first follow-up evaluation after TACE for HCC to
predict recurrence (7 min)
Shu-Hang Zhang; Jiangsu / China
11
Author Block: S-H. Zhang, Y-C. Wang; Jiangsu/CN
Purpose: The purpose of this study was to investigate whether the imaging features of lesions evaluated as LI-RADS nonviable or
equivocal during the first post-TACE imaging follow-up can predict recurrence in HCC patients.
Methods or Background: A total of 171 lesions from 149 patients were included in this study. All patients underwent at least three
enhanced magnetic resonance imaging examinations, with the first one conducted before TACE and the second one during the first
follow-up visit the one to two months after TACE. The lesions were divided into two groups, no recurrence during long-term follow-up
(n = 128) and recurrence during long-term follow-up (n = 43). Imaging features evaluated included irregular shape, size, internal
homogeneity, non-smooth margin, arterial peritumoural enhancement, rim enhancement in portal venous phase or delay phase,
peritumoural hyperintensity in T2WI or DWI. Using univariate and multivariate logistic regression analysis to investigate which
imaging features at the first follow-up can predict lesion recurrence at 6-12 months after TACE.
Results or Findings: After univariate and multivariate analysis showed that non-smooth margin (OR, 3.96, 95% CI, 1.66, 9.44, P =
0.002) and peritumoural hyperintensity in T2WI or DWI (OR, 7.74, 95% CI, 3.32, 18.06, P < 0.001) were independent risk factors for
recurrence of LI-RADS nonviable lesions at 6 to 12 months. The area under the ROC curve using these two factors to predict the
recurrence of LI-RADS nonviable lesions at 6 to 12 months was 0.754.
Conclusion: Non-smooth border and peritumoural hyperintensity in T2WI or DWI are imaging features to predict recurrence in HCC
patients with LI-RADS nonviable lesions evaluated on the first post-TACE imaging follow-up.
Limitations: This is a retrospective study where selection bias is an inherent flaw.
It can be included in subsequent research for analysis of patient baseline imaging features.
Ethics committee - additional information: This is a retrospective study. It was approved for retrospective analyses by the
institutional review board of this hospital, with an informed consent waiver.
MRI evaluations of focal liver reactions (FLR) following stereotactic body radiotherapy (SBRT) for small hepatocellular
carcinoma: relationship between MRI features and dose-volumetric parameters (7 min)
Runqian Huang; Guangzhou / China
Author Block: R. Huang; Guangzhou/CN

Purpose: The aim of this study was to assess the MRI features of FLR for patients with small hepatocellular carcinoma (HCC) treated
with stereotactic body radiotherapy (SBRT), and to investigate the relationship between MRI features of FLR and dose–volumetric
parameters.
Methods or Background: Between September 2018 and April 2021, 102 patients with small HCC treated with SBRT were evaluated
according to the entry criteria. After SBRT, all patients underwent regular follow-ups with blood tests and dynamic MR scans at a
median follow-up of more than 18 months. Two radiologists assessed imaging features of FLR in MRI sequences at 3 to 6 months. To
assess the deterioration of hepatic function, we evaluated the ALBI score. To identify the parameters of predicting the radiation-
induced hepatic toxicity, univariate and multivariate analyses were used to select independent risk factors and a nomogram model
was established.
Results or Findings: The MRI features of FLR were divided into three types according to T1WI, T2WI, arterial phase, portal-venous
phase and delayed phase signals: low/high/high/high/high in 27 patients (Type A), low/low/low/low/low in 48 patients (Type B) and
low/low/iso (or high)/iso (or high)/high in 27 patients (Type C). ALBI score at 3 to 6 months after SBRT was the significant factor that
differed between Type A (worse prognosis) and alternatives (Type B and C). Multivariate analysis showed that PTV (P=0.028) and
V5Gy (normal liver volumes receiving from 5Gy) (P=0.003) were the significant dose–volumetric parameters for predicting Type A.
Both factors were used to construct the nomogram model, which showed good performance (AUC=0.772).
Conclusion: The MR features of FLR can reflect the deterioration of the hepatic function for small HCC by SBRT. The PTV and V5Gy
were the significant parameters that predicted the risk of the deterioration of hepatic function for different MR types of FLR.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: No information provided by the submitter.
Ascites in patients with hepatocellular carcinoma undergoing transarterial chemoembolisation: the more, the worse?
(7 min)
Lukas Müller; Mainz / Germany
12
Author Block: L. Müller1, D. Bender1, F. Stöhr1, A. Mähringer-Kunz1, J. Mittler1, A. Weinmann1, R. Klöckner2, F. Hahn1; 1Mainz/DE,
2
Lübeck/DE
Purpose: Preliminary work has shown that portal hypertension plays a key role for the prognosis in patients with hepatocellular
carcinoma (HCC) undergoing transarterial chemoembolisation (TACE). Specifically, the presence of ascites appears to be a strong
negative predictor for these patients. However, it remains unclear whether different ascites volumes influence prognosis. Therefore,
the aim of this work was to investigate the influence of different ascites volumes on survival for patients with HCC undergoing TACE.
Methods or Background: A total of 351 patients with HCC treated at our tertiary referral centre between 2010 and 2020 were
included. In patients with ascites, the fluid was segmented, and the volume quantified by slice-wise addition using contrast-enhanced
CT imaging. The patients with ascites were subclassified using cut-off values previously suggested for patients with liver cirrhosis: A0
= 0ml, A1 ≤ 300ml, and A2 > 300ml ascites. Median overall survival (OS) was calculated.
Results or Findings: Ascites was present in 102 (29.1%) patients. Median OS without ascites was 16.5 months, and therefore
significantly longer than in patients with ascites (6.4 months, P<0.001). Patients were further classified as n(A0)=249(70.9%),
n(A1)=43(12.3%), n(A2)=59(16.8%). The corresponding median OS yielded 16.5, 9.9 and 4.0 months for A0-A2, respectively (overall
P<0.001, pairwise comparison A0 vs A1 P=0.03, A1 vs A2 P<0.001).
Conclusion: Ascites in patients with HCC undergoing TACE is associated with a bad prognosis. Our results indicate an association
between the amount of ascites and OS. Not only the presence but also the amount of ascites is relevant. The quantitative assessment
applied here can help improve clinical decision making.
Limitations: The retrospective design and single centre nature of the study were identified as limitations.
Funding for this study: This study was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation);
project number: 518477942.
Ethics committee - additional information: The study has been approved by the local ethics committee responsible (Rhineland-
Palatinate, Germany).
Adding estimates for central venous pressure boosts performance of non-invasive assessment of portosystemic
gradient prior to TIPS implantation (7 min)
Felix Hahn; Mainz / Germany
1 1 1 1 1 2 1 1 1 2
Author Block: F. Stoehr , L. Müller , T. Loew , C. Labenz , J. Mittler , R. Klöckner , M. B. B. Pitton , F. Hahn ; Mainz/DE, Lübeck/DE
Purpose: We aimed to evaluate the accuracy of non-invasive scores to predict markedly increased portosystemic gradient (PSG) in
patients with clinically significant portal hypertension (CSPH) undergoing TIPS procedure and to further investigate patients without
markedly increased PSG at the time of intervention.
Methods or Background: We included patients who underwent TIPS implantation at our tertiary care center between 2010 and
2022 with CT imaging and complete laboratory workup prior to TIPS and without history of splenectomy. Liver and spleen volume
were automatically assessed using a commercially available in-house trained segmentation software. Scores using CT data by
Iranmanesh et al. and Kihira et al. were assessed with regard to classification accuracy. Moreover, differences between the subgroups
of patients with and without PSG>10mmHg were analysed.
Results or Findings: Preliminary results after assessing 200 patients show a mean PSG prior to TIPS of 16.5±4.8mmHg. Out of these
patients, 22 with CSPH had a PSG≤10mmHg at the time of the TIPS procedure. Both scores showed high sensitivity (89% and 87%
resp.), but poor specificity (36% and 41% resp.), in predicting a markedly increased PSG. ROC-analysis yielded an AUC of 0.67 and
0.70. In the subgroup of patients with low PSG, central venous pressure (CVP) was significantly elevated compared to patients with
high PSG (13.4±5.9mmHg vs 7.7±3.6mmHg, p=0.003), while distribution of portal venous pressure was not significantly different
(p=0.51). Adding IVC diameter as an estimator for CVP in linear regression resulted in a significant AUC increase to 0.78 and 0.79
resp. (p=0.04 and p=0.01).
Conclusion: Including IVC diameter boosted the accuracy of CT morphological scores for assessing elevated PSG, especially
increasing specificity.
Limitations: This was a single-centre, retrospective study.
Ethics committee - additional information: The study was approved by the local ethics committee (Medical Association of
Rhineland Palatinate, Mainz, Germany). Informed consent was waived due to the retrospective nature of the study.
The comparison of preoperative and intraoperative graft volumes in living donor liver transplantation: effect of graft
type on accuracy (7 min)
Hande Özen Atalay; Istanbul / Turkey
13
Author Block: H. Özen Atalay, T. Kanmaz, A. Durur Karakaya; Istanbul/TR
Purpose: The purpose of this study was to conduct a retrospective evaluation of the difference between the preoperative estimated
and the actual intraoperative graft volume detected in living liver transplantation donors in terms of graft types.
Methods or Background: The present retrospective study was performed at a single centre, involving a total of 219 donors. The
analysis of total, right, left lobe and left lateral segment liver volumes were conducted in the venous phase abdominal CT by using
Myrian software. The preoperative graft volumes were compared to the intraoperative graft volumes. Intraclass correlation coefficient,
Blant Altman and one-sample tests were used to calculate correlation coefficient and to detect error of estimation.
Results or Findings: 128 donors underwent right hepatectomy, 81 donors had left lateral segmentectomy and 10 donors had left
hepatectomy. The estimated mean volumes of the right lobe, left lobe, and left lateral segment were 885.39 mL, 402.2 mL, and
241.14 mL, respectively. The mean intraoperative weights of the right lobe, left lobe, and left lateral segment were 820.65 mL, 504.2
mL, and 250.82 mL, respectively. The automated preoperative right lobe volume assessment had a higher correlation with
intraoperative graft weight (r=0,753) than the left lobe and left lateral segment volume assessment (r=0.468 and r=0.007,
respectively). The mean percentages of error in volume estimation were calculated at 12,20%, 23,79% and 68,53% for the right lobe,
left lobe and left lateral segment, respectively.
Conclusion: This study underscores the need for more accurate preoperative volume estimations, particularly for the left lobe and
left lateral segment. These areas showed lower correlation, higher error ratios, suggesting that extra caution is needed when planning
transplantation procedures involving these segments.
Limitations: Our retrospective study included a limited number of patients from a single institution.
Ethics committee - additional information: This study was approved by Koc University Biomedical Research Ethics Committee,
Istanbul/Turkey.
A functional liver imaging score (FLIS) as imaging parameter for predicting post-hepatectomy complication in patients
with liver cirrhosis (7 min)
Yea Hee Ji; Busan / Korea, Republic of
Author Block: Y. H. Ji, S-B. Hong, N. K. Lee, S. Kim; Busan/KR

Purpose: The aim of this study was to evaluate whether functional liver imaging score (FLIS) can predict post-hepatectomy
complication in patients with liver cirrhosis (LC).
Methods or Background: A retrospective review (2017-2020) of patients with LC who underwent gadoxetic acid-enhanced MRI and
hepatectomy was conducted. Univariable and multivariable logistic regression was used to identify clinicopathological and radiologic
findings associated with the development of major complications (Clavien–Dindo classification (CDC)≥III). Receiver operating
characteristic (ROC) curve analysis was performed to determine the cut-off value of FLIS for predicting CDC≥III. Spearman rank
correlation was used to assess the correlation between FLIS and other markers of hepatic function (MELD, ALBI, ICG-R15).
Results or Findings: Of the 106 patients finally included, 12 patients had a CDC≥III. Multivariable analysis showed that only FLIS
independently predicted post-hepatectomy complication (Odds Ratio 0.02; P = 0.01). ROC analysis suggested the FLIS≤4 was the
optimal cut-off for predicting CDC≥III (AUC value 0.94; sensitivity 91.67%; specificity 95.74%; positive likelihood ratio 21.54; and
negative likelihood ratio 0.09). In correlation analysis, FLIS significantly correlated, with weak-to-moderate strength with MELD score
(r = − 0.20, p = 0.045), ALBI score (r = − 0.23, p = 0.02), and ICG-R15 (r = − 0.43, p < 0.001).
Conclusion: In patients with LC, FLIS was an independent predictor of post-hepatectomy complications. FLIS showed excellent
diagnostic performance in predicting post-hepatectomy complications.
Limitations: The study was conducted retrospectively and had a relatively small sample size; therefore, there may be a selection
bias. Nonetheless, we gathered MRI and clinical data consecutively. Additionally, we attempted to prevent inappropriate inclusion or
exclusion. Second, our study was limited to patients with LC who had a high risk for surgery. Additional research is needed to
determine whether this can be applied to patients with chronic liver disease who undergo surgery.
Ethics committee - additional information: This study was approved by our institutional review board. The requirement for
informed patient consent was waived owing to the retrospective nature of the study.
14
RPS 102 - Breast imaging biomarkers
Categories: Artificial Intelligence & Machine Learning, Breast, Oncologic Imaging

CME Credits: 1
Moderator:
Fleur Kilburn-Toppin; Cambridge / United Kingdom
MRI radiomics analysis predicting early recurrence in breast cancer patients who are candidates for neoadjuvant
chemotherapy (7 min)
Charlotte Marguerite Lucille Trombadori; Rome / Italy
Author Block: C. M. L. Trombadori, A. D'Angelo, E. Boccia, L. Boldrini, G. Franceschini, D. Giannarelli, A. Franco, A. Fabi, P. Belli;
Rome/IT
Purpose: The aim of this study was to assess the role of pure radiomic predictive models and combined models with
clinical/radiological variables applied to Magnetic Resonance Imaging (MRI) in predicting early recurrence (ER: disease-free survival
<3 years after surgery) in breast cancer patients undergoing neoadjuvant chemotherapy (NAT). Identifying tools for non-invasive pre-
treatment predictors of clinical outcomes, particularly recurrence, is necessary for better patient stratification and treatment
selection.
Methods or Background: Patients with breast cancer who underwent staging MRI, NAT, and surgery at our centre (from 2012 to
2021) were included. Clinical variables evaluated included pathological complete response, ER, and tumour subtype. Radiological
variables included tumour response according to RECIST criteria. Four breast radiologists reviewed MRI, annotated regions of interest,
and extracted radiomic features. Pure radiomic models and combined models (clinical-radiological, radiological-radiomic, and clinical-
radiomic) were developed. The area under the curve (AUC) was calculated for each model, and the models were compared in terms of
accuracy, sensitivity, and specificity.
Results or Findings: A total of 211 patients were included, with an ER prevalence of 11.34%. Patients with complete or partial
response to NAT and Luminal tumour subtype had a lower likelihood of developing ER (p = 0.001 and p = 0.037, respectively). Two
radiomic features were statistically significant associated with ER: F_cm_2.5D.energy and F_cm_2.5D.joint.entr. The AUC values for
combined models were 0.77 (radiological-radiomic model), 0.68 (clinical-radiomic model), and 0.74 (clinical-radiological model). The
radiological-radiomic model was significantly more accurate in predicting ER than the pure radiological and radiomic models (p<0.001
and p<0.03, respectively).
Conclusion: The radiological-radiomic model, combining radiomic features and RECIST criteria, showed the most promising results in
predicting ER.
Limitations: The small sample size and monocentric nature of the study were identified as limitations.
Ethics committee - additional information: This study was approved by an ethics committee; ID cineca: 6081.
PET/CT radiomics integrated with clinical indexes as a tool to predict Ki67 in breast cancer: a pilot study (7 min)
Cong Shen; Xi'an / China
15
Author Block: C. Shen1, Y. Liao2, X. Yu2; 1Xi'an/CN, 2Shanghai/CN
Purpose: This study aims to assess the predictive value of radiomics features extracted from 18F-FDG PET/CT, in combination with
clinical characteristics, for estimating Ki67 expression in patients with breast cancer.
Methods or Background: A total of 114 patients diagnosed with breast cancer and examined using 18F-FDG PET/CT were included.
Patients were randomly assigned to a training set (n = 79, including 55 cases of Ki67+ and 24 cases of Ki67-) and a validation set (n
= 35, comprising 24 cases of Ki67+ and 11 cases of Ki67-). Thirteen clinical characteristics and 704 radiomics features were
extracted. Feature selection involved univariate logistic analysis, Max-Relevance and Min-Redundancy, least absolute shrinkage and
selection operator regression, and Spearman test. Three models were developed, including the clinical model, the radiomics model,
and the combined mode. Model performance was evaluated using receiver operating characteristic (ROC) curve, and clinical utility
was assessed through decision curve analysis (DCA).
Results or Findings: The N stage, tumour morphology, maximal standard uptake, and the longest diameter were significantly
differed between Ki67+ and Ki67- groups (all P<0.05). Seven radiomics features were selected for the radiomics model. The area
under the ROC curve (AUC) of the combined model in the training and test group was 0.90 (95% CI: 0.82–0.97) and 0.81 (95% CI:
0.64–0.99), respectively. The combined model significantly outperformed both the radiomics model and the clinical model alone (both
P<0.05). The DCA curve demonstrated the superior clinical utility of the combined model compared to the clinical model and
radiomics model.
Conclusion: PET/CT image-based radiomics features combined with clinical features have the potential to predict Ki67 expression in
BC.
Limitations: The retrospective nature of the study and its small sample size were identified as limitations.
Funding for this study: This study was funded by the National Natural Science Foundation of China (No. 82272073), the Key
Research and Development Plan of Shaanxi Province (2023-YBSF-480), and the Natural Science and basic research project of Shaanxi
Province (2023-JC-QN-0903).
Ethics committee - additional information: This retrospective study was conducted at the First Affiliated Hospital of Xi 'an
Jiaotong University (NCT05826197), and the study protocol was approved by the Ethics Committee of Xi 'an Jiaotong University (IRB-
SOP-AF-16).
Automatic breast segmentation-based radiomics for classifying breast composition and detecting neoplastic lesions on
chest CT (7 min)
Giridhar Dasegowda; Little Rock / United States
Author Block: G. Dasegowda1, M. Frölich2, S. Dalal3, S. R. Digumarthy3, P. Kaviani3, L. Karout3, R. Fahimi3, E. Garza Frias3, M. K. Kalra3;
1
Little Rock, AR/US, 2Munich/DE, 3Boston, MA/US
Purpose: The aim of this study was to evaluate if automatic breast segmentation-based radiomics can differentiate between benign
and malignant breast lesions and classify the breast based on tissue composition on contrast-enhanced chest CT.
Methods or Background: Our retrospective study included 882 female patients (mean age 55 ± 13 years) who underwent both
contrast-enhanced chest CT and mammography within one year. Patients with surgical clips, prior breast surgeries, and those with
artifacts projecting over the breast tissue on CT images were excluded. The tissue composition (dense, fibroglandular, fatty) and
BIRADS score reported on mammography examinations were recorded. Furthermore, when suspicious for malignancy, the pathology
report was used as the gold standard for classifying benign and malignant breast tissues. Thin-section CT images (1-1.25 mm) were
reconstructed and processed with a Radiomics software prototype (Frontier, Siemens Healthineers) for segmentation and feature
extractions of the left and right breast (1688 radiomic features) were analysed with multiple logistic regression and area under the
curve for precision-recall curve analysis (R Statistical software).
Results or Findings: Automated segmentation-based radiomics differentiated the breast tissue as dense (n=779), fibroglandular
(n=876), and fatty (n=108) with an 0.90 AUC (p<0.001). Of the 1764 breasts with a BIRADS score and pathology confirmation of
malignancy, there were 1545 benign and 219 malignant breast tissues. To differentiate benign and malignant lesions in all breast
tissue, radiomics had an AUC of 0.78 (p<0.001). To differentiate benign and malignant lesions in fatty and fibro glandular breast
tissue (excluding dense breast tissue), radiomics had an AUC of 0.82 (p<0.001).
Conclusion: Radiomics can reliably differentiate breast tissue composition as well as predict malignancy in fibroglandular and fatty
breast tissues with high accuracy. Radiomics can help assess breast tissues and suspicious breast lesions on contrast-enhanced chest
CT examinations.
Limitations: This was a single centre study.
Ethics committee - additional information: Mass General Brigham IRB approved this study.
Prediction of human epidermal growth factor receptor 2 (HER2) status in breast cancer by mammographic radiomics
features and clinical characteristics: a multicentre study (7 min)
Yalan Deng; Shanghai / China
16
Author Block: Y. Deng, Y. Lu; Shanghai/CN
Purpose: The study aimed to preoperatively evaluate the human epidermal growth factor 2 (HER2) status in breast cancer using
radiomics features extracted from digital mammography (DM) and clinical characteristics.
Methods or Background: This study included a cohort of 1512 Chinese women with invasive ductal carcinoma of no special type
(IDC-NST) from two different hospitals. 1332 from Institution A, used for training and testing the models, and 180 women from
Institution B, as the external validation cohort. The Gradient Boosting Machine (GBM) algorithm was employed to establish a
radiomics model and multiomics model using radiomics features and clinical characteristics. Model efficacy was evaluated by the area
under the curve (AUC).
Results or Findings: The number of HER2-positive patients in the training, testing and external validation cohort was 245 (26.3%),
105 (26.3.8%) and 51 (28.3%) respectively, with no statistical differences among the three cohorts (P = 0.842, Chi-square test). The
radiomics model, based solely on radiomics features, achieved an AUC of 0.814 (95% CI: 0.784-0.844) in the training cohort, 0.776
(95% CI: 0.727-0.825) in the testing cohort and 0.702 (95% CI: 0.614-0.790) in the external validation cohort. The multiomics model,
incorporating radiomics features with clinical characteristics, consistently outperformed the radiomics model with AUC values of 0.838
(95% CI: 0.810-0.866) in the training cohort, 0.788 (95% CI: 0.741-0.835) in the testing cohort, and 0.722 (95% CI: 0.637-0.811) in the
external validation cohort.
Conclusion: Our study demonstrates that a model based on radiomics features and clinical characteristics has the potential to
accurately predict the HER2 status of breast cancer patients across different centres.
Limitations: The limitations were (1) ROIs were manually delineated, (2) the focus was exclusively on the relationship between
radiomics features and HER2 status without analyzing other prognostic factors, and (3) the relationship between imaging equipment
and radiomics model efficacy was not explored.
Funding for this study: Funding for this study was received from:
1. Clinical Research Plan of SHDC (No. SHDC2020CR4069)
2. Medical Engineering Fund of Fudan University (No. yg2021-029)
3. Shanghai Sailing Program (No. 21YF1404800)
4. Youth Medical Talents –Medical Imaging Practitioner Program (No. 3030256001)
5. Shanghai Municipal Science and Technology Major Project (No. 2018SHZDZX01)
Ethics committee - additional information: This study was approved by the Ethics Committee of Fudan University Affiliated
Cancer Hospital.
Performance of radiomic features in STIR sequences in predicting histopathological outcomes of breast cancer (7 min)
Günay Rona; Istanbul / Turkey
Author Block: G. Rona1, M. Arifoğlu1, T. A. Serel2, Ö. Adıgüzel Karaoysal1, Ş. Kökten1; 1Istanbul/TR, 2Isparta/TR
Purpose: This study aimed to investigate the performance of radiomic features in STIR sequences in predicting the results of
histopathological outcomes of invasive breast cancer.
Methods or Background: Patients who underwent MRI before treatment were evaluated retrospectively. Histologic grade, ER, PR,
HER-2, Ki-67 expressions and molecular subtypes were noted. Lesions were manually segmented from STIR sequences in the 3D
Slicer program and volume of interest (VOI) was obtained. Machine learning (ML) analysis was performed using Python 2.3, the
Pycaret library program. Datasets were randomly divided into training (70%) and independent testing set (30%). The performances of
ML algorithms were evaluated by area under curve (AUC), accuracy, recall and precision values.
Results or Findings: 197 patients with a mean age of 50.72±46 (range 28-82) years were included in the study. The mean lesion
size was 23.71±14.86 (5-120) mm. 156 of the patients were luminal A+B (79.2%), 17 were HER-2 positive (8.6%), and 24 were TN BC
(12.2%). 156 (79.2%) of the patients were ER +, 41 (20.8%) were ER -, 126 (63.9%) were PR +, 71 (36.1%) were PR -, 58 (29.4%)
were HER-2 +, 139 (70.6%) were HER-2 -. 43 (21.8%) of the patients were grade 1, 104 (52.8%) were grade 2, and 50 (25.4%) were
grade 3.
The best results were obtained in predicting ER status and luminal A+B tumours. In the test set, AUC, accuracy, recall and precision
values in ER+/- discrimination were 0.7518, 0.8048, 0.9628, and 0.8194, respectively. AUC, accuracy, recall and precision values in
predicting luminal A+B tumours were 0.7229, 0.7958, 0.7958, and 0.6490 respectively.
Conclusion: Radiomic features obtained from STIR sequences have the potential to predict ER receptor status and luminal A+B
tumours.
Limitations: The limitations were that it was a retrospective study and a small patient population.
Funding for this study: No funding was provided for this study.
Ethics committee - additional information: This study was approved with the approval code: 202351425620.
Radiogenomics model based on quantitative spatial heterogeneity for predicting pathological complete response and
prognosis of triple-negative breast cancer (7 min)
Jiayin Zhou; Shanghai / China
17
Author Block: J. Zhou, Y. Chao, Y. Gu; Shanghai/CN
Purpose: The study aimed to characterize the spatial heterogeneity of triple-negative breast cancer (TNBC) on MRI and develop a
radiogenomics model for predicting both pathological complete response (pCR) and prognosis.
Methods or Background: In this prospective study, TNBC patients undergoing neoadjuvant chemotherapy were enrolled as the
radiomics development cohort (n=315); among these patients, those with genetic data were enrolled as the radiogenomics
development cohort (n=98). The external validation cohort (n=50) included patients from the DUKE database. Spatial heterogeneity
was characterized using features from the tumour body, intratumoral subregions, and peritumoral region. Three radiomics models
were developed by logistic regression after selecting features. Two fusion models were developed by further integrating pathological
and genomics features (PRM: pathology-radiomics model; GPRM: genomics-pathology-radiomics model). Model performance was
assessed with the AUC and decision curve analysis. Prognostic implications were assessed with Kaplan‒Meier curves and multivariate
Cox regression.
Results or Findings: For radiomics models, the multiregional model representing spatial heterogeneity (Model 3) exhibited better
pCR prediction with AUCs of 0.87, 0.79, and 0.74 in the training, internal validation, and external validation sets, respectively. GPRM
showed the best performance for predicting pCR in the training (AUC=0.97, P=0.015) and validation sets (AUC=0.93, P=0.019). Model
3, PRM and GPRM could stratify patients by disease-free survival, and a predicted non-pCR was associated with poor prognosis
(P=0.034, 0.001 and 0.019, respectively).
Conclusion: Imaging spatial heterogeneity could effectively predict pCR and prognosis of TNBC. The radiogenomics model could
potentially serve as a valuable biomarker to improve the prediction performance.
Funding for this study: No information provided by submitter.
Ethics committee - additional information: No information provided by submitter.
Prediction of cancer aggressiveness based on breast MRI features (7 min)

Veronica Magni; Milan / Italy
Author Block: V. Magni1, A. Benedek2, A. Colarieti2, F. Sardanelli1; 1Milan/IT, 2San Donato Milanese/IT
Purpose: This study aimed to investigate the value of breast MRI features for the prediction of tumour aggressiveness, considering
the association of Ki67 expression and tumour grade with perilesional oedema, rim enhancement, necrosis sign, and adjacent vessel
sign.
Methods or Background: Patients with histologically confirmed malignant breast lesions at preoperative breast MRI and available
results of final pathology on surgical specimens were included in this retrospective study. Exclusion criteria were incomplete or
suboptimal MRI examinations, incomplete histopathological data, breast implants, and neoadjuvant therapy. Pearson correlation
coefficient was calculated to evaluate the strength of association between variables, dichotomising Ki67 expression as low (when
<20% positive cells) and as high (when ≥20% positive cells). Multivariable binary logistic regression was then performed to identify
significant predictors of Ki67 expression and histological tumour grade.
Results or Findings: Among 50 malignant lesions included in the study, 23/50 (48%) showed high-Ki67 expression, while 27/50
(54%) showed low-Ki67 expression. Seven (14%) lesions were grade 1, 26 (52%) were grade 2, and 17 (34%) were grade 3. Ki67
expression showed a positive association with perilesional edema (ρ=0.729, p<0.001), rim enhancement (ρ=0.382, p=0.006),
necrosis sign (ρ=0.341, p=0.015), and adjacent vessel sign (ρ=0.327, p=0.020). At multivariable binary logistic regression,
perilesional oedema and rim enhancement were significant predictors for high-Ki67 expression, showing odds ratios of 39.7 (p=0.002)
and 13.6 (p=0.040) respectively. Perilesional oedema was significantly correlated with histological tumour grade (ρ=0.465, p<0.001).
Conclusion: Breast MRI features may have the potential to predict tumour aggressiveness, serving as prognostic and predictive
biomarkers usable in clinical practice. The systematic and standardised reporting of these findings in radiological reports should be
encouraged to obtain an initial assessment of tumour biological behaviour.
Limitations: The limitations were that it was a single-centre retrospective study with a small sample size.
Ethics committee - additional information: This study was approved by the Ethics Committee of IRCCS Ospedale San Raffaele.
The protocol code SenoRetro was approved on November 9th, 2017 and amended on April 4th, 2021.
Quantitative radiomic analysis in contrast-enhanced mammography for breast lesions characterisation (7 min)
Gianmarco Della Pepa; Milan / Italy
18
Author Block: G. Della Pepa1, C. Depretto1, W. Carli1, G. Irmici1, E. D'Ascoli1, C. De Berardinis1, C. Cazzella2, D. Ballerini1, G. P.
Scaperrotta1; 1Milan/IT, 2Bergamo/IT
Purpose: This study aimed to investigate the potential of radiomic quantitative texture analysis for characterising breast lesions on
contrast-enhanced mammography (CEM) and correlate them with their biological phenotypes.
Methods or Background: Patients who underwent CEM procedures at our institution since 2018 were considered. Among them, all
CEM-detected malignant lesions, confirmed via core needle biopsy and surgical intervention, were included in our study.
These lesions were firstly subjected to a semi-automatic segmentation, and then 93 radiomic features were extracted for each of
them, using the open-source Python package Py-Radiomics.
The association between each feature and the predetermined endpoints was evaluated through univariate logistic regression analysis.
The correlation was performed either with the singular molecular characteristics: the presence of estrogen (ER) and progesterone (PR)
receptors, HER2 status, Ki67 level either with the specific immunophenotype: Luminal A, Luminal B, HER2+ and Triple Negative (TN).
Results or Findings: In our preliminary results, 86 patients were selected, with a total of 89 breast lesions analysed. The logistic
regression isolated a subset of radiomic features correlating robustly with the biological phenotype. Second-order statistics textural
features of Neighbouring Gray Tone Difference Matrix (NGTDM) demonstrated a stronger correlation with the presence of both ER and
PR receptors, and multiple combinations of them resulted in a better correlation with Luminal A and Luminal B immunophenotype.
The Gray Level Run Length Matrix GLSZM contrast and first-order uniformity both correlate with the TN immunophenotype.
Conclusion: Radiomic quantitative texture analysis of breast lesions on CEM demonstrates promising capability in characterising
biological phenotype. Looking forward, it could lead to the construction of a nomogram to be used in clinical practice, potentially
helping decision-making processes before biopsy.
Limitations: The study is constrained by a limited sample size and by the lack of a distinct validation cohort.
19
RPS 105 - Automatic segmentation techniques
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Multidisciplinary, Research
CME Credits: 1
Moderator:
Liliana Caldeira; Cologne / Germany
Deep learning for segmentation and classification of cardiac implantable electronic devices on chest x-rays (7 min)
Felix Busch; Berlin / Germany
Author Block: F. Busch1, A. Zhukov1, P. Suwalski1, S. Niehues1, D. Poddubnyy1, M. Makowski2, K. K. Bressem1, L. C. Adams2; 1Berlin/DE,
2
Munich/DE
Purpose: The accurate classification of cardiac implantable electronic devices (CIEDs) on chest x-rays is crucial for effective patient
care. The aim of this study was to create an open-access deep learning algorithm capable of both segmenting and classifying CIEDs
on DICOM as well as smartphone-acquired images for bedside use.
Methods or Background: This retrospective study included patients with implantable pacemakers, cardioverter defibrillators,
cardiac resynchronisation therapy devices, and cardiac monitors who had undergone anterior-posterior or posterior-anterior chest
radiography from January 2012 to January 2022 at Charité – University Medicine Berlin. Utilising a U-Net architecture with a ResNet-50
backbone, we developed a model to segment and classify CIEDs based on their manufacturer and model, using both DICOM and
smartphone images. Performance metrics included the Dice coefficient for the segmentation model on the validation set (70-30
training/validation set split) and balanced accuracy for manufacturer and model classification on the test set (70-20-10
training/validation/test set split).
Results or Findings: The study encompassed 897 patients with 2,322 unique chest radiographs featuring 25 CIED models from six
manufacturers. To prevent misclassification of models less represented or not included in the training data, an "other" category was
implemented. Additionally, 11,072 images were captured using five different smartphones. The segmentation algorithm attained an
average Dice coefficient of 0.936 (interquartile range: 0.068), while the classification model achieved an overall accuracy of 0.927
(95% confidence interval (CI): 0.890-0.965) for manufacturer and 0.847 (95% CI: 0.799-0.888) for model classification.
Conclusion: We present a publicly accessible deep learning framework for the high-accuracy segmentation and classification of
CIEDs on chest x-rays. Notably, this research introduces the first classification algorithm specifically designed for accurate CIED model
identification based on both DICOM and smartphone images.
Limitations: The retrospective design of the study and the unequal representation of CIEDs were identified as limitations.
Ethics committee - additional information: This study was approved by an ethics committe; IRB-approval number: EA4/042/20.
Advancing total tumour volume estimation in colorectal liver metastases: development and evaluation of a self-
learning auto-segmentation model (7 min)
Inez M. Verpalen; Amsterdam / Netherlands
20
Author Block: J. I. Bereska1, M. Zeeuw1, L. Wagenaar1, M. G. Besselink1, H. Marquering1, J. Stoker1, Å. Fretland2, G. Kazemier1, I. M.
Verpalen1; 1Amsterdam/NL, 2Oslo/NO
Purpose: Total tumour volume (TTV) assessments have been shown to be prognostic of overall and recurrence-free survival in
patients with colorectal cancer liver metastases (CRLM). However, the labor-intensive nature of these assessments has hampered
their clinical adoption. This study aimed to develop an auto-segmentation model for CRLM on contrast-enhanced portal venous phase
CT scans to facilitate the clinical adoption of TTV assessments.
Methods or Background: We developed a self-learning-based segmentation model to segment CRLM using 760 portal venous
phase CTs (CT-PVP) of 363 patients with 13,739 CRLM from the Amsterdam University Medical Centre. We used a self-learning setup
in which we first trained a teacher model on 99 manually segmented CT-PVPs segmented by three radiologists and combined using
the Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm. The teacher model was then used to segment the
remaining 661 CT-PVPs for training the student model. We used Intraclass Correlation Coefficient (ICC) to compare the TTV obtained
from the student model's segmentations against that obtained from the STAPLE-combined radiologist's segmentations.
Results or Findings: We evaluated the student model in an external test set of 50 CT-PVPs from 35 patients with 72 CRLM from the
Oslo University hospital. The student model reached a DICE similarity score of 0.83 for segmenting CRLM. There was no significant
difference between the student model's DICE scores and interrater DICE scores. The ICC between the student model's and the
STAPLE-combined TTV was 0.97, signifying near perfect agreement.
Conclusion: Segmentation models can provide accurate and efficient assessments of TTV in CRLM patients.
Limitations: Our study's limitations include its retrospective design, lack of global data, and an external test cohort that differs from
the training set, underlining the need for prospective, internationally diverse studies for more robust validation.
Funding for this study: This study was funded by the KWF (project number 14002).
Ethics committee - additional information: The Medical Ethics Review Committee of the Amsterdam UMC, the Regional Ethical
Committee of South Eastern Norway, and the Data Protection Officer of Oslo University Hospital approved this study protocol.
A deep learning-based pipeline for cervical spinal cord segmentation and labelling on heterogeneous T1w brain images
(7 min)
Ricardo Magalhaes; Braga / Portugal
Author Block: R. Magalhaes, A. Liseune, S. van Eyndhoven, T. Billiet, N. Barros, D. Smeets, D. M. Sima; Leuven/BE
Purpose: The aim of this study was to develop a robust and automated deep learning-based method for cervical spinal cord
measurements on T1w brain images.
Methods or Background: Measuring spinal cord (SC) cross-sectional area (CSA) is valuable for monitoring multiple sclerosis (MS),
but challenging in daily clinical practice. We propose a fully automated processing pipeline that performs this measurement in mere
minutes and is robust for a wide range of imaging protocols. The pipeline was developed using T1w brain scans from MS patients, with
ground truth masks generated using an in-house semi-automated pipeline that ensures SC coverage and segmentation quality (187
training, 44 validation). Starting from an input T1w image, the pipeline uses icobrain to perform neck cropping and subsequently
applies three cascaded U-net deep learning models that respectively segment, smooth and label the spinal cord, from which the
measurements are derived. We report performance on an independent data set containing 10 MS subjects with 53 scans from
different scanners.
Results or Findings: Dice scores for the segmentation of the full SC and for labelling vertebrae C1 to C4 were 0.89 and 0.85, 0.87,
0.85 and 0.83, respectively. Intra-scanner measurement reproducibility on the test set was high, with an average relative CSA error of
1.5% (intrascanner) and 4.6% (interscanner).
Conclusion: Trained on a heterogeneous set of T1w brain scans, the pipeline enables reliable and accurate quantification of cervical
SC using standard brain scans, extending icobrain software's capabilities.
Limitations: The method requires brain images covering at least a portion of the cervical spine.
Funding for this study: This study is partly funded by Flanders Innovation & Entrepreneurship (VLAIO) project HeKDiscoMS
(HBC.2021.0500) and by the CLAIMS project, supported by the Innovative Health Initiative Joint Undertaking (JU) under grant
agreement No 101112153. The JU is supported by the European Union's Horizon Europe research and innovation programme and
COCIR, EFPIA, EuropaBio, MedTech Europe, Vaccines Europe, AB Science SA and icometrix NV.
Ethics committee - additional information: The test data set was acquired from 10 MS patients who participated in a study at the
University Hospital Brussels, Belgium. The study was approved by the local ethics committee and all patients signed informed consent
forms.
Development data comes from subjects for which icometrix analysed MR scans as part of clinical practice who had agreed to allow
icometrix to use an anonymised version of the already analysed MR images for post-market research purposes. Icometrix processes
personal data received from the hospitals in conformity with the applicable data protection and privacy legislation.
Multi-organ CT-based automatic segmentation via semi-supervised learning (7 min)

Alejandro Vergara; Valencia / Spain
21
Author Block: A. Vergara, A. Jimenez-Pastor, C. Kerckhaert, A. Alberich-Bayarri; Valencia/ES
Purpose: To overcome the scarcity of manually annotated data sets for multi-organ segmentation, we propose a novel approach that
combines fragmented data sets to train a single model capable of performing multiorgan segmentation.
Methods or Background: Four distinct data sets were collected: AbdomenCT1k (liver, kidneys, pancreas, spleen), CT-org (lungs,
bladder), VerSe (spine), and CTPelvic1K (hips, sacrum), collecting a total of 1543 cases. Each data set was used to train a 2D U-Net
with deep supervision. These submodels generated pseudo-labels for data sources that were not included in their training, resulting in
a combination of both original strong labels and soft labels. Finally, a unified model was trained using all data sets and labels.
Results or Findings: The submodels achieved a mean Dice Score Coefficient (DSC) of 0.91. The final model improved the DSC for
each structure by a mean of 0.15, attaining a maximum DSC of 0.98 for liver and a minimum of 0.78 for bladder.
Conclusion: Our study introduces an innovative method for training a single model using diverse data sources, leveraging a pseudo-
label semi-supervised strategy to achieve robust multiorgan segmentation. This approach enables the generation of a larger
annotated data set from smaller, specialised ones lacking all desired labels. Additionally, employing a unified model, as opposed to
separate models for each data source, offers advantages in terms of reduced inference time and resource efficiency.
Limitations: The accuracy of the pseudo-labels used for training the final model is crucial. Gross errors or inaccuracies in these
labels could propagate through the subsequent stages of training, affecting the final model's performance.
Funding for this study: The IMAS project (High Sensitivity and Low Dose Molecular Imaging) was funded by the Spanish Ministry of
Science and Innovation and European Funds.
External validation of an AI-based system for teeth detection and segmentation in panoramic RX images (7 min)
Marco Bologna; Milan / Italy
Author Block: G. Rubiu, M. Bologna, M. Cè, M. Cellina, E. Liaci, D. Sala, D. Fazzini, M. Alì, A. Bubba; Milan/IT
Purpose: Accurate instance segmentation of teeth in orthopanoramics (OPT) is challenging due to variations in tooth morphology and
overlapping regions. In this study, we validate a model for the instance segmentation and numbering of teeth on OPT images.
Methods or Background: The instance segmentation model was trained using Mask-RCNN architecture. The data for training and
internal validation were taken from the TUFT dental database. The number of predicted labels was 52 (20 deciduous and 32
permanent). The size of training and internal validation and sets were 760 and 190 respectively, and the split was performed
randomly. The model was trained for 300 epochs, using a batch size of 10, a base learning rate of 0.001, and a warmup multistep
learning rate scheduler (gamma=0.1). Data augmentation was performed by changing brightness, contrast, crop, and image size. The
external validation was performed on an external data set of 769 OPT. Only images of patients with permanent teeth were used for
the validation. The following quality metrics were used: sensitivity, positive predicted value (PPV), accuracy, and percentage of cases
without errors.
Results or Findings: On the test set, average sensitivity and PPV were both 98%: The overall accuracy was 97%, while the
percentage of panoramic OPT without any error was 71%.
Conclusion: The robustness of an instance segmentation model for the identification of teeth in OPT images was further evaluated
on an external data set with successful results. This model may help speed up and automate tasks like teeth counting and identifying
specific missing teeth, improving the current clinical practice.
Limitations: A multicentric data set was not considered and could be considered for future studies.
Ethics committee - additional information: This study was approved by the ethics committee of Milan, Area 1 (protocol code
7331/2019).
ACMA-net and unscented Kalman filter-based accurate coronary artery segmentation: an application of deep learning
to computed tomography angiography image (7 min)
Bao Li; Beijing / China
22
Author Block: B. Li, C. Wen, H. Sun, W. Wang, J. Liu, Y. Liu; Beijing/CN
Purpose: Accurate coronary computed tomography angiography (CCTA) image segmentation is a prerequisite for high-precision
reconstruction of three-dimensional (3D) coronary artery models, which can visually demonstrate stenotic lesion information and
develop treatment plans. However, due to the complex structure and small size of coronary arteries, and the interference in image
acquisition, the 3D models reconstructed by existing segmentation technology present insufficient precision. This results in the 3D
coronary artery models being noisy and prone to disconnection. To overcome the challenges in accurate segmentation of coronary
arteries, this study proposes a deep learning-based two-stage algorithm.
Methods or Background: In the first stage, we added an atrous convolution feature fusion module (ACFFM) and a multiaxis
attention module (MAM) to 3D U-Net, called ACMA-Net, to enhance the feature expression ability of the network and effect the
preliminary segmentation of coronary arteries. CCTA images of 323 patients were clinically collected to train the network. In the
second stage, the preliminary segmentation results were skeletonised and endpoint detection was performed. The regions of coronary
artery disconnection were determined by finding minimum distance between the main trunk branch and the endpoint of each
disconnected branch. The disconnected skeleton was repaired after reconnection by the unscented Kalman filter (UKF) algorithm.
Results or Findings: We evaluated the proposed method on the constructed test set of 50 patients, and the Dice and Jaccard scores
were 0.940 and 0.888, respectively, outperforming existing deep learning methods.
Conclusion: This study proposed a coronary segmentation method that effectively reduces the phenomenon of coronary
disconnection and improves the accuracy and continuity of coronary segmentation using a small-size data set. This provides excellent
technical support for the patient-specific 3D demonstration of coronary arteries.
Limitations: More CTA images will be clinically collected to further validate the segmentation method.
Funding for this study: Funding was provided by the National Key Research and Development Program of China (Grant No.
2021YFA1000201), and the National Natural Science Foundation of China (Grant No. 12202022, 11832003, 32271361, 12102014).
Ethics committee - additional information: The Ethics Committee of Peking University People's Hospital approved this study.
Deep learning-based bowel automatic segmentation and visualisation of Crohn's disease using multilabelled
continuous MRE images (7 min)
Li Huang; Guangzhou / China
Author Block: L. Huang1, Z. Zhong2, B. Huang2, S-T. Feng1, X. Li1; 1Guangzhou/CN, 2Shenzhen/CN
Purpose: Recognition of bowel segments from magnetic resonance enterography (MRE) images is quite challenging and time-
consuming due to unclear boundary, shape, size, and appearance variations. We established a publicly available whole bowel
segments MR data set with benchmark results and visualisation.
Methods or Background: We retrospectively collected T2-weighted coronal MRE data from 70 patients with Crohn's disease (CD).
The bowel images per patient were divided into ten segments (stomach, duodenum, small intestine, appendix, caecum, ascending
colon, transverse colon, descending colon, sigmoid colon, and rectum), with fine pixel level annotations labelled by two experienced
radiologists. Then, nnU-Net model, a deep learning-based segmentation method that automatically configures all hyperparameters
based on the data set characteristics, was employed on this data set (training set, n=56; test set, n=14). To reinforce the mutually
exclusive relationship between tags, a topological interaction loss function was utilised. The segmentation algorithm was assessed
using the dice similarity coefficient (DSC).
Results or Findings: Evaluating the performance of bowel segmentation, the mean DSC in the test set was 0.778. Our nnU-Net
method in segmenting digestive tract can achieve DSC of 0.963 ± 0.042 in stomach, 0.886 ± 0.049 in duodenum, 0.936 ± 0.024 in
small intestine, 0.378 ± 0.441 in appendix, 0.598 ± 0.294 in cecum, 0.825 ± 0.131 in ascending colon, 0.819 ± 0.231 in transverse
colon, 0.819 ± 0.186 in descending colon, 0.801 ± 0.220 in sigmoid colon and 0.859 ± 0.130 in rectum, respectively. Segmentation
results with predicted bowel boundary can be shown by two- or three-dimensional visual representation.
Conclusion: We presented a new data set containing labels for all digestive tract segments on MRE images. Accurate deep learning-
based bowel automatic segmentation and visualisation of CD can facilitate the application of artificial intelligence in CD.
Replicability of outcome prediction across IPF patient cohorts based on machine learning features learned without
supervision (7 min)
Jeanny Pan; Vienna / Austria
23
Author Block: J. Pan, J. Hofmanninger, K-H. Nenning, F. Prayer, S. Röhrich, N. Sverzellati, V. Poletti, H. Prosch, G. Langs; Vienna/AT
Purpose: Idiopathic pulmonary fibrosis (IPF) is most common interstitial lung disease. This study validated the transferability of an
outcome prediction method based on previously identified disease patterns on a new cohort.
Methods or Background: Four lung CT patterns associated with disease progression had been previously identified from 74 IPF
patients using unsupervised machine learning. We studied a different multicentre cohort (various manufacturers, slice thickness and
reconstruction kernels) of 164 patients to investigate the transferability of the progression patterns. We tested outcome prediction
based on patterns in a single CT scan, and based on additional pattern changes in subsequent scan pairs. In both experiments,
patients were clustered based on similarities in their progression pattern profiles, and Kaplan-Meier survival curves were analysed for
each cluster to test if the outcome was significantly different.
Results or Findings: Of the 164 patients, 59 died and 16 had transplants before the censoring date, with an average time of 211.73
weeks from the baseline scan, while for the remaining 89, it was 197.34 weeks. Predicting survival outcomes with a single scan profile
yields a hazard ratio (HR) of 5.39 (p<0.01). Consistent with the results on the initial cohort, incorporating the change of pattern
profiles between two scans further improved the prediction, yielding an HR of 6.03 (p <0.01).
Conclusion: The replication of outcome prediction with previously identified progression markers in a new cohort of IPF patients
demonstrated significant predictive value for outcome. The dynamic changes in marker profiles between scans enhanced the hazard
ratio. This underscores the potential of quantitative marker profiles in disease monitoring for IPF patients. Future studies may explore
the broader applicability of the method to other interstitial lung diseases.
Limitations: We did not perform specific analysis to differentiate the impact of centres and manufacturers.
Funding for this study: Funding for this project was received via the FWF, ONSET Project P35189.
Ethics committee - additional information: No information was provided by the submitter.
24
BS 1 - Pearls and pitfalls in paediatric and foetal imaging
Categories: Imaging Methods, Musculoskeletal, Neuro, Paediatric

ETC Level: LEVEL II
CME Credits: 1
Moderator:
Maria Argyropoulou; Ioannina / Greece

Pearls and pitfalls of imaging paediatric skeletal dysplasias (15 min)

Amaka Cynthia Offiah; Sheffield / United Kingdom
1. To become familiar with a pattern recognition approach to the main skeletal dysplasias.
2. To be able to recognise key imaging features of the most common dysplasias.
3. To be aware of imaging findings that may be present in acquired disorders.
Pearls and pitfalls of imaging of the paediatric head trauma (15 min)
Élida Vazquez; Barcelona / Spain
1. To become familiar with age-related patterns of traumatic brain injury.

2. To acquire knowledge about imaging protocols that enhance lesion conspicuity.
3. To be aware of false negatives associated with the imaging modality and timing after trauma.
Pearls and pitfalls of imaging the foetus (15 min)

Eléonore Blondiaux; Paris / France
1. To become familiar with the role of gestational age in evaluating foetal imaging.
2. To acquire knowledge of the role of inadequate imaging protocols applied.
3. To be aware of artefacts that can obscure anatomic details or introduce pseudo-structures.
Questions and answers (10 min)
25
HW 1Pa - Peripheral zone: how to detect typical and not-so-typical tumours
Categories: Genitourinary, Oncologic Imaging

ETC Level: LEVEL II+III
CME Credits: 1
We would like to thank our workshop sponsors. For more information click here.
Please note we will be utilising equipment from certain vendors during the workshop sessions; however, a wide range of alternative
options from other vendors is also available.
In this session participants will be able:

1. To understand the importance of optimal image quality for prostate MRI interpretation.
2. To become familiar with the typical features of PI-RADS 4 and 5 lesions.
3. To develop practical skills in minimising PI-RADS 3 lesions.
Instructors (60 min)

Tristan Barrett; Cambridge / United Kingdom
Francesco Giganti; London / United Kingdom
26
E³ 22A - Hip and pelvis
Categories: Education, General Radiology, Musculoskeletal

ETC Level: LEVEL I+II
CME Credits: 1
Moderator:
Andrea B. Rosskopf; Zurich / Switzerland

Athletic pubalgia and extraarticular soft tissue injuries (25 min)

James Teh; Oxford / United Kingdom
1. To list the possible causes of athletic pubalgia and groin-related pain.

2. To be aware of the spectrum of imaging findings of peri-symphyseal injuries.
3. To be familiar with some of the possible interventions for groin pain.
Hip joint injuries and hip impingement: basic claims for reporting (25 min)
Diana Afonso; Lisbon-Funchal / Portugal
1. To understand the different concepts of hip intraarticular impingement and to understand the mechanisms of labral and chondral
damage.
2. To understand the different concepts of hip extraarticular impingement.
3. To report the essential information to the referring clinician.
Panel discussion (8 min)
27
RC 111 - Post-neurosurgery: what to expect on imaging
Categories: Neuro
CME Credits: 1
Moderator:
Sofie Van Cauter; Holsbeek / Belgium

Sofie Van Cauter; Holsbeek / Belgium
Operative approaches in the skull: common findings and complications (15 min)
Stephen Price; Cambridge / United Kingdom
1. To name three routes of CSF leakage post-operatively.

2. To describe the concept of 'brain shift' using image guidance.
3. To list three potential causes of a post-operative subdural collection.
4. To describe why contrast-enhancing tumours may be left behind following surgery.
Post-neurosurgery artifacts on MRI imaging (15 min)

Torstein Ragnar R. Meling; Copenhagen / Denmark
1. To illustrate imaging features and artefacts caused by hemostatic agents in cranial surgery.
2. To explain the CT and MRI aspect of duraplasty materials and bone flap fixation devices.
3. To identify how to recognise imaging complications resulting from implanted materials.
CSF shunts: post-operative evaluation and complications (15 min)

Joanna Bladowska; Wroclaw / Poland
1. To define the most common types of CSF shunts as well as the application of different imaging methods for the evaluation of shunt
malfunction.
2. To list and describe the most common complications (including mechanical failure, infection, ventricular loculation, overdrainage
and the specific ones related to the shunt type) and discuss the key findings that may be useful for the correct diagnosis.
3. To be able to identify possible outliers and pitfalls on imaging.
Panel discussion: How to improve post-operative reporting in neuroradiology (10 min)
28
RC 116 - Normal variants and mimics in oncologic imaging
Categories: Chest, Genitourinary, GI Tract, Oncologic Imaging

ETC Level: LEVEL I
CME Credits: 1
Moderator:
Inês Santiago; Lisbon / Portugal

Inês Santiago; Lisbon / Portugal
Chest imaging (15 min)

Marie-Pierre Revel; Paris / France
1. To understand how benign pathology may mimic cancer.

2. To learn about normal variants which may resemble cancer.
3. To know how to avoid discrepancies in interpretation.
Gastrointestinal imaging (15 min)

Luis Curvo-Semedo; Coimbra / Portugal
1. To understand how benign pathology may mimic GI cancer.

2. To learn about normal variants which may resemble GI cancer.
3. To know how to avoid discrepancies in interpreting GI lesions.
Female pelvis imaging (15 min)

Lucia Manganaro; Rome / Italy
1. To understand how benign pathology may mimic gynaecological cancer.

2. To learn about normal variants which may resemble gynaecological cancer.
3. To know how to avoid discrepancies in interpreting gynaecological lesions.
Panel discussion: Errors and discrepancies in oncologic imaging (10 min)
29
E³ 25A - Abdominal vasculature
Categories: Vascular
CME Credits: 1
Moderator:
Ulrike I. Attenberger; Bonn / Germany

Atherosclerotic disease and pre-EVAR workup (23 min)

Akos Varga-Szemes; Charleston / United States
1. To discuss indications for EVAR procedures.

2. To discuss the role of CTA and MRA in the EVAR pre-interventional workup.
3. To discuss the elements of the radiology report essential in EVAR pre-interventional workup.
Non-atherosclerotic disease, including compression syndromes (23 min)

Dominika Suchá; Utrecht / Netherlands
1. To discuss the role of CT and MR in imaging suspected or known non-atherosclerotic disease of the abdominal aorta and the
visceral branch arteries.
2. To show examples of non-atherosclerotic disease involving the abdominal aorta and the visceral branch arteries, including
compression syndromes.
3. To discuss the essential elements of the radiology report in patients with suspected or known non-atherosclerotic disease involving
the abdominal aorta and the visceral branch arteries, including compression syndromes.
30
RC 103 - Pathology in and around the heart: from outside to inside
Categories: Cardiac, Imaging Methods, Professional Issues, Vascular

CME Credits: 1
Moderator:
Jens Bremerich; Basel / Switzerland

Jens Bremerich; Basel / Switzerland
Imaging in pulmonary hypertension (PH) (15 min)

Karl-Friedrich Kreitner; Mainz / Germany
1. To outline the challenges for imaging in PH.

2. To discuss CT and MRI used to investigate PH with their strengths and limitations.
3. To update the role of CT and MR imaging in current guidelines.
Imaging in pericardial disease (15 min)

Aparna Deshpande; Leicester / United Kingdom
1. To discuss the various imaging modalities used to investigate pericardial disease with their strengths and limitations.
2. To describe the imaging appearances of various pericardial diseases.
3. To outline the use of MRI and CT in the diagnostic work-up of constrictive pericarditis.
Non-invasive imaging in cardiac arrhythmias (15 min)

Alexis Jacquier; Marseille / France
1. To understand the role of MR in patient with recent onset of severe arrhythmia.

2. To understand the principle of ventricular arrhythmia and the role of MR to guide ablation.
3. To understand the role of MR to predict the risk of ventricular arrhythmia.
Panel discussion: To outline effective imaging strategies with regard to current guidelines (10 min)
31
RC 113 - Considerations on radiation doses and associated risks in case of oncological patients'
pathways: frequency and type of examinations in focus
Categories: EuroSafe Imaging/Radiation Protection, Imaging Methods, Multidisciplinary, Physics in Medical Imaging, Professional
Issues
ETC Level: LEVEL I
CME Credits: 1
Moderator:
Mika Kortesniemi; Hus / Finland

Appropriate imaging in oncology (15 min)

Denis Joseph D'Almada Remedios; Harrow / United Kingdom
1. To learn about the generic justification of imaging modalities used for diagnosis staging, planning and follow-up.
2. To understand how the frequency of examination might be adjusted for specific clinical situations for individual justification.
3. To be aware of the role of radiology in the management of patients with special innovative therapies, e.g. immunotherapy.
Recurrent medical imaging procedures in case of oncological patient's pathways (15 min)
Agnieszka Kuchcinska; Warsaw / Poland
1. To evaluate which malignancies could be associated with the highest level of radiation doses due to medical imaging.
2. To learn about possible cumulative effective dose levels in the case of patients taking part in successful clinical trials: a case study:
adult patients with melanoma.
3. To understand how to use cumulative effective dose concept in optimising planned exposure series during treatment and follow-up:
a paediatric patient with osteosarcoma case study.
Radiation protection issues concerns: cumulative doses and possible radiation risk (15 min)
Marco Brambilla; Novara / Italy
1. To understand the possible risk associated with single medical low-dose exposure and risk associated with a series of recurrent
exposures.
2. To learn about the possibilities of dose tracking systems that help manage patients' cumulative radiation doses.
3. To learn about global discussion and current consensus on the possible approach to optimisation and management of planned
medical exposure series.
Panel discussion: How to choose the appropriate medical imaging modality and frequency of examinations in case of
patients receiving innovative oncological therapy? (10 min)
32
OF 1R - Latest developments in contrast media utilisation
Categories: Contrast Media, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods, Radiographers
CME Credits: 1
This session focuses on the latest developments in contrast media utilisation in medical imaging. This session will feature three talks
by distinguished experts in the field, each shedding light on critical aspects of contrast media use across different imaging modalities,
namely CT, MRI and Ultrasound. This session promises to be a valuable resource for radiographers, radiologists, healthcare
professionals, researchers, and educators seeking to stay at the forefront of contrast media utilisation in medical imaging.
Consequently, attendees will gain insights into the latest trends and breakthroughs in contrast media use, ultimately contributing to
enhanced patient safety and better-informed clinical decisions.
Moderators:
Sonyia Lorraine McFadden; Belfast / United Kingdom
Laura Romanini; Brescia / Italy
Chairpersons' introduction (5 min)

Sonyia Lorraine McFadden; Belfast / United Kingdom
Laura Romanini; Brescia / Italy
CT: how low can we go with modern protocols? (16 min)

Efthimios M. Agadakos; Athens / Greece
MRI: current issues with gadolinium-based agents (16 min)

Peter C Murphy; Cork / Ireland
Ultrasound: indications and protocols for contrast agent administration (16 min)
Barbara Kraus; Wolkersdorf / Austria
Open forum discussion (7 min)
33
RC 117 - Near misses and great pickups in trauma imaging
Categories: Emergency Imaging, Multidisciplinary

ETC Level: LEVEL I
CME Credits: 1
Moderator:
Stefan Wirth; München / Germany

Stefan Wirth; München / Germany
Head and neck injuries beyond the spine (15 min)

Massimo Caulo; Chieti / Italy
1. To discuss common mistakes in the interpretation of head and neck trauma.

2. To describe strategies to minimise errors.
3. To understand the value of developing a search pattern for the detection of injuries.
Chest injuries (15 min)

Inès Saab; Paris / France
1. To review the role of the different imaging modalities in thoracic trauma.

2. To describe strategies to minimise errors.
3. To become familiar with common and no so common traumatic injuries to ensure an accurate diagnosis.
Abdominal and pelvic injuries (15 min)

Michael N. Patlas; Toronto / Canada
1. To discuss common mistakes in the interpretation of abdominal trauma.

2. To analyse factors leading to errors.
3. To summarise the advantages of multi-phase imaging and multi-planar reconstructions for the detection of traumatic injuries.
Panel discussion: Tips and tricks to avoid near misses (10 min)
34
CUBE 1 - New frontiers in IR
Categories: Interventional Radiology

Peripheral IR Day - Topic Coordinator: Dr. Raúl García Marcos
The "Tools of the Trade" session is an innovative session format that introduces the audience to the most important devices (tools)
used in interventional radiology. A specialist leads these sessions, describing the devices and its use, and demonstrating its
application on anatomical phantoms. Participants also have the opportunity to touch and explore these devices, which are circulated
in the audience.
Moderator:
Raúl García Marcos; Valencia / Spain

New frontiers in IR (28 min)

Ruben Eduardo Pacios Blanco; Madrid / Spain
1. To discuss performance improvement in IR.

2. To learn how to take advantage of AI.
3. To discuss new challenges in IR.
35
ST 6 - The Cube - Unconventional Interventional
Categories: Education, Interventional Radiology, Students

Step into the world of Interventional Radiology with The Cube @ ECR. Immerse yourself in hands-on experiences using one of the
most extensive collections of ready-to-use endovascular simulators and flow models. Engage with medical professionals and device
tutors, explore an exclusive 'unboxing' area at the Cube's Adventure Park to interact with and understand IR devices firsthand.
Join us for a comprehensive 4-day programme on neuro, peripheral, central and onco IR, featuring academic content presented by
European IR experts and their teams, covering everything from basic to advanced levels. The Cube offers an exciting blend of
challenges, surprises, and prizes, making it the perfect space for colleagues intrigued by the captivating field of Interventional
Radiology. Come and enjoy the experience!
Moderator:
Interview (30 min)

Miltiadis Krokidis; Athens / Greece
Maximilian De Bucourt; Berlin / Germany
36
NH 2 - The arrival of immunotherapy: new horizons for radiologists
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Molecular Imaging, Oncologic Imaging
ETC Level: LEVEL III
CME Credits: 1.5
Moderator:
Ahmed Ba-Ssalamah; Vienna / Austria

Immunotherapy: primers for radiologists (20 min)

1. To have an overview of the mechanisms of action of immunotherapy on the immune system and on the cancer cells.
2. To learn about the different response patterns and their appearance on cross-section imaging modalities.
3. To understand the limitation of morphological assessment in response evaluation.
Assessing immunotherapy response with functional imaging and radiomics (20 min)
Evis Sala; Rome / Italy
1. To discuss the role of functional imaging in assessing immunotherapy response.

2. To review the added value of radiomics in evaluation of response to immunotherapy.
3. To highlight the potential role of integrated multiomics in immunotherapy response assessment.
Imaging appearances of immunotherapy adverse effects (20 min)

Sandra Baleato Gonzalez; SANTIAGO DE COMPOSTELA / Spain
1. To review the pathophysiology of irAES.

2. To list the most common adverse effects, the incidence, and the median time to onset.
3. To recognise the imaging features of toxicities associated with IT based on different imaging techniques.
Panel discussion: How can imaging assess prognosis and response to immunotherapy? (25 min)
37
RPS 211 - Normal brain development and psychiatric disorders
Categories: Multidisciplinary, Neuro, Research

CME Credits: 1.5
Moderator:
Paul M. Parizel; Perth / Australia
Brain responses to complex addition problems in children, adolescents and adults (7 min)
Asya Istomina; Moscow / Russia
Author Block: A. Istomina1, A. Faber1, M. Ublinskiy1, A. Manzhurtsev1, M. Arsalidou2; 1Moscow/RU, 2Toronto, ON/CA
Purpose: Addition is a basic mathematical operation that is taught first in school, and it is the most popular math operation in
functional magnetic resonance imaging (fMRI) studies. fMRI studies consistently show that solving addition problems elicit fronto-
parietal and cingulo-opercular areas in adults. Results in children is less consistent and no study to our knowledge examines addition
problem-solving with 2-digit and 3-digit problems in children, adolescents, and adults. This current fMRI study highlights
developmental effect in complex addition problem-solving.
Methods or Background: Structural (TR=8.4 ms, matrix=240×222, voxel size=1.0×1.0×1.0 mm; FOV=240×240×170 mm; TE=3.9
ms; flip angle=8°) and functional (TR=2500 ms; TE=35 ms; FOV=230×230×150; 260 measurements per run; voxel
size=3.0×3.0×3.0 mm) brain data of 20 children (9 female, 11-13 years), 20 adolescents (9 female, 14-16 years) and 20 adults (12
females; 18–29 years) were acquired using a magnetic resonance Philips Achieva dStream 3.0T scanner. Participants performed 2-
digit and 3-digit addition tasks in a block design that lasted 32 seconds each. Participants were instructed to provide as many correct
answers as possible. Data processing and analysis were carried out using AFNI software.
Results or Findings: Solving difficult addition problems elicits activity in common and distinct regions across age groups. Common
areas include middle and superior frontal gyri, inferior parietal lobule, insular cortex. Adults also engaged the bilateral cerebellar and
supplementary motor areas, whereas in children and adolescents, cerebellar activation was in the right hemisphere.
Conclusion: Results suggest that cognitive strategies may not be fully developed in children and adolescents. The agreement in
brain areas among adults, contrasted with their absence in children and adolescents, offers insights into neural processing during
math tasks. This highlights developmental distinctions in brain function and cognitive capacities across these groups.
Limitations: Motion is the most prevalent artifact, especially in pediatric samples.
Funding for this study: The study was funded by the Brain Program of the IDEAS Research Center.
Ethics committee - additional information: The local ethics committee approved all materials and procedures.
Association of variability in body size with imaging metrics of brain structural health (7 min)
Jing Sun; Beijing / China
38
Author Block: J. Sun, H. Lv, Z. Wang; Beijing/CN
Purpose: This study aimed to examine the associations of long-term variability in body mass index (BMI) and waist-to-hip ratio (WHR)
with neuroimaging metrics that approximate brain structural health.
Methods or Background: This cohort study recruited 1,114 participants aged 25-83 years from the subset of brain MRI acquisition
within Kailuan study from 2020 onward. BMI and WHR variability during 2006-2018 were calculated as the average slope
incorporating seven biennial measurements separately. We investigated the associations of BMI and WHR variability with MRI markers
of brain tissue volumes, white matter microstructural integrity, white matter hyperintensity, and cerebral small vessel diseases.
Results or Findings: Progressive weight gain during follow-up was associated with decreased global fractional anisotropy (beta =
-0.19, 95% CI -0.35 to -0.02) and increased mean diffusivity (beta = 0.16, 95% CI 0.01 to 0.31) and radial diffusivity (beta = 0.18,
95% CI 0.03 to 0.33). Weight loss was associated with lower periventricular white matter hyperintensity load (beta = -0.24, 95% CI
-0.47 to -0.02) and lower risk of moderate-to-severe basal ganglia enlarged perivascular spaces (OR = 0.41, 95% CI 0.21 to 0.83). In
adults with central obesity, WHR loss was associated with larger volumes of gray matter (beta = 0.50, 95% CI 0.11 to 0.89),
hippocampus (beta = 0.62, 95% CI 0.15 to 1.09), and parahippocampal gyrus (beta = 0.85, 95% CI 0.34 to 1.37).
Conclusion: Progressive weight gain is associated with impaired white matter integrity. Weight and WHR losses are associated with
improved general brain structural health. The present study contributes to a better understanding of the integrated associations
between variations in general and central adiposity and brain structural health, especially in the early stages of the prevention of
adiposity-related Alzheimer’s disease and dementia.
Limitations: None
Funding for this study: This study was supported by grants 62171297 and 61931013 from the National Natural Science Foundation
of China (No. [2015] 160 from the Beijing Scholars Program, No. ZYLX202101 from Beijing Hospitals Authority Clinical Medicine
Development of Special Funding Support, No. 2021-135 from Beijing Municipal Health Commission-Beijing Key Clinical Discipline
Funding.
Ethics committee - additional information: The Medical Ethics Committee of Kailuan General Hospital approved the META-KLS
study (IRB number: 2021002). Written informed consent was obtained from all the participants.
Resolving heterogeneity in post-traumatic stress disorder using individualised structural covariance network analysis
(7 min)
Li Chen; Chengdu / China
Author Block: L. Chen1, N. Pan1, L. Li2, G. J. Kemp3, S. Wang1, Q. Gong1, X. Suo1; 1Chengdu/CN, 2Changsha/CN, 3Liverpool/UK
Purpose: The heterogeneity of post-traumatic stress disorder (PTSD) is an obstacle to both understanding and therapy, and this has
prompted a search for internally-homogeneous biological subgroups within the broad clinical diagnosis. Our goal was to do this using
the individual differential structural covariance network (IDSCN).
Methods or Background: We constructed cortical thickness-based IDSCN using T1-weighted images of 89 individuals with PTSD
(mean age 42.8 years, 60 female) and 89 demographically-matched trauma-exposed non-PTSD (TENP) controls (mean age 43.1
years, 63 female). The IDSCN metric quantifies how the structural covariance edges in a patient differ from those in the controls. We
examined the structural diversity of PTSD and variation among subtypes using a hierarchical clustering analysis, the correlation
between IDSCN and individually clinical symptoms, and sex-by-subtype interactions.
Results or Findings: PTSD patients exhibited notable diversity in distinct structural covariance edges, but mainly affecting three
networks: default mode, ventral attention and sensorimotor. These changes predicted individual PTSD symptom severity. We
identified two neuroanatomical subtypes: the one with greater PTSD symptom severity showed lower structural covariance edges in
the frontal cortex and between frontal, parietal, and occipital cortex regions which are functionally implicated in selective attention,
response selection and learning tasks. And there was no significant PTSD subtype x sex interaction.
Conclusion: Deviations in structural covariance in large-scale networks are common in PTSD, but fall into two subtypes. This work
sheds light on the neurobiological mechanisms underlying the clinical heterogeneity, and may aid in personalised diagnosis and
therapeutic interventions.
Limitations: First, this was one dataset and a single modality; future studies could use multimodal imaging and validation in an
independent dataset. Second, it was a cross-sectional design; longitudinal research would be needed to establish whether these two
PTSD subtypes showed different responses to treatment.
Funding for this study: This work was supported by the National Natural Science Foundation of China (Grant Nos. 82001800,
81761128023, 81820108018, and 82027808), and the Young Elite Scientists Sponsorship Program by CAST (Grant No.
2022QNRC001).
Ethics committee - additional information: The study protocol was reviewed and approved by the Sichuan University Research
Ethics Committee. Each participant provided full-informed written consent. This study conforms to the provisions of the Declaration of
Helsinki.
Default-mode network overlap with the mirror-neuron system activation task in trauma-exposed veterans (7 min)
Oleksandr Zakomornyi; Kyiv / Ukraine
39
Author Block: O. Zakomornyi, L. Myroniak, I. Kobzar, O. Omelchenko, I. Dykan; Kyiv/UA
Purpose: The purpose of this study was to detect functional connectivity of the mirror-neuron system and the overlap with the
default-mode network as a potential marker of PTSD in trauma-exposed veterans.
Methods or Background: Fifteen trauma-exposed veterans (mean PCL5 value = 24) were studied with the fMRI using the event-
related paradigm of different right-hand motion types observation with subsequent execution. A 3T scanner was used with the 16ch
head-neck coil. Invivo fMRI stimulation hardware with mirror goggles and a display screen was used. EPI pulse sequence with
TR/TE=3000/100. FSL software library was used for the analysis.
Results or Findings: Mirror neuron system fMRI activation was detected during the hand motion observation periods. In particular,
activation of the posterior inferior frontal gyrus of the right and left hemispheres and extensive activation of inferior parietal lobules
bilaterally were observed. In addition to the described activation of the precuneus, the inferior parietal lobules region was detected
during the described motion observation phases. Default-mode network deactivation was not detected in its typical regions.
Conclusion: "Social-brain" networks like default-mode network and mirror-neuron system were detected to function coherently, while
the deactivation of the default-mode network was decreased, which might be the imaging signature of PTSD in veterans and might be
used for the detection of the PTSD severity markers.
Limitations: The study applies to the patient population with subacute trauma exposure period.
Funding for this study: No additional funding was received.
Ethics committee - additional information: Local ethics committee approval was received before the study.
Abnormal reward circuits in adolescents with first-episode non-suicidal self-injury based on subcortical volume and
lateralisation index (7 min)
Nian Liu; Nanchong / China
Author Block: R. Yu, N. Liu; Nanchong/CN

Purpose: This study aims to investigate the structural characteristics of the reward circuit in adolescents with non-suicidal self-injury
(NSSI) in the absence of comorbidities using subcortical volume and lateralisation index(LI). The objective is to elucidate the
pathophysiological mechanisms of NSSI and provide structural evidence for NSSI.
Methods or Background: This study prospectively enrolled 30 primary NSSI patients and 29 healthy controls (HC). Clinical
characteristics included the self-rating anxiety scale (SAS), Beck Scale for Suicidal Ideation (BSI), Brief Assessment of Cognition in
Schizophrenia (BACS), and Ottawa Self-Injury Inventory (OSI). Subcortical volume segmentation was conducted using Freesurfer
software. LI quantified differences in left and right subcortical volume. Statistical analysis of clinical features, subcortical volume, and
LI were performed using wilcoxon rank-sum and independent sample t-tests. Pearson or Spearman correlation was used to explore
relationships between subcortical volume or LI and clinical characteristics.
Results or Findings: The NSSI group had higher SAS scores compared to the HC group (P<0.05). BACS scores in the NSSI group
were lower than in the HC group (P<0.05). Compared to the HC group, the NSSI group exhibited reduced volume in the left
accumbens and putamen, along with an increased LI of the accumbens (p<0.05). In the NSSI group, left putamen volume was
negatively correlated with OSI scores (p=0.045), left accumbens volume was positively correlated with BACS scores (p=0.033), and LI
of the accumbens was negatively correlated with OSI scores (p=0.044).
Conclusion: Our study reveals that in primary NSSI adolescents, subcortical volume of the left putamen and left accumbens are
diminished, with LI indicating potential left-sided vulnerability. These findings provide a new sight and structural evidence for the
reward circuits mechanism of NSSI.
Limitations: This is a cross-sectional study, so the exploration of the efficacy is limited.
Funding for this study: The Bureau of Science & Technology and Intellectual Property Nanchong City (NO. 22SXQT0305). The
Opening Project of Functional and Molecular Imaging Key Laboratory of Sichuan Province (NO. SCU-HM-202307001).
Ethics committee - additional information: Ethical approval was obtained from the ethics board of the Affiliated Hospital of North
Sichuan Medical College(2022ER406-1).
The association among individual grey matter volume of frontal-limbic circuitry, fatigue susceptibility, and comorbid
neuropsychiatric symptoms following COVID-19 (7 min)
Wenrui Bao; Xi'an / China
40
Author Block: W. Bao, Z. Luo, M. Zhang, X. Niu; Xi'an, China/CN
Purpose: Our study aimed to identify brain structural markers of post-COVID fatigue and to further investigate the correlation
between individual grey matter volume, fatigue susceptibility, and concomitant neuropsychiatric symptoms.
Methods or Background: Individuals following a mild SAS-COV2 infection (COV+) underwent neuropsychological measurements (n
= 335) and MRI scans (n = 271) within one month, and 191 (70.5%) of the individuals were followed up at 3three months after
infection. 67 healthy controls (COV-) completed the same protocol on recruitment to the study.
Results or Findings: The analysis of the whole brain revealed that there was no difference in grey matter volume during the acute
phase between the study groups. In the COV+ group, however, fatigue severity was linked only to the volume of the right dorsal
anterior cingulate cortex (dACC) and dorsolateral prefrontal cortex (DLPFC), which are both part of the frontal-limbic system.
Furthermore, it should be noted that fatigue acted as a mediator in the associations between the volume of identified brain regions
and COVID-related sleep, PTSD, and anxiety. It is important to highlight that the initial structural differences in the DLPFC observed
during the early stage can predict fatigue symptoms three months after COVID-19 infection.
Conclusion: Our findings provide novel evidence on the neuroanatomical basis of fatigue vulnerability and emphasize fatigue as an
important link from GMV in frontal-limbic regions to comorbid neuropsychiatric symptoms in the early recovery after infection. This
may facilitate public health interventions by early identifying individuals at risk of developing post-COVID fatigue and implement
treatment targeted at fatigue, further reducing the long-term adverse effects of the disease.
Limitations: Our data were collected in adults with mild COVID-19 who were in good health, which may limit the generalisability of
the findings to a broader population.
Funding for this study: This study was funded by the Shaanxi Province health brain science and imaging new technology research
and innovation platform-2023TD-09.
Ethics committee - additional information: This study was approved by the Ethics Committee of The First Affiliated Hospital of
Xi'an Jiaotong University: XJTU1AF2023LSK-013.
In-vivo mapping of the brain network connectivity alterations among adolescents with attention-deficit hyperactivity
disorder by resting-state functional magnetic resonance imaging (7 min)
Rania Essam-el-dein Mohamed; Cairo / Egypt
Author Block: R. E-e-d. Mohamed; Tanta/EG

Purpose: This study aimed to investigate the abnormal changes in the functional brain activity and connectivity of resting-state brain
networks (RSNs), in the resting-state conditions, among adolescents with different sub-types of ADHD.
Methods or Background: The current study included 60 participants; Group A included 30 adolescents with ADHD and Group B
included 30 normal typically developing controls. The ages of all the participants ranged between 13 and 18 years. All participants
underwent a neuropsychological assessment by applying the DSM-5 criteria to them; and rs-fMRI examinations, including blood
oxygenation level-dependent BOLD-MRI and pseudo-continuous pulsed arterial spin labeling MRI (pCASL-MRI).
Results or Findings: In comparison with the normal controls, the adolescents with ADHD showed significant differences in the
resting-state functional connectivity of the resting-state networks including dorsal attention network (DAN) (p<0.001), default mode
network (DMN) (p=0.003), auditory network (AN) (p<0.001), sensorimotor (SMN) (p=0.019), executive control network (ECN)
(p=0.021), and salience network (SN) (p<0.014). Noteworthy, in comparison to adolescents with ADHD of the inattentive type, this
study revealed that adolescents with ADHD of the combined sub-type showed clusters with higher regional homogeneity (ReHo)
values in the right cerebellum and left middle temporal gyrus than those with ADHD of the inattentive sub-type.
Conclusion: The rs-FMRI with its techniques including rs-BOLD-MRI and pCASL-MRI can be considered an advantageous choice for
clinical applications, particularly in adolescents with different sub-types of ADHD by virtue of their clinical convenience and being non-
invasive as it can provide an in-vivo imaging tool for mapping the brain activity in the resting state condition.
Limitations: The small size of the studied sample might be a limitation for the study.
Ethics committee - additional information: As per international standard or university standard written ethical approval has been
collected and preserved by the author(s). As per international standards or university standard, parental’ written consent has been
collected and preserved by the author(s).
Network controllability developmental patterns in first-episode drug-naïve schizophrenia (7 min)

Siyi Li; Chengdu / China
41
Author Block: S. Li, W. Zhang, S. Lui; Chengdu/CN
Purpose: Schizophrenia typically manifests during late adolescence and early adulthood. Studying early-stage patients is vital for
understanding brain function without the influence of antipsychotic medications. While previous research has identified brain region
and network deficits, little attention has been given to network controllability. Controllability is a predictor of the brain's capacity to
transition between different states, with modal controllability being a phenotype associated with substantial changes in brain states.
This study aims to investigate network controllability deficits in the early stages of schizophrenia while controlling for the influence of
antipsychotic medication.
Methods or Background: The study included 135 drug-naïve first-episode schizophrenia patients aged 16-30 and 119 matched
healthy controls. Resting-state fMRI data were acquired using a 3-Tesla GE scanner. DPASF software processed data and computed
voxel-wise functional connectivity. Linear Mixed-Effects Models incorporated regional modal controllability and factors including
controllability, category (schizophrenia or healthy), gender, age, education level, and category interactions. Inter-group comparisons
assessed regional modal controllability and age-related changes.
Results or Findings: Schizophrenia patients exhibited significantly lower regional modal controllability in the left inferior parietal
gyrus compared to healthy controls (p=0.02, corrected). As schizophrenia patients aged, their regional modal controllability
decreased, whereas healthy controls showed increased modal controllability with age. A significant difference in age-related changes
in regional modal controllability emerged between diagnostic groups (p=0.009, corrected).
Conclusion: This study uncovers abnormal network development patterns in early-stage schizophrenia marked by reduced regional
modal controllability, notably in the left inferior parietal gyrus. Understanding network controllability in schizophrenia's early stages is
crucial and may inform treatment and intervention strategies.
Limitations: The study did not include long-term follow-up data to track how network controllability deficits evolve over time in early-
stage schizophrenia. Longitudinal data could provide a more comprehensive understanding of disease progression.
Funding for this study: This study was funded by the Miaozi Project in Science and Technology Innovation Program of Sichuan
Province: Grant No. 2021028.
Ethics committee - additional information: This study was approved by the Ethics Committee of West China Hospital, Sichuan
University.
Altered cerebellar effective connectivity in first-episode schizophrenia and their long-term changes after treatment (7
min)
Xia Wei; Chengdu / China
Author Block: X. Wei1, H. Cao2, C. Luo1, Q. Zhao1, C. Xia1, Z. Li1, Z. Liu1, Q. Gong1, S. Lui1; 1Chengdu/CN, 2Manhasset, NY/US
Purpose: We aimed to the find the detailed directional connectivity disruption of cerebellar cortex.
Methods or Background: Cerebellar functional dysconnectivity plays a key role in schizophrenia (SZ) pathology. One hundred and
eighty drug naïve first-episode patients with SZ (54 were reassessed after one-year treatment) and 166 healthy controls (HCs) were
included. Their resting-state functional magnetic resonance images were used to perform Granger causal analysis, in which nine
functional systems of the cerebellum were defined as seeds. The observed directional connectivity alterations at baseline were further
assessed at follow-up and associated with psychotic symptom changes.
Results or Findings: We observed increased information projections in the first episode SZ from cerebellum to cerebrum (i.e., from
the cerebellar attention and cingulo-opercular systems to the bilateral angular gyri, and from the cerebellar cingulo-opercular system
to the right inferior frontal gyrus). In contrast, decreased information projections in the first episode SZ were mainly from cerebrum to
cerebellum (i.e., from the right inferior temporal gyrus, the left middle temporal gyrus, the left putamen, and the right angular gyrus
to the cerebellar language system) and inside the cerebellum (i.e., from the cerebellar cingulo-opercular system to the right
cerebellum Crus II). With symptom remission after one-year antipsychotic treatment, the information projections from cerebrum to
cerebellum were partly restored. The normalization extent of the information projection from the left middle temporal gyrus to the
cerebellar language system was positively correlated with the change in general psychopathological symptoms.
Conclusion: These findings suggest that the decreased information projections from the cerebrum into the cerebellum during the
acute phase of schizophrenia may be a state-dependent alteration that is related to symptoms and medication. While the increased
information projections from the cerebellum may reflect a persistent pathologic trait.
Limitations: Not applicable.
Funding for this study: This study was supported by the National Natural Science Foundation of China (Project Nos. 82120108014,
82071908), National Key R&D Program of China (Project Nos.2022YFC2009901,2022YFC2009900/2022YFC209903), China
Postdoctoral Science Foundation (Project No. 2022M722270), the Youth Science Fund of the Natural Science Foundation of Sichuan
Province, China (Project No. 2022NSFSC1435), CAMS Innovation Fund for Medical Sciences (CIFMS) (Project No. 2021-I2M-C&T-A-022),
Chengdu Science and Technology Office, major technology application demonstration project (Project Nos. 2022-YF09-00062-SN,
2022-GH03-00017-HZ), Sichuan Science and Technology Program (Project No. 2021JDTD0002), Sichuan University Postdoctoral
Interdisciplinary Innovation Fund (Project No. JCXK2209), the Fund of the Beijing Medical Award Foundation (Project No.
YXJL-2022-0665-0189), Dr. Su Lui acknowledges the support from Humboldt Foundation Friedrich Wilhelm Bessel Research Award and
Chang Jiang Scholars (Program No. T2019069).
Ethics committee - additional information: This study was approved by the Ethics Committee on Biomedical Research, West
China Hospital of Sichuan University
42
Evaluating diffusion tensor imaging of corpus callosum in patients with psychosis (7 min)
Rajul Rastogi; Moradabad / India
Author Block: R. Rastogi, D. Goel; Moradabad/IN

Purpose: Psychosis is a complex group of disease showing significant alterations in corpus callosum affecting the cognitive behavior
of the patient. Diffuse tensor imaging (DTI) is an advanced MR imaging tool that depicts the microstructural changes in the white
matter of the brain. Since DTI can predict callosal changes before visible thinning, hence it can be used to determine the prognosis of
patients and manage patients accordingly.
Methods or Background: Twenty patients with proven psychosis for more than one year underwent noncontrast MRI brain and DTI
on 1.5T scanner followed by evaluation of callosal MD, FA & ADC measurement as well as changes in its thickness. This data was
compared to 20 age & sex matched subjects who were normal healthy volunteers.
Results or Findings: Fourteen out of 20 patients revealed alterations in MD, ADC and FA of the corpus callosum with only 3 out of
14 revealed thinning of corpus callosum with ventriculomegaly.
Conclusion: Callosal DTI MR imaging in psychotic patients may be a novel screening tool of disease progression and prognostication
allowing early identification of patients who may require adjuvant psychotherapy or more aggressive management. It may also serve
as an objective tool for evaluating the effects of various drug in the management of retarding or cessation of the progression of
psychosis.
Limitations: Small sample volume, single centre study
Ethics committee - additional information: This study was ethically approved.
Resting state functional MRI in patients with obsessive-compulsive disorder to detect the areas of activation (7 min)
Anand Narsingam Parimalai; Chennai / India
Author Block: A. N. Parimalai, A. Chellathurai, M. K. Logudoss, D. Lakshmi; Chennai/IN

Purpose: The aim of the study is to analyse the various functionally connected networks in obsessive-compulsive disorder using Rs-
fMRI (resting state functional MRI) and to compare the same with normal individuals, thereby defining the pathophysiology of
obsessive-compulsive disorder.
Methods or Background: All the patients who were diagnosed under DSM-5 criteria and were registered and undergoing treatment
in the psychiatry Department in Government Stanley Medical College, Chennai. All our images were acquired using a 1.5 T MRI
system and the resting state functional images were acquired using an echo-planar imaging (EPI) sequence.
Results or Findings: OCD patients showed significant decrease/absence of functional connectivity in executive control network and
significant increase/presence of functional connectivity in orbito-fronto-striatal thalamic network
Conclusion: The altered functional connectivity in executive control network and orbito-fronto -striatal thalamic network forms the
root cause for the obsessions & compulsions in OCD
Limitations: Not applicable
Funding for this study: No funding was received for this study
Ethics committee - additional information: It is an educational study and no ethical approval was sought.
Neural correlates of auditory verbal hallucinations: a combined structural and resting state fMRI (7 min)
Faten Aldhafeeri; Hafar al-Batin / Saudi Arabia
43
Author Block: F. Aldhafeeri, K. Al Enazi; Hafar al-Batin/SA
Purpose: Auditory verbal hallucinations (AVH) refer to the perception of voices by subjects in the absence of external auditory
stimulation. The objective of this study was to investigate the underlying neural mechanisms associated with experiencing AVH while
excluding any presence of psychological or neurological condition.
Methods or Background: This study was approved by the local research ethics committee. A total of 14 subjects suffering from AVH
and 15 age and sex matched healthy controls were recruited in this study. All participants underwent structural magnetic resonance
imaging (MRI), diffusion tensor imaging (DTI), and resting-state functional MRI (rs-fMRI). The findings acquired from each modality
were utilised for statistical comparison between the control group and AVH group
Results or Findings: Structural MRI revealed a remarkable cortical thickness reductions in the frontal and temporal lobes and
cingulate cortex in the AVH group compared to the controls. Rs-fMRI exhibited reduced functional connectivity (FC) in the AVH group
compared to the controls in the following regions: inferior frontal gyrus, anterior cingulate, and temporal cortex bilaterally. AVH
subjects have demonstrated increased FC in the right thalamus, left hippocampal region and amygdala. Compared to the healthy
control group, AVH demonstrated reduced mean fractional anisotropy (FA) in the right fronto-occipital fasciculus and the corpus
callosum and higher mean diffusivity (MD) in the left inferior fronto-occipital fasciculus and anterior thalamic radiation
Conclusion: Using a multimodel MRI techniques, current findings shed the light on the brain mechanisms of AVH by revealing the
causal involvement of key brain areas involving the auditory, language, and memory networks.
Limitations: This study is limited by a small sample size; therefore, we recommend further studies with larger sample sizes to better
understand the neural correlates of AVH.
Funding for this study: This study did not receive any funding.
Ethics committee - additional information: This study was approved by the Local Research Ethics Committee at University of
Hafr Albatin, SA.
44
HW 2Sa - The imaging dilemma in acute stroke: CT vs MR
Categories: Education, Emergency Imaging, Imaging Methods, Neuro, Vascular

CME Credits: 1

1. To become familiar with stroke and therapeutic decisions by investigating the role of MRI and CT in guiding therapeutic decisions
related to strokes.
2. To learn about imaging techniques to differentiate stroke mimics from genuine stroke cases.
3. To develop practical skills in identifying stroke mimics by exploring conditions that resemble strokes but have
different underlying causes.
4. To become familiar with identifying stroke mimics by exploring conditions that resemble strokes but have
5. To understand the interventional-diagnostic correlation with the aim of a better prognostic outcome.

Marios-Nikos Psychogios; Basel / Switzerland
Jessica Jesser; Heidelberg / Germany
45
RPS 210 - Exploring knee imaging
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Musculoskeletal

CME Credits: 1.5
Moderator:
Reto Sutter; Zürich / Switzerland
Assessing lateral posterior meniscal root insertion variations of the knee on a clinical 7 Tesla MRI in an asymptomatic
cohort (7 min)
Roy P. Marcus; Zurich / Switzerland
Author Block: R. P. Marcus, A. A. Marth, S. Zimmermann, R. Sutter; Zurich/CH

Purpose: The aim of this study was to assess variations of posterolateral meniscal root insertions using a high-resolution sequence
on a clinical 7 Tesla (T) MRI scanner.
Methods or Background: Fifty-eight knees of healthy subjects were examined on a clinical 7 T MRI scanner (Magnetom Terra,
Siemens Healthineers). Multiplanar reconstructions of a sagittal isotropic 0.24 mm double echo steady state 3D-sequence were used.
The number of root insertion reins and insertion localisations was recorded for the lateral meniscus by two radiologists.
Results or Findings: The majority of subjects presented with double root insertion reins (66.7%), followed by single (31.6%) and
triple insertion reins (1.8%). The majority of single posterior lateral rein roots inserted into the intercondylar area (77.8%) and the
remaining 22.2% inserted into the posterior slope of the posteromedial tubercle of the tibia. The majority of double reins inserted with
a major rein along the intertubercular area to the posteromedial insertion of the ACL and a minor rein to the lateral tubercle (78.9%).
The inverted variant (major rein inserts into the lateral tubercle and the minor rein inserts next to the tibial ACL-insertion) was visible
in 5.3%. The remaining double reins inserted into the posterolateral slope of the medial and lateral eminence. Interrater reliability for
describing the meniscal root reins was perfect (κ = 1) and identifying the insertion sites was strong (κ = 0.808).
Conclusion: Posterolateral meniscal roots consisted mostly of double root insertion reins, followed by single and triple reins. Five
insertion locations were detected, with the majority inserting with a major rein next to the posteromedial insertion of the ACL and a
minor rein to the lateral tubercle of the tibia.
Limitations: Small sample size.
Funding for this study: This was an in-house study and no funding was received for it.
Ethics committee - additional information: This study was approved by the Local Ethics committee of the Canton of Zurich.
Stability of deep learning-based image quality improvement in MRI of the knee: correlation with arthroscopy (7 min)
Seokhee Park; Seoul / Korea, Republic of
46
Author Block: S. Park, S. H. Choi, H. Lee, S. Kim, J-Y. Jung; Seoul/KR
Purpose: The study aimed to validate whether deep learning (DL)-based image-quality improvement changes the diagnostic
performance and visibility of core features for meniscal, cartilage, and ligament knee lesions.
Methods or Background: This retrospective study include 106 patients, who underwent knee MRI and arthroscopy within a three-
month interval. Fat-suppressed 2D fast spin-echo (FSE) were processed with pre-trained DL. Two musculoskeletal radiologists
independently reviewed both the original and DL-processed MRI for medial meniscus (MM), lateral meniscus (LM), articular cartilage
(AC) and cruciate ligaments (CL). Sensitivity and specificity were compared with arthroscopic results used as the reference standard.
Additionally, two radiologists compared the target lesions between arthroscopic images and original or DL-processed MRI with
different denoising levels using pe-defined similarity score.
Results or Findings: In reader 1, the sensitivity and specificity (%) of original vs. DL-IQI-processed MRI were 91.0 vs. 92.5 (P>0.99)
and 92.3 vs. 97.4 (P=0.50) for MM, 88.9 vs. 93.3 (P=0.50) and 90.1 vs. 91.8 (P>0.99) for LM, 87.2 vs. 94.8 (*P<0.05) and 92.8 vs.
85.7 (P=0.625) for AC, and 91.3 vs. 91.3 (Non-Applicable, NA) and 98.8 vs. 98.8 (NA) for CL. In reader 2, the sensitivity and specificity
of original vs. DL-IQI-processed MRI were 86.6 vs. 91.4 (P=0.25) and 97.4 vs. 97.4 (NA) for MM, 86.7 vs. 86.7 (P>0.99) and 88.5 vs.
88.5 (P>0.99) for LM, 80.8 vs. 84.6 (P=0.37) and 96.4 vs. 89.3 (P=0.5) for AC, and 78.3 vs. 78.3 (NA) and 98.8 vs. 98.8 (NA) for CL.
The mean agreement scores between arthroscopy and MRI for target lesions were 7.18(O) and 7.33-7.36(DL) for meniscal tears
(n=130), and 2.53(O) and 3.23-3.24(DL) for cartilage lesions (n=84).
Conclusion: DL-processing of knee MRIs improved the identification of cartilage lesions, without affecting overall diagnostic
performances as correlated with arthroscopic results.
Limitations: This was a retrospective, single-centre study.
Funding for this study: No funding was obtained for this study.
Ethics committee - additional information: This study was approved by the Institutional review board in our hospital: 20230822-
F-261.
M score as an alternative to T score in the evaluation of femoral osteoporosis? (7 min)

Muhammet Batuhan Gökhan; Ankara / Turkey
Author Block: M. B. Gökhan, N. Çay; Ankara/TR

Purpose: This study aimed to contribute to the diagnosis of osteoporosis-osteopenia with MRI-based “M-scores” obtained from T1
weighted hip MRI.
Methods or Background: Between February 2019 and May 2023, patients aged 18 years or older, who underwent DXA and also
had hip MRI within six months before or after DXA, were included. A reference group consisting of 190 females with a normal BMI who
had hip MRI and aged between 20 and 29 years was also included.
Images were retrospectively evaluated by two radiologists, blinded to the DXA scores. Signal intensities of locations corresponding to
DXA measurement sites, subcutaneous fat intensities, muscle intensities, and noise from artifact-free areas, were measured. Signal-
to-noise(SNR), signal-to-fat (SFR) and signal-to-muscle ratios(SMR) were calculated. M scores for SNR, SMR and SFR were calculated
based on the measurement values of the reference group, similar to the T score. Correlation and ROC analysis were performed for all
M-scores .
Results or Findings: Included 330 patients (mean age 57.61); 30(%9.1) osteoporotic for hip at DXA. There were a strong negative
correlation between M-scores and T-scores (SNR-M-r=0.722, SFR-M-r=0.750, SMR-M-r=0.662) for neck and moderate negative
correlation for total hip (SNR-M-r=0.599, SFR-M-r=0.625, SMR-M-r=0.568). Cut-off points for femoral neck osteoporosis detection were
determined as follows: SNR-M-score at 2.23 (sensitivity 93%, specificity 71%), SFR-M-score at 3.052 (sensitivity 84%, specificity 85%),
and SMR-score at 1.538 (sensitivity 100%, specificity 78%). And cut-off points for hip were determined as following: SNR-M-score at
1.22 (sensitivity 85%, specificity 68%), SFR-M-score at 1.313 (sensitivity 75%, specificity 82%), and SMR-M-score at 1.576 (sensitivity
67%, specificity 83%).
Conclusion: Our study suggests the potential of MRI-based M scores for diagnosing hip osteoporosis, but further validation on a
larger population is needed.
Limitations: The limitations of the study is sample size.
Ethics committee - additional information: The study was approved by Ankara Bilkent City Hospital Ethics Committee:
E2-23-4074.
Extension and costs of low-value MRI of the knee in Norway (7 min)

Bjørn Morten Hofmann; Gjøvik / Norway
47
Author Block: B. M. Hofmann1, V. Håvik2, I. Ø. Brandsæter1, E. R. Andersen1, E. Kjelle1; 1Gjøvik/NO, 2Oslo/NO
Purpose: As low-value imaging is a challenge for the quality of care and patient safety, we need more evidence about the extension
and costs of low-value imaging. Accordingly, the purpose of this study is to provide knowledge about the extension of knee MRIs that
are of low value and to estimate their opportunity costs.
Methods or Background: Contingent extraction of imaging and diagnostic data from the control and reimbursement of healthcare
claims (Helfo) registry in Norway in 2021 analysed with descriptive statistics.
Results or Findings: Between 60 and 80 per cent of knee MRIs in Norway could be of low value, as they yield no relevant diagnoses
or actions prior to or after the MRI examination. This corresponds to between 24,000 and 35,000 MRIs, at a cost of USD 8 to 12 million
annually. Fewer than 7% of the patients had a relevant diagnosis code or a prior imaging examination when having the MRI, and less
than 15% got a knee-related diagnosis code from specialist care within six months after the MRI.
Conclusion: Most MRIs of the knee in Norway could be of low value to the patient. Reducing low-value MRIs can free resources for
high-value imaging, reduce wait times, increase the quality of care, in addition to improve patient safety, healthcare efficiency, and
professional integrity.
Limitations: Quality of registry data, inclusiveness of disease and treatment codes, investigative time span.
Funding for this study: This study was funded by the Research Council of Norway: Project number 302503.
Ethics committee - additional information: This is a register study where no patient data are used outside data provider
organization and so no ethical approval was sought for this study.
Investigation of the anatomic risk factors in acute anterior cruciate ligament rupture coexisted with ramp lesions of
the medial meniscus by quantitative MRI (7 min)
Ziyi Tang; Chongqing / China
Author Block: Z. Tang, Y. Luo, D. Liu, S. Zhou, H. Yang; Chongqing, China/CN

Purpose: This study aimed to investigate the predictive anatomic risk factors of knees in patients with acute non-contact anterior
cruciate ligament (aACL) rupture that coexisted with ramp lesions of medial meniscus.
Methods or Background: A total of 202 subjects were retrospectively divided into three groups: (1) an aACL rupture with ramp
lesions group (n=76); (2) an isolated ACL rupture group (n=56) and (3) a normal control group (n=70). Quantitative morphological
parameters on MRI were measured: the diameter of medial femoral condyle (MFC), the anterior-posterior length and depth of medial
tibial plateau (MTP AP length and MTP depth), the lateral posterior tibial slope (LPTS) and medial posterior tibial slope (MTPS), the
asymmetry of LPTS and MPTS (LMPTS), the lateral meniscal slope (LMS), and medial meniscal slope (MMS).
Results or Findings: The MTP AP length, MTP AP length/MFC diameter ratio, MTP depth, LPTS and the asymmetry of LMPTS showed
significant differences among the three groups (p<0.001). The risk factors associated with the ramp lesions including a longer MTP AP
length (OR 1.17, 95% CI 1.00-1.44, p=0.044), deeper MTP depth (OR 1.91, 95% CI 1.22-3.00, p=0.005) and larger MTP AP length/MFC
diameter ratio (OR 1.11, 95% CI 1.01-1.22, p = 0.036). The highest AUC was the MTP AP length/MFC diameter ratio (0.74; 95% CI,
0.66-0.82), with a sensitivity of 73% and specificity of 68% to predict ramp lesions.
Conclusion: Several occult anatomic characteristics of the knee, especially the morphology of the medial tibia plateau, can be
additional risk factors for aACL rupture that coexisted with ramp lesions.
Limitations: Not all cases were diagnosed with ramp lesions under arthroscopy. However, the ramp lesions exhibited the typical MRI
characteristics with specificity as high as 98% in previous studies.
Ethics committee - additional information: The study was approved by The First Affiliated Hospital of Chongqing Medical
University (No.2021-203).
Leveraging subtraction STIR-T1 MAVRIC sequences for precise discrimination of artefacts vs peri-implant infections in
total knee replacement prostheses (7 min)
Sriram Rajan; New Delhi / India
48
Author Block: S. Rajan, V. K. Venugopal, H. Mahajan; New Delhi/IN
Purpose: This study aimed to explore the utility of a new subtraction technique which involved subtracting STIR from T1 images to
unveil the persisting infective signals on STIR images while reducing the T1 artefacts.
Methods or Background: Our study encompassed 80 patients suspected of peri-implant infections surrounding total knee
replacement prostheses, all of whom underwent MRI scans with the standard T1 and STIR MAVRIC protocol. Subsequently, the
anonymised and randomised images were evaluated by two radiologists, possessing 8 and 21 years of experience, who arrived at a
consensus regarding the presence or absence of peri-implant infections or marked them as indeterminate cases. Further, subtraction
images, derived from STIR minus T1 weighted images, were meticulously reconstructed and colour-coded (hot-iron) before being
fused with conventional T1 images. These fused images underwent a second evaluation by consensus to ascertain the presence or
absence of peri-implant infections or indeterminate cases.
Results or Findings: The initial assessment revealed 38 cases detected with infection, 25 cases absent of infection, and 17 cases
marked as indeterminate. However, upon implementing the subtraction technique, the numbers shifted to 34 cases with infection, 32
cases without infection, and 14 indeterminate cases. To ascertain the statistical significance of these results, a chi-square goodness-
of-fit test was conducted, revealing a Chi-square value of 18.66. This value signifies a significant difference between the observed and
expected frequencies, further affirming the utility of the subtraction technique.
Conclusion: Our study underscores the substantial advantages of employing subtraction STIR-T1 MAVRIC sequences in distinguishing
between infections and artefacts, outperforming conventional sequences.
Limitations: This study was limited by the small size of the dataset
Ethics committee - additional information: This study was IRB approved.
Deep learning-based MRI quantification of the thigh muscles in sarcopenia (7 min)

Zaizhu Zhang; Beijing / China
Author Block: Z. Zhang1, W. Yu1, F. Santini2, J. Zhu1, F. Wang1; 1Beijing/CN, 2Basel/CH

Purpose: This study aimed to explore whether MRI markers of cross-sectional volume (CSV), fat fraction (FF) and T2 values are
affected in sarcopenia and to examine associations between MRI markers and components of sarcopenia (muscle mass, handgrip
strength, and gait speed) in Chinese subjects.
Methods or Background: Postmenopausal women were prospectively recruited for this cross-sectional study. Muscle mass was
measured by dual-energy X-ray absorptiometry (DXA). Jamar hydraulic hand dynamometer was used to assess muscle strength and
gait speed was used to assess physical performance. A 3.0T MRI machine was used to scan bilateral thigh muscles. A DL-based
automated segmentation software (Deep Anatomical Federated Network, DAFNE) was used for the segmentation of anatomical
images. MRI measures were compared between sarcopenic and non-sarcopenic groups and correlations with other variables were
assessed. Their optimal cut-off points (Youden index) for sarcopenia diagnosis were identified in relation to the AWGS2 criteria.
Results or Findings: 68 participants (mean age, 56.6 years ± 5.7 [SD]) were studied, and 25 (36.8%) were diagnosed as
sarcopenic. The sarcopenic group had lower CSV (737.8 cm3 vs 862.2 cm3; P<0.001) and elevated T2 values (45.8 ms vs 44.4 mol/L;
P=0.018) than the non-sarcopenic group, although there was no evidence of a difference in FF. CSV was positively associated with
DXA-derived muscle mass and handgrip strength (P<0.001), whereas T2 values were inversely associated with gait speed (P<0.05).
The CSV and T2 value optimal cut-off points in discriminating sarcopenic patients were identified at 798.7 cm3 (AUC-ROC = 0.883)
and 44.8 ms (AUC-ROC = 0.666) respectively.
Conclusion: CSV and T2 values can differentiate sarcopenic versus non-sarcopenic subjects, representing an imaging biomarker of
sarcopenia.
Limitations: This preliminary study was limited in size, and therefore could not account for all confounders in the sarcopenic group.
Ethics committee - additional information: This study was approved by the institutional review board of the Peking Union Medical
College Hospital, Beijing, China.
Five-minute knee MRI: an AI-based super resolution reconstruction approach for compressed sensing. A validation
study on healthy volunteers (7 min)
Robert Angelo Terzis; Cologne / Germany
49
Author Block: R. A. Terzis, T. M. Dratsch, R. Hahnfeldt, L. Basten, P. Rauen, K. Sonnabend, D. Maintz, A-I. Iuga, G. Bratke; Cologne/DE
Purpose: This study aimed to investigate the feasibility of a complete five-minute 2D knee MRI protocol obtained by combining
Compressed Sensing (CS) and a new AI-aided super-resolution reconstruction method.
Methods or Background: 20 volunteers were examined on a 3T-MRI scanner (Ingenia Elition X, Philips) using a 2D knee protocol
similar to clinical standards: a sagittal fat-saturated proton-density-, and a T1-weighted- sequence in coronal, sagittal and transversal
orientation. Images were acquired in standard and low resolution and reconstructed with a traditional CS and a new, AI-aided
technique (CS-SuperRes). Two experienced radiologists assessed subjective image quality using 8 criteria on a 5-point scale, and
signal-to-noise ratio was measured as an objective parameter.
Results or Findings: CS-SuperRes showed superior ratings over equivalent CS reconstructions, particularly for low-resolution images
(overall image quality: 4.3±0.4 vs. 3.4±0.4, p<0.05). For the low-resolution scans, CS-SuperRes reconstructions matched the quality
of standard-resolution traditional CS across all metrics. Overall image impression scored 4.3±0.4 with CS-SuperRes vs. 4.0±0.5 with
traditional CS (p<0.05). This allowed a significant decrease in scan duration without image quality loss, from 11:01 minutes to just
4:46 minutes (a 57% reduction) for the entire protocol.
Conclusion: The AI-aided CS-SuperRes reconstruction method can reduce MRI scan times by 57%, retaining image quality
comparable to traditional CS. This allows for a complete 2D knee MRI scan in just 5 minutes with preserved image excellence,
facilitating increased scanner throughput and improved patient workflow.
Limitations: This was a single-centre study which used a single MRI scanner and a small sample of healthy volunteers. Nevertheless,
the multiscale network ensures data consistency and objective findings corroborate subjective interpretations. Moreover, some of the
scans showed incidental findings, which were visible with all reconstruction methods.
Funding for this study: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-
profit sectors.
Ethics committee - additional information: This prospective single-centre study was carried out in accordance with the ethical
standards in the 1964 Declaration of Helsinki and its later amendments and was approved by the institutional review board.
dGEMRIC evaluation of autologous conditioned adipose tissue with leukocyte-poor platelet rich plasma for treatment
of mild knee osteoarthritis (7 min)
Vid Matišić; Zagreb / Croatia
Author Block: V. Matišić, V. Molnar, E. Pavelić, P. Brlek, Ž. Jeleč, E. Rod, I. Borić, D. Primorac; Zagreb/HR
Purpose: This study was conducted to assess the clinical and radiomorphologic effects of autologous conditioned adipose tissue
(ACA) with leukocyte-poor, platelet-rich (LP-PRP) plasma knee injections in treating patients with mild knee osteoarthritis (KOA). The
main objective of this study was to compare the results of the treatment to the control group treated with the standard hyaluronic
acid injection.
Methods or Background: 16 patients (8 male and 8 female) with KOA were included in this study. Patients received an
intraarticular, ultrasound-guided injection of ACA and LP-PRP. Contralateral knees of the same patients were treated with hyaluronic
acid (HA) to compare the dGEMRIC indices with knees with KOA. Clinical results and dGEMRIC scores were noted by filling out
questionnaires (VAS, WOMAC, KOOS) and imaging at 3-time points occurring prior to intervention and 3 and 6 months after the
intervention. Each subject received 0.2 mmol/kg gadolinium diethylene triamine penta-acetic acid for the purpose of dGEMRIC per
visit.
Results or Findings: Clinical scores demonstrated improvement at both 3 and 6-month follow-ups. However, dGEMRIC indices
showed no significant change at both time points. Furthermore, no difference between the affected and the unaffected knees treated
with HA was demonstrated.
Conclusion: The lack of objective findings after ortho-biologic treatment is a common finding across the published literature. Hence,
the major orthopaedic guidelines do not include or advise against the use of these treatment modalities. Further research should
standardise the dose of given mesenchymal stem cells and PRP products to try and produce higher-quality evidence for their use and
identify the best patients to receive these treatments.
Limitations: This study was limited by the small patient size and the undetermined amount of MSCs given to any of the patients.
Ethics committee - additional information: This study was approved by the Ethics Committee of St. Catherine Specialty Hospital
(protocol code 21/3-1, 9 March 2021)
Anterior tibial translation in the anterior cruciate ligament tear: a magnetic resonance imaging study (7 min)
Henrique Saraiva Ponte; Faro / Portugal
50
Author Block: H. S. Ponte, M. C. P. Ribeiro; Lisbon/PT
Purpose: Anterior Cruciate Ligament (ACL) tears represent about 50% of all injuries located in the knees, leading to a high direct and
indirect burden on healthcare systems. Magnetic Resonance Imaging (MRI) is considered the most accurate in detecting this
pathology. The objective of this study was to evaluate, through MRI, the correlation between anterior tibial translation (ATT)
measurement, a secondary sign of this pathology and the ACL tears, as well as its sensitivity and specificity. We wanted also to
emphasise the importance of secondary signs of ACL tears with MRI as a useful and sensitive identification and diagnosis tool for this
pathology, leading to a change in its approach.
Methods or Background: This quantitative- and descriptive-correlational study included 151 subjects who underwent knee MRI
[1,5T] and were chosen consecutively. 78 were non-pathological, 62 had some type of injury to the ligament and 11 subjects
underwent ACL ligamentoplasty. Subjects eligible had to be 18 years old or older without any other knee injuries, and who have not
undergone knee surgery, except for ACL ligamentoplasty. Sagittal PDw knee MRI were analysed retrospectively, in PACS platforms,
and ATT measurements were performed. The results of ligament injury type were confirmed through the clinical report.
Results or Findings: There is a positive correlation between acute ACL tears and ATT. It was also possible to determine that the ATT
measured with MRI has low sensitivity and high specificity: 29.03% and 89.74% on the lateral compartment, and 43.55% and 80.77%
in the medial compartment, respectively.
Conclusion: We can prove that the ATT, by MRI, can be a useful tool to help radiologists detect/exclude ACL tears in patients with
acute injury, improving the sensitivity of this examination.
Limitations: This study was limited by the various measurement techniques and MRI reports by different radiologists.
Ethics committee - additional information: This study was approved by the ethics committee of Hospital Beatriz Ângelo
Performance of low-field MR for knee cartilage quantification: results on comparison with arthroscopy (7 min)
Francesco Pucciarelli; Rome / Italy
Author Block: F. Pucciarelli, M. Zerunian, M. Polici, B. Masci, A. Del Gaudio, G. Argento, D. De Santis, D. Caruso, A. Laghi; Rome/IT
Purpose: This study aimed to evaluate the reliability of T2-mapping on knee cartilage on low-field (0.3 T) MRI to detect low-grade
chondropathy and to compare it to arthroscopy.
Methods or Background: 55 patients with planned arthroscopic surgery were prospectively enrolled. MRI examination was
performed with a low-field scanner (0.3 T) with a dedicated sequence (3D SHARC) for the evaluation of cartilage T2 relaxation time.
Image analysis was performed by two radiologists with dedicated experimental software specifically set on MatLab. A colour map,
superimposed on the cartilage, consisted of different colours reflecting different T2 values. Each knee was divided into 14 regions,
according to the Whole-Organ Magnetic Resonance Imaging Score (WORMS) classification, and a single free-hand region of interest
(ROI) was drawn in each region. Arthroscopic knee surgery was performed by an orthopaedic surgeon blinded to MRI data and
chondropathy grade was assessed by Outerbridge classification. The T2-relaxation time of each ROI was annotated and compared to
arthroscopy results.
Results or Findings: A total of 52 participants were included and a total of 728 ROIs were analysed. Comparison of negative
patients (grade 0) and positive patients (grade I, II, III and IV) showed significant (p<0.001) lower T2-values for grade 0 (42,35±11,67
ms vs 48,36±12,36 ms, respectively). Sub-analysis of different grades showed significant (p<0.001) lower T2-values between grade 0
and I (42,35±11,67 ms vs 46,38±9,09 ms, respectively) and grade 0 and II (42,35±11,67 ms vs 47,42±10,51 ms, respectively). No
significant difference (p=0.089) was found between grades I and II (46,38±9,09 ms vs 47,42±10,51 ms, respectively). No significant
difference (all p>0.05) was found between the two readers.
Conclusion: T2 mapping applied to low-field magnetic resonance allows discrimination between negative and positive patients with
chondropathy and allows the detection of patients with low degrees of chondropathy.
Limitations: This was a single-centre study with a small sample size.
Ethics committee - additional information: Written informed consent was acquired for all patients and Institutional Review Board
approval was obtained.
To evaluate the impact of cartilage damage on knee joint functionality through dynamic CT imaging (7 min)
Manou Acke; Elsene / Belgium
51
Author Block: M. Acke, T. Scheerlinck, B. Keelson, N. Van Vlasselaer, S. Héréus, G. Van Gompel, E. Cattrysse, N. Buls; Brussels/BE
Purpose: This study aimed to evaluate the impact of cartilage damage on knee joint functionality through dynamic CT imaging.
Methods or Background: In an ex-vivo thiel embalmed leg, cartilage was removed progressively from the patellofemoral joint in
three stages: lateral facet, crista and medial facet. After each stage, dynamic CT scans were acquired while inducing a cyclic flexion-
extension motion of the leg. Dynamic CT scans were acquired in cine mode with 16 cm z-coverage, 50 cm FOV, 80 kVp, 452 mA, 280
ms rotation time and 6 s total scan duration. Bones were segmented and registered through a multi-atlas approach using SimpleITK-
Elastix. Surface proximity maps of the joint were created to visualise the intra-articular joint space. The intra-articular contact area
was computed as the area with a distance to the adjacent bone smaller than the cartilage thickness (5,7 mm). Differences in contact
areas between normal and damaged stages were evaluated using a Wilcoxon Signed Rank test.
Results or Findings: The contact area after maximum intervention increased from 241 mm² to 613 mm² for extension, and from 67
mm² to 480 mm² for maximum flexion (p=0.014). Apart from the first stage, each intervention resulted in an increased intra-articular
contact area (p=0.02 – 0.014).
Conclusion: Dynamic CT scans can indicate an increasing trend of intra-articular contact area with escalating cartilage degeneration.
Limitations: The feasibility was shown for a single, ex-vivo sample.
Funding for this study: Funding for this study was provided by the UZBrussels.
Ethics committee - additional information: Prior to commencing this study, ethical approval (B.U.N. 1432023000100) was
obtained from the VUB-UZ Brussels University Hospital’s ethical committee after submitting the necessary documentation. The
cadaver used in the study was legally donated, and either the subject or their legal next of kin explicitly consented to the use for
research purposes.
52
EU 2 - ESR iGuide: clinical decision support for imaging referrals
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Informatics, Professional Issues
ETC Level: ALL LEVELS
CME Credits: 1.5
Moderator:
Boris Brkljačić; Zagreb / Croatia

Implementing a clinical support tool at a national level: opportunities and challenges (12 min)
Maria Papapanayidou; Nicosia / Cyprus
1. To learn about the ESR iGuide capabilities and integration processes at a national level.
2. To appreciate the impact of ESR iGuide on imaging referrals based on imaging guidelines.
3. To understand utilisation of ESR iGuide for audit purposes.
CT and MRI imaging in Sweden: retrospective appropriateness analysis of large referral samples (12 min)
Henrietta Stahlbrandt; Eksjo / Sweden
1. To learn about the results of the national CT and MRI audit in Sweden.
2. To appreciate the importance of appropriateness in imaging.
3. To understand utilisation of ESR iGuide for audit purposes.
The role of CDS in referral vetting (12 min)

Ronan P Killeen; Dublin / Ireland
1. To learn about the role CDS can play in vetting of imaging referrals.
2. To appreciate the impact of CDS of the efficiency of vetting.
3. To understand the use of CDS as a tool in the vetting process.
Data analysis and indirect benefits of ESR iGuide (12 min)

Marta Serrallonga Mercader; Vic / Spain
1. To learn about internal hospital planification for a new version update.

2. To appreciate how ESR iGuide can be used to analyse hospital data about justification of exams.
3. To understand what patients know about their exam and its justification.
ESR iGuide implementation update: Israel (12 min)

Jacob Sosna; Jerusalem / Israel
1. To learn about the ESR iGuide integration into a proprietary hospital information system.
2. To appreciate the role of CDS in radiology and hospital IT strategy.
53
HW 2Pc - Mimickers: not every lesion is a tumour

CME Credits: 1

1. To become familiar with the typical mimickers in the central zone.
2. To understand the differential diagnosis of a hypertrophied AFS and a tumour.
3. To know the mimickers in the peripheral zone.
4. To know the mimickers in the transition zone.

Geert M. Villeirs; Gent / Belgium
Pieter Julien Luc De Visschere; Ghent / Belgium
54
SF 2b - Chest x-ray: will it stay or will it go?
Categories: Artificial Intelligence & Machine Learning, Chest, Education, Evidence-Based Imaging
CME Credits: 1.5
The session will include a brief overview of the role of AI in the screening of several cancer tumours. Cancer prevention, screening and
early detection offer the best chance of beating cancer and saving lives. The extension of screening programmes and shortage of
radiologists are jeopardising elements for the fully implementing of population-based cancer programmes. The lectures will focus on
the feasibility, testing and planning of innovative AI solutions, with the identification of and addressing barriers and facilitators for the
utilisation of AI in screening services within the different health systems.
Moderator:
Mariana Benegas Urteaga; Barcelona / Spain

Mariana Benegas Urteaga; Barcelona / Spain
Will the use of AI improve chest x-ray diagnosis sufficiently to prevent a change to CT? (20 min)
Replacing chest x-ray by ultra-low dose CT: pros and cons (20 min)
Thomas Frauenfelder; Zurich / Switzerland
Which patients should go straight to CT without a CXR (20 min)

Anthony Edey; Bristol / United Kingdom
Panel discussion: Is there still a role for chest x-ray imaging in the next decade? (25 min)
55
SF 2a - HCC subtypes: from histology to imaging

CME Credits: 1.5
Moderator:
Valérie Vilgrain; Clichy / France

Histological and molecular classification of HCC: impact on patient management and outcome (25 min)
Aurélie Beaufrère; Clichy / France
1. To list and describe the histological subtypes of HCC according to the WHO classification.
2. To summarise the molecular classification of HCC.
3. To demonstrate the clinical impact of the histological and molecular classifications of HCC.
Imaging of classic (NOS) and rare HCC subtypes (25 min)

Roberto Cannella; Palermo / Italy
1. To become familiar with the typical and atypical imaging appearance of classic (NOS) HCC on CT and MRI.
2. To know the imaging features of macro trabecular-massive or steatohepatitis HCC and other rare subtypes.
3. To understand the role of imaging in the diagnosis of HCC subtypes.
HCC alterations and aggressiveness: impact on imaging (25 min)

Raneem Albazaz; Leeds / United Kingdom
1. To be familiar with imaging features that help predict tumour aggressiveness.

2. To understand radiological characteristics that can identify cholangiocarcinoma-containing tumours.
3. To be aware of preoperative prediction models that combine clinical and imaging data to guide management.
Panel discussion: How can imaging anticipate histologic results? (10 min)
56
IF 2 - Integrated diagnostics in radiology
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Oncologic Imaging, Professional Issues, Translational
Imaging
CME Credits: 1.5
This session will discuss the bleeding edge of integrated diagnostics in radiology - where imaging biomarkers will be used as an
integral part of a patient's health data and allow for a comprehensive view of disease and prognosis. Speakers will discuss how
imaging biomarkers are currently used and highlight strategies and potentials for integration in multidisciplinary settings. Such
integrative overviews will allow for the creation of digital twins opening up new opportunities for research and personalised medicine.
Moderator:
Tobias Penzkofer; Berlin / Germany

The EIBALL perspective on imaging biomarkers: where are we at? (20 min)
Aad Van Der Lugt; Rotterdam / Netherlands
Integrated diagnostics: strategies for radiology (20 min)

Geraldine B. Mcginty; New York / United States
eHealth and AI-integrated diagnostics with multidisciplinary digitised data (15 min)
Horst Karl Hahn; Bremen / Germany
Digital twin: the boundary between reality and dream (15 min)
Simone Novelli; Roma, RM, Italia / Italy
Panel discussion: Where do we see integrated diagnostics in 10 years? (15 min)
57
CTiR 2 - CTiR – Next-generation imaging: clinical trials and AI
Categories: Artificial Intelligence & Machine Learning, Research

CME Credits: 1
This session celebrates the 10th anniversary of the Clinical Trials in Radiology (CTiR) programme at the European Congress of
Radiology (ECR). Renowned speakers will explain how they successfully planned, conducted and published their CTiR at ECR and in
high-impact journals. This will help the radiology community by providing 'how to do it' role model information for future CTiRs. Two
distinguished discussants will put the two exemplary studies into perspective by explaining how they have helped to shape clinical
practice. In addition, the audience will learn about the research and educational importance of CTiR as well as its implications for
value-based radiology and will have the opportunity to interact with the faculty in the discussion section of this session.
Moderators:
Marc Dewey; Berlin / Germany
Adrian Brady; Cork / Ireland
Chairpersons' introduction: the importance of clinical trials (6 min)

How to do a clinical trial in radiology: the example of the DENSE trial (9 min)
Ritse Maarten Mann; Nijmegen / Netherlands
How did the DENSE trial change our clinical practice? (4 min)
Fiona J. Gilbert; Cambridge / United Kingdom
How to do a clinical trial in radiology: the example of the DISCHARGE trial (9 min)
Maria Bosserdt; Berlin / Germany
How did the DISCHARGE trial change our clinical practice? (4 min)
Rozemarijn Vliegenthart; Groningen / Netherlands
The research importance of clinical trials in radiology (4 min)

Regina G. H. Beets-Tan; Amsterdam / Netherlands
The educational importance of clinical trials in radiology (4 min)

Annemiek Snoeckx; Zandhoven / Belgium
58
The implications of clinical trials in radiology for value-based radiology (4 min)
Discussion (16 min)
59
E³ 223 - Urogenital
Categories: Genitourinary
CME Credits: 1.5
Moderator:
Valeria Panebianco; Roma / Italy

Renal and adrenal imaging (28 min)

Marie-France Bellin; Le Kremlin Bicêtre / France
1. To describe the normal imaging anatomy and variants of the kidneys and the adrenal glands.
2. To understand the imaging features of benign and malignant tumours of the kidneys.
3. To describe the imaging features of benign and malignant tumours of the adrenal glands.
4. To explain the imaging features of infectious disorders of the kidneys
Imaging of the ureter and bladder (28 min)

Alberto Hebert Vargas; New York / United States
1. To explain the imaging anatomy and variants of the ureter and bladder.
2. To understand the diagnostic evaluation and imaging features of obstructive uropathy.
3. To describe the imaging features of benign and malignant tumours of the ureter and bladder.
Prostate imaging (28 min)

1. To describe the MRI anatomy of the prostate.

2. To describe the imaging features of benign prostatic hypertrophy.
3. To understand the imaging features of inflammatory disorders of the prostate.
4. To explain the imaging features of prostate cancer using the prostate imaging reporting and data system (PIRADS).
60
RC 212 - Novel techniques in paediatric pulmonary imaging
Categories: Chest, Imaging Methods, Paediatric

CME Credits: 1
Moderator:
Pierluigi Ciet; Rotterdam / Netherlands

DUAL energy and photon counting CT (12 min)

Joseph Cao; Durham / United States
1. To discuss technical differences between energy integrating detector (EID) and photon counting (PCD) multi-energy CT.
2. To review dual energy applications in the paediatric chest on a photon counting detector platform.
3. To learn unique features of photon counting CT not available in prior generation dual energy CTs.
Ultrafast and real-time MRI of the lungs (12 min)

Franz Wolfgang Hirsch; Leipzig / Germany
1. To be informed about the new possibility of examining the child's lung in real-time with MRI, without motion artefacts and without
sedation.
Practical applications of lung MRI in children (12 min)

Kushaljit Singh Sodhi; Chandigarh / India
1. To name and identify three practical lung MRI sequences in children and
common practical applications of lung MRI in children.
2. To compare results of lung MRI with CT scan and other diagnostic modalities in children.
3. To describe and reflect on the use of functional lung MRI in children.
Ultrasound of the lungs (12 min)

Jovan Lovrenski; Novi Sad / Serbia
1. To identify and reflect on both the advantages and disadvantages of paediatric lung ultrasound.
2. To compare the use of lung ultrasound to other diagnostic modalities.
3. To adapt the use of lung ultrasound to everyday clinical situations.
Panel discussion: Will MRI take over lung imaging in the future? (10 min)
61
RPS 209 - Advances in liver loco-regional treatment
Categories: Abdominal Viscera, Interventional Oncologic Radiology, Interventional Radiology, Oncologic Imaging, Vascular
CME Credits: 1.5
Moderator:
Nicolas Frédéric Mathieu Villard; Lausanne / Switzerland
Unraveling the controversy: transarterial chemoembolisation vs radiofrequency ablation in early hepatocellular

carcinoma: a systematic review and meta-analysis (7 min)
Saad Ashraf Alsaad; Jerash / Jordan
Author Block: M. S. Abdelgalil1, B. M. Elfakharany2, M. El-Samahy3, I. M. lkhawaldeh4, N. Refaey1, S. A. Alsaad5, D. Alaraby5, M. Abd-
Elgawad6; 1Cairo/EG, 2Alexandria/EG, 3Zagazig/EG, 4Al-Karak/JO, 5Jerash/JO, 6Fayoum/EG
Purpose: Our study aims to compare the efficacy and safety of transarterial chemoembolisation (TACE) versus radiofrequency
ablation (RFA) in treatment for early HCC patients near critical liver regions.
Methods or Background: Data from diverse sources (Cochrane Library, PubMed, Scopus, Web of Science) were collected. Inclusion
criteria: patients with single tumours <5 cm or three tumours ≤3 cm, without spread. Outcomes assessed: overall survival (OS),
recurrence-free survival rate (RFS), progression-free survival (PFS), tumour response (TR), and adverse event rate (AER).
Results or Findings: Eighteen studies (4,537 patients) were reviewed. Pooled analysis favored RFA with better three-year and five-
year OS rates compared to TACE (RR = 0.85, 95% CI [0.78, 0.92], p < 0.00001; RR = 0.80, 95% CI [0.73, 0.89], p < 0.0001). However,
one-year OS showed no significant difference except in specific subgroups (patients aged ≥60 years, AFP ≥100 ng/ml, or receiving
Adriamycin), where RFA performed better (RR = 1.02, 95% CI [0.99, 1.05], p = 0.29). Propensity score-matching analyses confirmed
superior three-year and five-year OS rates for RFA (RR = 0.91, 95% CI [0.85, 0.96], p = 0.001; RR = 0.85, 95% CI [0.79, 0.92], p <
0.001). Conversely, TACE demonstrated significant increases in three-year and five-year RFS rates in propensity-score matching
cohorts (RR = 1.32, 95% CI [1.10, 1.60], p = 0.003; RR = 1.26, 95% CI [1.10, 1.46], p = 0.001).
Conclusion: In early-stage HCC, TACE does not demonstrate superior oncologic outcomes compared to RFA, except for its ability to
reduce adverse effects. This suggests a need to explore alternative treatments to RFA
Limitations: The dominance of retrospective studies introduces potential bias related to patient selection. Heterogeneity in
outcomes, influenced by factors like age and prior chemotherapy, was prevalent. Publication bias was also identified.
Ethics committee - additional information: Since, this was a meta-analysis, it did not require ethical approval
Salvage therapy outcomes of degradable starch microspheres transarterial chemoembolisation (DSM-TACE) in patients
with uveal melanoma liver metastases (7 min)
Johannes Maximilian Ludwig; Heidelberg / Germany
62
Author Block: J. M. Ludwig, H. Steinberg, J. Haubold, L. Umutlu, M. Forsting, J. T. Siveke, B. M. Schaarschmidt, J. Theysohn; Essen/DE
Purpose: The aim of this study was to assess DSM-TACE as salvage therapy in patients with unresectable uveal melanoma liver
metastases and to identify prognostic factors for survival.
Methods or Background: Fifty-five patients (49.1% male, median age 65 yrs) first treated between 08/2016-06/2021 were assessed
retrospectively. One hundred and sixty two DSM-TACE (median of 2/patient, range: 1-12) with melphalan (in 98%) were performed.
Cox Proportional Hazard Model for uni-(UVA) & multivariate (MVA) analyses (Hazard ratio;95%CI,p-value) and Kaplan-Meier analysis
for determining median overall survival and time to progression (OS/TTP in months; 95%CI) were performed. Response assessment
was performed according to the RECIST criteria.
Results or Findings: The median OS of the study cohort was 7.97 (6.7-9.3) months following the first DSM-TACE. UVA identified low
lactate dehydrogenase (LDH <2x the upper level of normal) (0.26;0.12-0.57,p=<0.001), normal serum protein
(0.32;0.2-0.7,p=0.008), hepatic tumour burden ≦25% (0.39;0.19-0.7,p=0.007), and monthly tumour growth rate ≦20% before the
first DSM-TACE (0.32;0.14-0.7,p=0.005) as predictors of prolonged OS. MVA confirmed low LDH (median OS: 11.4 vs. 4.3 months,
p=0.021) and low TGR (median OS: 9.9 vs. 6.4 months, p=0.005) as independent predictors. Patients with low LDH and TGR (40.9%)
survived longer, with a median OS of 20.4 (8.4-1.6) months compared to patients with high LDH or high TGR (46%) with 6,4 months
(4,5-8,9) or patients with high TGR and high LDH (14%) with 2.4 months (1,1-7,2). Median TTP was four months (3.1-5.7). The best
response observed was PR in 14%, SD in 67%, and PD in 19%, with a median OS of 25 (13.5-31.6), 8 (7.1-.8), and 4.4 (2.1-20.4)
months, respectively.
Conclusion: DSM-TACE is a promising salvage therapy for patients with unresectable uveal melanoma liver that can achieve disease
control and prolong survival. Low pretreatment LDH and low prior tumor growth rate are independent predictors of prolonged survival.
Limitations: Single-centre, retrospective study is its limitation.
Ethics committee - additional information: This study is a retrospective single-centre database analysis that has been approved
by the local institutional review board with waived informed consent (5: 20-9799-BO).
Percutaneous combined treatment (transarterial chemoembolisation + thermal ablation) in patients with

hepatocellular carcinoma enlisted for liver transplant: pathological findings and follow-up (7 min)
Federica Piccione; Turin / Italy
Author Block: F. Piccione, M. Calandri, A. Doriguzzi Breatta, P. Fonio, M. M. E. Fronda, C. Gazzera, E. Susanna; Turin/IT
Purpose: The study aimed at evaluating the efficacy of a combined thermal ablation (TA) and trans-arterial chemoembolisation
(TACE) treatment as a bridge or downstaging method for liver transplantation (LT) in patients with hepatocellular carcinoma (HCC) >
3 cm in size.
Methods or Background: A retrospective review encompassed 700 consecutive patients subjected to OLT for HCC, focusing on 36
patients who underwent combined TA and TACE as bridge or downstaging. Primary objectives included assessing explanted liver
pathology, emphasising necrosis of the targeted lesion, post-OLT overall survival (OS) and post-OLT recurrence free survival (RFS). A
comparison in terms of post-OLT OS and RFS with 170 patients subjected to TA alone for nodules < 3 cm in size was also performed.
Results or Findings: Out of the 36 patients, 63.9% underwent the combined therapy as a bridge, while 36.1% required downstaging
to meet the Milan criteria. The average treated node size was 4.25 cm (+/- DS). Half received radiofrequency (RF) treatment and the
other half underwent microwave (MW) treatment. All nodes underwent drug-eluting beads (DEB) TACE. The mean necrosis percentage
was 65.9% in the RF+TACE group and 83.3% in the MW+TACE group (p-value = 0.099). OS was 100% at 1 year, 100% at 3 years, and
94.7% at 5 years. RFS was 97.2% at 1 year, 94.4% at 3 years, and 90% at 5 years. Despite the different sizes of the lesions, OS and
RFS did not show significant differences between the study cohort and the cohort of patients subjected to TA alone.
Conclusion: Combined therapy (TA and TACE) is effective for HCC > 3 cm, particularly for bridging and downstaging to OLT,
achieving OS and RFS rates exceeding 80% at 1, 3, and 5 years.
Limitations: Retrospective single-centre study limits the scope of this study.
Funding for this study: No funds were received for this study.
Ethics committee - additional information: Waiting for a response.
Adjuvant transarterial chemoembolisation in resected macrotrabecular-massive hepatocellular carcinoma (ATAC-

MACRO): a multicentre real world study (7 min)
Yi Yang; Beijing / China
63
Author Block: Y. Yang1, J. Zhou2, X. Chang3, H. Zhao1; 1Beijing/CN, 2Zhengzhou/CN, 3Jinan/CN
Purpose: The objective of this study was to evaluate the effectiveness and safety of adjuvant transarterial chemoembolisation
(TACE) in patients with resected macrotrabecular-massive hepatocellular carcinoma (MTM-HCC).
Methods or Background: MTM-HCC is a novel morphological subtype associated with early recurrence, no adjuvant therapy for
resected MTM-HCC has been established. This retrospective study included patients with resected MTM-HCC with/without adjuvant
TACE at four centers between January 2015 and December 2022. The end point was recurrence free survival (RFS), overall survival
(OS) and safety.
Results or Findings: A total of 559 eligible patients were classified into the adjuvant TACE group (n=173, mean [SD] age, 55.0 [6.0]
years; 147 male [85.0%] individuals) and the observation group (n=386, 56.0[6.0] years, 318 males [82.4%]) based on whether or not
to receive postoperative adjuvant TACE. After minimising the biases between two groups using propensity score matching analysis,
both RFS (HR 0.62 [95% CI, 0.48 to 0.80]; P < 0.001) and overall survival (OS, HR 0.59 [95% CI, 0.42 to 0.84]; P = 0.013) in the
adjuvant TACE group were significantly better than the observation group. By Cox regression models, mALBI grade, types of
hepatectomy, number, satellite lesion, without adjuvant TACE were identified as independent risk factors for RFS, and mALBI grade,
maximum tumour size, microvascular invasion, without adjuvant TACE were identified as independent risk factors for OS. The
incidence of surgery-related adverse events (AEs) had no significant difference between the two groups (P = 0.609).
Conclusion: Adjuvant TACE significantly improved RFS and OS of patients with resected MTM-HCC with acceptable toxicity. The
findings suggest that adjuvant TACE should be recommended in resected MTM-HCC.
Limitations: It was a retrospective study conducted in China, and the present study's findings still need to be prospectively validated
by large external cohort studies.
Funding for this study: This study was supported by National Natural Science Foundation of China (No. 81972311, 82141127),
CAMS Innovation Fund for Medical Sciences (CIFMS) (No. 2021-I2M-1-066), Non-profit Central Research Institution Fund of Chinese
Academy of Medical Sciences (No. 2019PT310026), Natural Science Foundation of Shandong Province (ZR2020QH177), Henan
Provincial Medical Science and Technology Research Project (LLRGJ20220191), Key Scientific Research Projects of Colleges and
Universities in Henan Province (23A320033), and Henan Provincial Science and Technology Project (232102311080).
Ethics committee - additional information: The study was approved by the institutional review board of ethics committee of each
centre. The requirement for written informed consent for the study was waived.
Two-dimensional perfusion-angiography permits direct visualisation of redistribution of flow in hepatocellular

carcinoma during balloon-occluded TACE (7 min)
Paolo Vetri; Rome / Italy
Author Block: P. Lucatelli, P. Vetri, S. Ciaglia, B. Rocco, E. Damato, M. G. Travaglini, P. Ricci, C. Catalano; Rome/IT
Purpose: The purpose of this study was to demonstrate in vivo redistribution of the blood flow towards HCC’s lesions by utilizing two-
dimensional perfusion-angiography in balloon occluded-TACE procedures.
Methods or Background: Thirty patients with 35 HCC nodules treated in the period between January 2019 and November 2021. For
each patient, a post-processing software leading to a two-dimensional perfusion-angiography was applied on each angiography
performed via balloon microcatheter, before and after inflation. On the color map obtained, reflecting the evolution of contrast
intensity change over time, five regions of interests (ROIs) were assessed: one on the tumour (ROI-t), two in the immediate
peritumoural healthy liver parenchyma (ROI-ihl) and two in the peripheral healthy liver parenchyma (ROI-phl). The results have been
interpreted with a novel in-silico model that simulates the haemodynamics of the hepatic arterial system.
Results or Findings: Among the ROIs drawn inside the same segment of target lesion, the time-to-peak of the ROI-t and of the ROI-
ihl have a significantly higher mean value when the balloon was inflated compared with the ROIs obtained with deflated balloon
(10.33 ± 3.66s vs 8.87 ± 2.60s (p=0.015) for ROI-t; 10.50 ± 3.65s vs 9.23 ± 2.70s (p=0.047) for ROI-ihl). The in-silico model
prediction time-to-peak delays when balloon was inflated, matches with those observed in vivo. The numerical flow analysis shows
how time-to-peak delays are caused by the obstruction of the balloon-occluded artery and the opening of intra-hepatic collateral.
Conclusion: The measurements identify predictively the flow redistribution in the hepatic arteries during b-TACE, supporting a
proper positioning of the balloon microcatheter, in order to provide the operator with an advantageous flow distribution for
chemoembolization.
Limitations: Limitations to the study are the fact that it is a retrospective analysis and that it is monocentric.
Ethics committee - additional information: Not applicable for this study.
Interventional treatment of unresectable or recurrent hepatic cholangiocarcinoma: transarterial chemoembolisation

(TACE) with combined thermal ablation vs TACE alone (7 min)
Thomas J. Vogl; Frankfurt a. Main / Germany
64
Author Block: T. J. Vogl, M. H. H. Albrecht, N-E. A. N-E. Mohammed; Frankfurt a. Main/DE
Purpose: The purpose of this study was to retrospectively evaluate local tumour control and survival rates after targeted local
therapy using transarterial chemoembolisation (TACE) with or without local thermal ablation in patients with unresectable or recurrent
hepatic cholangiocarcinoma (CCC).
Methods or Background: From 01/07 to 12/17, 152 patients (69 males/83 females; mean: 58.7 years) with CCC were
retrospectively evaluated. The study included patients with both unresectable (80.2%) and recurrent lesions (19.8%). Patients were
treated with at least three cycles of TACE (range; 3-26). 32 patients received combined TACE and local thermal ablation. Local tumour
response was assessed by contrast-enhanced MRI (CE-MRI) based on the modified RECIST criteria (mRECIST) and the survival
evaluated using the Kaplan-Meier method.
Results or Findings: The mean survival for all patients was 28.7 months (range:21.8-35.7). Patients with additional thermal ablation
showed significantly longer survival vs. those with TACE as monotherapy (median: 27 vs. 18 months). Tumour response after three
cycles of TACE was either stable (35.5%), partial response (41.4%), progressive (23%), or complete response (0%) and the response
at the last follow-up was 25.7%, 15.2%, 59.2%, and 3.5%, respectively. Significant prognostic factors were found: nodal and/or
systemic metastases, pre-therapeutic tumour load, initial local tumour response and associated application of local thermal ablation.
Conclusion: Targeted therapy of unresectable or recurrent CCC using TACE with an added value of thermal ablation may provide a
therapeutic option for local tumour control and may improve patient's survival.
Limitations: Retrospective non-randomised study design limits the scope of this study.
Ethics committee - additional information: Institutional review board approval was obtained prior to commencing this
retrospective study.
Cone beam-assisted transarterial chemoembolisation in malignant liver tumours: evaluation of parenchymal blood
volume (7 min)
Author Block: T. J. Vogl, E. Emrich, S. Bernatz, T. Gruber-Rouh; Frankfurt a. Main/DE

Purpose: The study was aimed to assess the prognostic value of parenchymal blood volume (PBV) after transarterial
chemoembolisation (TACE).
Methods or Background: One hundred and thirty seven patients (60 females/77 males; median age 61; range 22–86) with
malignant liver tumours of hepatocellular carcinoma (HCC) (n=27), colorectal cancer (CRC) (n=30) or other primaries (n=80) treated
with TACE between 07/2016 and 07/2018 were retrospectively evaluated. Examinations were performed using Artis Pheno- and/or
Artis Zeego-Dyna CT. CTs were reworked at a dedicated workstation to create a PBV map, which was overlapped with the associated
MRI to determine tumour diameter and PBV. Patients were divided into 2 groups: PBV<50 or >50ml/1000 ml and subdivided
regarding the primaries. Median survival was calculated using Kaplan-Meier and compared using the log‐rank test.
Results or Findings: One hundred and eleven patients with 2-6 TACEs (average: 3) were included for measuring reduction PBV and
tumor size: In 27/111 patients with initial PBV <50 ml/1000 ml, tumour volume, independent of the primary, was reduced by 13.26%,
PBV by 23.11%. In 84/111 patients with PBV>50 ml/1000 ml, tumour volume was reduced by 24.01%, PBV by 44.69%. In the overall
study population (n=137) patients with PBV>50 ml/1000 ml (n=101) survived on average 15.05 months, patients with PBV<50
ml/1000 ml (n=36) 10.01 months (p<0.002). The subgroup analysis showed a longer median survival in the HCC group at PBV>50
ml/1000 ml of 18.09 months. For CRC and other primaries survival time for high and low PBV is almost identical with 11.64 months for
CRC and 12.03 months for other primaries.
Conclusion: High initial PBV values in the HCC group showed a prolonged survival time. In the other subgroups (CRC, other
primaries) there was no effect on survival time regarding the initial PBV value.
Limitations: Retrospective study design, preparation and analysis of PBV cards is very time-consuming, different pretreatments of
the primary tumours, partly no longer traceable, all these are the limitations of this study.
Funding for this study: There was no funding for this study.
Ethics committee - additional information: A consenting ethics vote was issued by the Frankfurt Ethics Committee on January 17,
2019 (business no: 376 / 18).
Long-term outcome of transarterial radioembolisation for patients with hepatocellular carcinoma (7 min)
Riccardo Muglia; Bergamo / Italy
65
Author Block: R. Muglia, P. Marra, C. Prussia, M. De Giorgio, M. Viganò, A. Gerali, G. L. Poli, S. Fagiuoli, S. Sironi; Bergamo/IT
Purpose: Transarterial radioembolisation (TARE) is a treatment indicated across many stages of hepatocellular carcinoma (HCC). We
aimed to assess radiological response, safety and overall survival (OS) of TARE for HCC, identifying predictors of OS.
Methods or Background: We included consecutive patients treated with TARE from 2012 to 2021 in a single centre with following
criteria: presence of at least 1 measurable HCC without extrahepatic metastases, Child-Pugh score (CPS) A/B, ECOG performance
status 0/1. Only the first TARE was considered in those patients who received >1 procedure. The radiological response by mRECIST
criteria was evaluated 3/4 months after TARE. Uni- and multivariable analyses were used to explore the features at time of TARE and
post-TARE radiological evaluation.
Results or Findings: Among the 142 patients (median age 67 years, 85% males, 92% cirrhotics, BCLC-A 29%, BCLC-B 35%, BCLC-C
36%, HCV+ 46%, CPS-A 85%, median alpha-fetoprotein 27 ng/mL) the median OS was 16.7 months with a 3-yrs cumulative survival
rate of 28%. 31%, 39%, 9% and 21% of patients had complete response (CR), partial response (PR), stable disease (SD) and
progressive disease (PD), respectively. BCLC stage and alpha-fetoprotein levels at time of TARE, delta alpha-fetoprotein (from TARE to
post-TARE evaluation) and radiological response were statistically related to OS. Alpha-fetoprotein >21.4 ng/mL and BCLC-C at time of
TARE were significantly related with death [HR 1.48 (95%CI 1.00-2.18, p=0.048) and 1.71 (95%CI 1.05-2.79, p=0.031),respectively]
although only radiological incomplete response (PR, SD, PD) had higher HR for death (3.34, 95%CI 2.03-5.79, p<0.0001), at
multivariate analysis. Adverse events occurred in 27% of patients (1 severe).
Conclusion: TARE is effective for HCC patients across different disease stages, and response to treatment remains the most
important predictor of OS.
Limitations: Possible biases in patients selection leading to cohort inomogeneity limit this study.
Funding for this study: This study was not supported by any funding.
Ethics committee - additional information: All procedures performed were in accordance with the ethical standards of the
institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable
ethical standards.
Percutaneous thermal segmentectomy for liver malignancies over 3 cm: retrospective multicentric study of mid-term
oncological performance and predictors of sustained complete response (7 min)
Bianca Rocco; Rome / Italy
Author Block: B. Rocco1, R. Argirò1, V. Semeraro1, Q. Lai1, C. Catalano1, R. Lezzi1, L. Crocetti2, P. Lucatelli1; 1Rome/IT, 2Pisa/IT
Purpose: Report the mid-term oncological results of a multicenter retrospective experience on percutaneous thermal
segmentectomy (single-step combination of balloon-occluded MWA -bMWA- followed by balloon-occluded TACE -bTACE-) in patients
with liver malignancies >3 cm and to identify risk factors for the loss of sustained complete response in the target lesion.
Methods or Background: Sixty-three patients (40/23, male/female) with liver primary malignancies (hepatocellular
carcinoma,HCC=49; intra-hepatic cholangiocarcinoma,iCC=4) and metastasis (n=10) were treated. Median diameter of the target
lesion was 4.5 cm (range: 3-7 cm). bMWA was performed, in a single-step procedure, after balloon-microcatheter inflation, followed
eventually by bTACE (with epirubicin or irinotecan). Oncological results at 1 month and at 3 months interval until 12-15 months, were
evaluated using m-RECIST (HCC) and RECISTv1.1 (metastasis/iCC).
Results or Findings: Median follow-up was 9.2 months. At one month follow-up 79.4% of patients presented a complete response
and the remaining 20.6% of patients were partial responders. The 3-6 months follow up was reached by 59/63 patients: 83.3%(48/59)
of patients showed a sustained complete response, while 10.2%(6/59) and 8.5%(5/59) of patients had respectively a partial response
and a progressive disease. At the last available follow-up, the global median time of sustained complete response was 9 months, with
65.9% of cases showing a complete response. While, no relevant risk factors were identified for obtaining a complete response after
the bMWA-bTACE, an initial diameter of the target lesion ≥ 5cm showed to be the unique independent variable for the risk of failure in
maintaining a complete response at 6 months (OR=8.58, 95%CI=1.38-53.43; P=0.02).
Conclusion: Percutaneous thermal segmentectomy allows to achieve promising oncological results in patients with >3 cm tumors,
with tumour dimension ≥5 cm being the only risk factor associated with the failure of a sustained complete response.
Limitations: Retrospective nature
Ethics committee - additional information: Sapienza ethics committee approved of this study.
Radiomics analysis of pre-treatment imaging in hepatocellular carcinoma (HCC) undergoing trans-arterial

chemoembolisation (TACE): correlations with treatment response and survival (7 min)
Elio Damato; Rome / Italy
66
Author Block: P. Lucatelli, E. Damato, E. P. Perrone, B. Rocco, M. Forlino, A. De Santis, C. Catalano; Rome/IT
Purpose: Radiomics uses radiological imaging to generate multi-dimensional data, defined as features. The novelty of radiomics is
the possible correlation with clinical endpoints, mostly in oncological diseases. This a retrospective study investigate correlations
between pre-treatment imaging radiomics and clinical outcomes in patients with hepatocellular carcinoma (HCC) undergoing
transarterial chemoembolisation (TACE).
Methods or Background: We selected pretreatment data (clinical, CT scan and laboratory) of 64 HCC Patients treated with TACE.
With an open-source software we extract 68 features. Therapeutic outcome was divided in "response" (complete and partial
response) and "non-response" (stable disease). Response to TACE was assessed with mRECIST criteria. Primary endpoint was
correlation with clinical response to treatment. Secondary endpoint was overall survival.
Results or Findings: Primary endpoint: clinical data related to response (Chi-Square test) were age (upper median, p=0.027), Child-
Pugh Score (A vs B, P=0.009) and albumin (upper median, P=0.009). Twelve radiomics features were related with response (Mann-
Whitney test, p<0.05), namely: CONVENTIONAL, SHAPE, GLRLM, NGLDM, GLZLM family. Secondary endpoint: clinical data related to
survival (Cox regression model) were: age, number of HCC nodules, albumin, history of ascites and hepatic encephalopathy. Kaplan
Meier curve showed that Patients that had at least two of the aforementioned clinical variable experienced inferior survival, Log Rank:
p<0,01. Radiomic parameters related to survival were: GLCM_Entropy_median, GLZLM_SZHGE_median. Kaplan Meier curve showed
that Patients with high median scores of GLCM_entropy and GLZLM_SZHGE experienced inferior survival, Log Rank: p=0.022.
Conclusion: Our study showed that some radiomic features have predictive or prognostic value in HCC Pts undergoing TACE. Clinical
and laboratory data are always essential. Integration of the two models can help management of HCC patients.
Limitations: Limited sample size. Further studies are needed to develop standardised models for performing radiomic analysis.
European multicentre retrospective study of balloon-occluded transarterial chemoembolisation (B-TACE) for

hepatocellular carcinoma (HCC) long term (31 months) follow-up (7 min)
Elio Damato; Rome / Italy
Author Block: P. Lucatelli1, E. Damato1, B. Rocco1, T. Debaere2, G. Verset3, F. Fucilli4, A. Paccapelo5, C. Catalano1, C. Mosconi5;
1
Rome/IT, 2Villejuif/FR, 3Brussels/BE, 4Triggiano/IT, 5Bologna/IT
Purpose: The aim of this retrospective multicentric European study was to evaluate the long-term tumor response rates of balloon-
occluded transarterial chemoembolisation (B-TACE) in HCC patients.
Methods or Background: B-TACE procedures response were evaluated according to m-RECIST criteria with multiphasic CT/MRI
imaging at 1 months after the procedure and then every 3-6 months. Data were collected from six European centers from 2015 to
2022. Values assessed at latest available follow-up were disease control (DC), progressive disease (PD) and Progression free survival
(PFS), for both target lesions and patient overall.
Results or Findings: Of the 91 patients of the original cohort, 19 patients were lost at follow-up, resulting in a final cohort of 72
patients. Mean maximum diameter of target lesion was 36.5±19.4 mm, with a median follow-up of 31.6 months. Overall Response
was DC 33.3% (24 patients) and PD 66.7% (48 patients). Target lesion long-term response was DC 69.4% (50 patients) and PD 30.6%
(22 patients). Median Progression-Free Survival was 9.3 months for Overall Response and not Reached for Target Lesion Response
(Local Recurrence-Free), among the 22 patients with local recurrence the mean time observed for PD was 9.8±19.4 months (median
7.7, range 1-31).
Conclusion: The study proves that B-TACE permits to achieve an high sustained disease control rate of the target lesion whereas
progression occurred within first 7 months from initial treatment.
Limitations: Small sample size
MRI-based investigation of dynamic changes in viable hepatocellular carcinoma following transarterial

chemoembolisation (7 min)
Weilang Wang; Nanjing / China
67
Author Block: W. Wang, Y-C. Wang, S-H. Zhang; Nanjing/CN
Purpose: The aim of this study was to analyse the dynamic changes in MRI radiological features of viable tumours at different follow-
up times after the initial Tanscatheter arterial chemoembolisation (TACE) treatment for hepatocellular carcinoma (HCC) to improve
the understanding and diagnostic accuracy of viable tumours.
Methods or Background: The study analysed a prospective, single-arm, multicentre clinical trial data set (NCT03113955) and
another three centres of clinically diagnosed patients with intermediate-stage HCC who received TACE as their first treatment. Viable
tumours in two follow-up MRI were evaluated according to the Liver Imaging Reporting and Data System (LI-RADS). Radioligical
features were evaluated including signal characteristics of T1WI, T2WI, and DWI, the presence or absence of wash in and wash out, as
well as the size, location, and shape of the viable tumours.
Results or Findings: There were 157 HCCs in multicentre data set and 297 HCCs in another three centers met inclusion criteria.
After exclusion, there were total 160 viable tumours on both follow-up MRI. Compared with the radiological features of the baseline,
the T2WI hyperintensity (P= .002), wash out (P< .001) and the tumour short diameter (P< .001) of the first follow-up viable tumours
significantly decreased. Compared with the radiological features of the first follow-up viable tumours, only the DWI hyperintensity of
the second follow-up viable tumours increased (P= .009). For the shape features, both follow-ups showed mostly regular shapes
(66.9%, 60.6%). Another location feature, the first follow up viable tumour showed mostly internal location (51.3%), but the second
follow up viable tumour showed mostly non-internal location (66.9%).
Conclusion: The radiological features of viable tumours at early follow up post-TACE showed more variability compared to the
baseline HCC, while those at second follow-up remained variable but less than the first follow up.
Limitations: Not applicable for this study.
68
SF 2c - Imaging after treatment in head and neck cancer
Categories: Head and Neck, Hybrid Imaging, Oncologic Imaging

CME Credits: 1.5
This session will address four critical issues concerning imaging after head and neck cancer treatment. The first question focuses on
grading the likelihood of recurrence by assigning a standardised value to each suspicious imaging feature, which is then added to a
final scale. NIRADS will significantly improve patient classification homogeneity in multidisciplinary team analysis and treatment
planning. The second issue concerns the extrapolation of imaging biomarkers that have been shown to help predict treatment
outcomes. The following questions will be asked: Should pre-treatment imaging data be used to extract biomarkers? Could they be
obtained early during (chemo)radiation treatment? The final two issues will focus on expected and unexpected changes in tumours
and surrounding tissues (including organs at risk) due to chemoradiation- and immunotherapy.
Moderator:
Roberto Maroldi; Concesio / Italy

Prognostic value of imaging biomarkers for treatment outcome (15 min)

Ann Dorothy King; Hong Kong / China
Imaging head and neck cancer after chemoradiation therapy (15 min)
Steve Connor; London / United Kingdom
Immunotherapy in head and neck: expected and non-expected findings (15 min)
Alexandra Borges; Lisbon / Portugal
NIRADS PET CT and MRI (15 min)

Barton F. Branstetter IV; Wexford / United States
Post-treatment interactive cases presentation (15 min)

Minerva Becker; Geneva / Switzerland
Panel discussion: How to practically organise the follow-up: my advice (10 min)
69
PC 2b - Strategies for success in cardiac imaging
Categories: Cardiac, Imaging Methods, Management/Leadership, Multidisciplinary, Professional Issues

CME Credits: 1.5
Moderator:
Matthias Gutberlet; Leipzig / Germany

Matthias Gutberlet; Leipzig / Germany
Success in cardiac imaging: a multimodality perspective (from CMR to CCTA) (15 min)
1. To learn how to start and grow a career in CMR.

2. To acknowledge that cardiac imaging is not only about CMR (and not only about radiology).
3. To learn how to grow into becoming a multimodality cardiac radiologist.
Success in cardiac imaging: a clinical perspective from Scandinavia (15 min)

Anagha P. Parkar; Bergen / Norway
1. To share tips to establish cardiac CT in a general hospital.

2. To share tips and tricks to prevail in a hostile environment.
3. To discuss how to establish a team of radiographers and radiologists to deliver a robust cardiac CT service.
A split soul lives in my breast: an interdisciplinary perspective (from cardiologist to radiologist) (15 min)
Gianluca Pontone; Milan / Italy
1. To learn that fundamental clinical knowledge is mandatory in cardiovascular imaging.

2. To become aware that your clinical partners are not your enemy.
3. To learn that being a cardiologist is no obstacle to becoming a good radiologist.
An academic perspective: from a dedicated heart centre to a university hospital (15 min)
Pál Maurovich-Horvat; Budapest / Hungary
1. To learn how a sub-specialisation in a dedicated heart centre can help to start with cardiac imaging.
2. To acknowledge that a good cardiac radiologist is a specialised generalist or generalised specialist.
3. To realise that cardiac imaging is still an interesting field of research but already part of radiologist's everyday life.
Panel discussion: Is there a single strategy for success in cardiac imaging across Europe? (25 min)
70
E³ 219 - The "new" digital transition in the radiology department
Categories: Artificial Intelligence & Machine Learning, EuroSafe Imaging/Radiation Protection, Management/Leadership
CME Credits: 1.5
Moderator:
Guy Frija; Paris / France

Dose monitoring systems: how to optimise radiation exposure (25 min)

1. To explain how dose monitoring systems work.

2. To explain the advantages of using dose monitoring systems for radiation protection.
3. To present future developments of dose monitoring systems.
Business analytics (BA) for data-driven decisions (25 min)

Andrea Laghi; Roma / Italy
1. To explain how business analytics software works.

2. To explain the advantages of using business analytics software for management decisions.
3. To present future developments in business analytics.
AI implementation: benefits, threats, and barriers (25 min)

Tim Leiner; Rochester / United States
1. To list different AI-based solutions for radiology.

2. To explain the advantages of using AI in clinical practice.
3. To describe potential threats in AI implementation.
Discussion (10 min)
71
VBR - Value-based radiology: a vague idea or practical reality?
Categories: Audit, Education, Evidence-Based Imaging, Interventional Oncologic Radiology, Interventional Radiology,
Multidisciplinary
ETC Level: LEVEL II
CME Credits: 1.5
Moderator:
Andrea Grace Rockall; Godalming / United Kingdom

How can we measure value in diagnostic radiology? (17 min)

Nuria Bargallo I Alabart; Barcelona / Spain
1. To learn about the meaning of value-based care.

2. To appreciate the different ways that quality and outcomes can be measured in imaging departments.
3. To understand the importance of patient experience in imaging departments.
How can we measure value in interventional radiology? (17 min)

Douglas Mulholland; Dublin / Ireland
1. To learn the methods used to measure quality outcomes in IR.

2. To appreciate the challenges specific to IR related to patient experience.
3. To understand what practical steps imaging departments can take to offer high-value IR.
What do patients value from imaging departments: findings of ESR patient survey (17 min)
Judy Birch; Poole / United Kingdom
1. To learn about the key findings of the ESR patient survey.

2. To be aware of the main patient concerns and needs of radiology service.
3. To understand the central importance of measuring patient experience in imaging departments.
How can we translate value into resourcing of radiology: a European perspective? (17 min)
Michael Fuchsjäger; Graz / Austria
1. To learn how quality measures can be costed in imaging departments.

2. To appreciate some challenges in allocating costs to quality measures.
3. To understand how value-based care can lead to improved resources for imaging departments.
Panel discussion: Can value-based radiology (as opposed to the volume of work) become the real measure of
radiology's impact? (20 min)
72
PC 2a - Challenges in recruitment and retention of radiographers
Categories: Education, Management/Leadership, Professional Issues, Radiographers

CME Credits: 1.5
Moderators:
Charlotte A. Beardmore; London / United Kingdom
Dávid Sipos; Pécs / Hungary

Opportunities and innovations for increasing student training capacity (22 min)
Carst Buissink; Groningen / Netherlands
1. To promote radiography training programmes (EQF Level 6) as the basis for professional recognition.
2. To develop simulation and virtual reality as training opportunities and for skills improvement.
3. To promote clinical research as the first step to developing and improving the radiographer's role.
Safe staffing ratios/quality and safety workload, the need for a framework for radiographers (22 min)
Diego Catania; Milan / Italy
1. To outline the current situation in radiology departments regarding quality and safety.
2. To discuss mechanisms and innovations for responding to staff shortages.
3. To enhance the importance of a framework for professional regulation and mobility.
Solving retention issues for radiographers: a multifactorial approach (22 min)

Patrick Vorlet; Lausanne / Switzerland
1. To outline the reasons for radiographers progressing in their careers.

2. To discuss opportunities to help retain radiographers, both within the current workplace and in the profession.
3. To appreciate the potential role of national bodies and the European Federation of Radiographer Societies (EFRS) in providing an
environment to support retention.
Panel discussion: What are the most successful strategies for solving radiographer staffing issues? (12 min)
73
RPS 203 - Strain analysis in cardiac MRI
Categories: Cardiac, Imaging Methods, Research

CME Credits: 1.5
Moderator:
Ursula Reiter; Graz / Austria
The value of cardiac MRI-derived strain measures in left atrial function assessment and prognostication of
hypertrophic cardiomyopathy patients: a feature tracking cardiac magnetic resonance study (7 min)
Maedeh Dastmardi; Tehran / Iran
Author Block: M. Dastmardi, S. Asadian, N. Rezaeian; Tehran/IR

Purpose: Cardiac MRI has an established position in HCM diagnosis,and the prognostic role of multiparametric cardiac magnetic
resonance(CMR) imaging in HCM is under active research.We aimed to investigate the role of left atrial(LA) strain measures derived
from the feature tracking(FT)-CMR method in predicting adverse events in HCM patients.
Methods or Background: In this retrospective study,the clinical and CMR data of 132 subjects, including 80 HCM cases(42y±17,
60% male)and 52 healthy controls(39y±10.5, 50% male), who underwent CMR examination from October 2017 to June 2022 was
investigated.FT-CMR-derived LA volume and strain values (reservoir, conduit, and booster)were registered. Moreover, the occurrence
of adverse events, including SCD, implantable cardioverter defibrillator(ICD) insertion, deterioration of the cardiac systolic
function(left ventricle ejection fraction[EF] decline≥10%),and aborted SCD, were sought in the patients' medical records.
Results or Findings: During the follow-up of 29 m±12.5, 16(20%) patients developed the combined endpoint. HCM patients had
significantly higher LA diameter, lower LA EF, and lower LA strain values compared to the healthy controls(all Ps<0.05).The univariate
Cox regression analysis demonstrated LAEF(HR, 0.46; CI, 0.24 to 0.90, P=0.02), LA reservoir strain(HR, 0.93; CI, 0.91 to 0.95,
P<0.001),and LA conduit strain(HR, 0.82; CI, 0.54 to 0.91, P<0.001) as predictors of adverse events. Multivariate COX regression
analyses identified LA reservoir strain as an independent predictor of combined adverse events(HR, 0.91; CI, 0.87 to 0.98; P<0.001).
Conclusion: LA dysfunction could be well diagnosed utilizing the FT-CMR method. Low LAEF as well as decreased LA reservoir and
conduit strain values, were associated with poor outcomes in HCM. Furthermore, LA reservoir strain was the potent and independent
predictor of adverse events.
Limitations: There are several limitations to consider in this study. First, it was conducted at a single center. Second, we did not
include LA strain rate and diffuse LV tissue characteristics.
Funding for this study: The authors did not receive any funding.
Ethics committee - additional information: This study was approved by the Research Ethics Committee of Iran University of
Medical.
The additive effects of coronary obstruction on right ventricular diastolic and systolic dysfunction in type 2 diabetes
mellitus patients: evaluated by CMR feature tracking (7 min)
Jin Wang; Chengdu / China
74
Author Block: J. Wang, Z-G. Yang, Y. Li, Y-K. Guo, H. Fang, W-F. Yan, Y. Jiang; Chengdu/CN
Purpose: Obstructive coronary artery disease and type 2 diabetes mellitus (T2DM) commonly coexist, and coexisting OCAD are at
increased risk for adverse cardiovascular events in T2DM patients. This study aimed to investigate the impact of coronary obstruction
on right ventricle (RV) systolic and diastolic function in T2DM patients, and explore independent predictors of reduced RV function.
Methods or Background: Cardiac magnetic resonance (CMR) scanning was performed on 274 T2DM patients {161 patients without
OCAD [T2DM(OCAD−)] and 113 with [T2DM(OCAD+)]} and 83 control subjects. CMR-derived parameters, including RV structure,
function, global strains [including systolic peak strain (PS), peak systolic (PSSR) and diastolic strain rate (PDSR) in radial,
circumferential and longitudinal directions] were measured and compared among observed groups. Univariable and multivariable
linear regression analyses were performed to identify independent predictors of RV dysfunction.
Results or Findings: Compared to the control subjects, T2DM (OCAD-) patients exhibited reduced RV systolic and diastolic function,
as evidenced by a decrease in all RV global strains except for radial PS, PSSR, and PDSR (all P < 0.05). T2DM(OCAD+) patients
showed a significantly more severe impairment of RV systolic and diastolic function than T2DM(OCAD−) patients and control subjects
(all P < 0.05). The presence of OCAD was independently correlated with reduced GCPS ( β = −0.149, P < 0.05), the radial ( β =
−0.204, P < 0.001) and the longitudinal PS ( β = −0.155, P < 0.05) in the context of T2DM. Among T2DM(OCAD+) patients, Gensini
score was associated with decreased GCPS (β=−0.388, P < 0.001).
Conclusion: In the context of T2DM, coronary artery obstruction exacerbated RV diastolic and systolic dysfunction. The presence of
OCAD and Gensini score were independent predictors of decreased RV function.
Ethics committee - additional information: Not applicable
Preoperative left atrial strain predicts outcome in patients with HOCM after myectomy (7 min)
Shujuan Yang; Beijing / China
Author Block: S. Yang, S. Zhao, S. Wang; Beijing/CN

Purpose: This study aims to investigate the prognostic value of preoperative FT-derived LA strain among patients with HOCM
following septal myectomy.
Methods or Background: In this retrospective, single-center study, 802 adult patients with HOCM who underwent septal myectomy
from January 2010 through December 2017 were included. All patients underwent preoperative CMR with complete clinical outcome
follow-up. Preoperative LA phasic strains were measured using CMR-FT. Primary endpoint included all-cause mortality, heart
transplantation, unscheduled hospitalization for heart failure, and stroke. Secondary endpoint included all-cause mortality and heart
transplantation.
Results or Findings: Of the 802 patients in this study, 116 experienced primary endpoints, 38 experienced secondary endpoints,
and 18 experienced SCD/aborted SCD during a median follow-up of 56.7 (IQR, 40.2-78.6) months. LA reservoir strain held the best
performance for predicting the occurrence of adverse outcomes. Compared with the group with LA reservoir strain > 22.9% (median),
the annual event rate of primary endpoint(4.2% vs. 1.7%, P < .0001), secondary endpoint (1.5% vs. 0.4%, P =.0005), and sudden
cardiac death (SCD) (0.7% vs. 0.2%, P = .047) was significantly higher in the group with LA reservoir strain ≤ 22.9%. In multivariable
Cox proportional hazards regression analysis, LA reservoir strain≤22.9% was independently associated with primary endpoint (HR
1.98, 95%CI 1.26-3.11, P=.003) and secondary endpoint(HR 3.71, 95%CI 1.73-7.92, P=.001). Incorporating LA reservoir strain, the
performance of the 2014 HCM Risk-SCD score improved for sudden cardiac death prediction (C-statistic: from 0.62 to 0.69; log-
likelihood: from -110.31 to -108.27, P=.04).
Conclusion: In patients with HOCM undergoing septal myectomy, reduced preoperative LA reservoir strain assessed by CMR-FT is
independently associated with postoperative adverse outcomes.
Limitations: Selection bias and referral bias might be induced in this observational study.
Funding for this study: This study is supported by grants from the National Key Research and Development Program of China (No.
2021YFF0501404, No. 2021YFF0501400) and the key projects of the National Natural Science Foundation of China (No. 81930044).
Ethics committee - additional information: The study complied with the Declaration of Helsinki and was approved by the
institutional review boards of Fuwai Hospital.
Subacute and convalescent phases of Takotsubo syndrome: CMR Feature-Tracking evaluation of left atrial and
ventricular involvement (7 min)
Giacomo Carlo Pambianchi; Rome / Italy
75
Author Block: G. C. Pambianchi, L. Ruoli, G. Cundari, L. Marchitelli, C. Catalano, N. Galea; Rome/IT
Purpose: The aim of this study was to evaluate the ventricular and atrial contractile dysfunction during the subacute and
convalescent phases of Takotsubo (TTS), assessing myocardial strain by Cardiac Magnetic Resonance Feature-Tracking technique
(CMR-FT).
Methods or Background: We retrospectively selected 50 patients with clinical-radiological diagnoses of TTS who underwent CMR
within 30 days since the onset of specific symptoms; 19 patients were studied during the subacute phase (saTTS: ≤7 days) and 31
during the convalescence (cTTS: 8–30 days). We measured the left ventricular (LV) longitudinal, circumferential, radial systolic strain
(lvGLS, lvGCS, lvGRS), the left atrial (LA) reservoir (laS_r), conduit (laS_cd), and booster pump strain (laS_bp). CMR parameters were
compared with those of 30 healthy subjects.
Results or Findings: All the patients were female (100%), with a mean age of 69±13 years for TTS and 63±11 years for controls.
When compared to controls, TTS patients showed reduced values of LV and LA strain parameters (p<0.05 for all). The saTTS showed
increased laS_bp (12.7±2.6% vs 9.8±1.9%; p<0.001), reduced lvEF (47.4±11.9% vs 54.8±9.9%; p:0.028), lvGLS (-11.9±3.9% vs
14.6±3.8%; p:0.044), as compared to cTTS. The days passed from the onset of symptoms and CMR was inversely related to laS_bp (r:
-0.484; p<0.001), with a direct correlation with the lvGLS (r: 0.470; p:0.001) with Pearson. In ROC analysis, laS_bp showed the best
discriminatory power between saTTS and cTTS, with an AUC of 0.815 (95% CI: 0.684-0.945, p<0.001).
Conclusion: During the early phases of the TTS LA Booster Pump, function is significantly increased and gradually decreases over
time. laS_bp could be an early marker of cardiac function impairment, even better the LV Strain and EF.
Limitations: The population under analysis was numerically limited and exclusively female.
CMR follow-up of the patients wasn't included in the study. Single-centre study.
Ethics committee - additional information: This study was approved
Impairment of left atrial and ventricular strain as early markers of myocyte dysfunction in hereditary transthyretin (h-
TTR) mutation carriers (7 min)
Luca Conia; Rome / Italy
Author Block: L. Conia, L. Ruoli, G. Stancanelli, G. Cundari, C. Catalano, N. Galea; Rome/IT

Purpose: The aim of the study was to investigate left atrial and ventricular contractile function and myocardial tissue characteristics
by CMR in carriers of an hereditary transthyretin (h-TTR) mutation related with cardiac amyloidosis (CA) in pre-hypertrophic phase,
compared with CA patients with left ventricular hypertrophy (LVH) and healthy controls.
Methods or Background: Twenty patients with definite diagnosis of CA (myocardial biopsy and/or genetic test and/or myocardial
scintigraphy) and ten subjects with h-TTR mutation related to CA without LVH, who performed CMR in a 3T scanner between
2020-2023 were enrolled and compared with ten healthy age- and sex- matched controls. CMR image analysis was performed using a
dedicated software (Cvi42, circle cardiovascular imaging) and LV volumes, maximal wall thickness, native-T1, LV global longitudinal
(GLS), circumferential (GCS), atrial reservoir (ARS), conduit (ACS) and booster-pump (ABPS) strain were measured.
Results or Findings: Native-T1 values were higher in patients with hypertrophic CA (1348±87 ms) compared to h-TTR-mutation
carriers (1227±33 ms) and healthy controls (1225±35 ms) with a p-value < 0,0003, whereas no significant difference was found
between TTR-mutation carriers and healthy controls (p-value > 0,999). ARS, GLS and GCS were higher in healthy controls
(respectively 25.6±2; 20.3±2; 21.9±5 %) compared to TTR-mutation carriers (respectively 22.0±4; 15.0±5; 15.5±5 %) with p-values
< 0,05; there was non significant difference of ACS and ABPS between the two groups (p-value> 0,5).
Conclusion: In h-TTR mutation carriers CMR evaluation showed no significant alterations in T1 mapping values compared to healthy
subjects, whereas a significant reduction was found in atrial and ventricular strain features, indicating a potential role as early
markers of myocyte dysfunction.
Limitations: Small study population. Single center study. Lack of histologically evaluations in h-TTR carriers. Lack of a clinical and
CMR follow-up.
Acute cardiac injury after postoperative left-sided breast cancer radiotherapy assessed by CMR feature-tracking strain
(7 min)
Linlin Zheng; Chengdu / China
76
Author Block: L. Zheng, M-X. Yang, C. Luo, J. Xu, H. Wang, J. Li, J. Zhang, P. Zhou, J. Yin; Chengdu/CN
Purpose: Chest radiotherapy could lead to cardiac toxicities; this study aims to assess the acute cardiac injury after left-side breast
cancer radiotherapy using CMR feature-tracking strain.
Methods or Background: Patients treated with breast/chest-wall radiotherapy for left-side breast cancer were prospectively
enrolled, and the mean heart dose (MHD) was calculated. CMR scans, consisting of cardiac cine and T2-weighted imaging, were
performed within two weeks before and after radiotherapy. Left ventricular ejection fraction (LVEF), global radial strain (GRS), global
circumferential strain (GCS), global longitudinal strain (GCS), and myocardial edema were analyzed by CVI 42 post-processing
software.
Results or Findings: Thirty women patients aged 46±9 years were included (16 received breast radiotherapy and 14 received
chest-wall radiotherapy). The MHD in all patients was 5.33 (3.13-9.19) Gy; and the MHD of chest-wall group was higher compared with
that of breast group (9.65±2.05Gy vs. 3.45±1.15Gy, P<0.001). After radiotherapy, LVEF showed no significant changes but
myocardial edema was observed in nine patients in the whole cohort. Compared with baseline scan, a decrease of GCS (-20.04±1.72
vs. -18.98±2.04, P=0.029) and GLS (-15.64±2.49 vs. -14.32±2.19, P=0.001) was observed in all patients after radiotherapy, while no
differences of GRS were shown. In the chest-wall group, GRS [37.67±6.89 vs. 34.97 (29.95-38.49), P=0.038], GCS (-20.33±1.29 vs.
-18.61±2.59, P=0.028) and GLS (-16.29±2.68 vs. -14.41±2.46, P=0.007) all demonstrated a decrease after radiotherapy; however,
no change of GRS, GCS and GLS was observed in the breast group.
Conclusion: Left-side breast cancer radiotherapy could cause acute cardiac injury, which is more pronounced in patients with higher
cardiac radiation dose. CMR strain allows the sensitive detection of radiation-related cardiac dysfunction soon after radiotherapy.
Limitations: The changes of left ventricular strain after radiotherapy in long-term follow-up and its clinical significance require
further studies.
Funding for this study: This study was supported by National Natural Science Foundation of China (82202094) the Science and
Technology Department of Sichuan Province (2022NSFSC1600)
Ethics committee - additional information: The clinical protocol in this study was approved by the institutional ethics review
board of our hospital.
Unveiling subclinical myocardial changes in a cohort of treatment-naive cancer population with diffuse large B-cell
lymphoma (DLBCL) and breast cancer: is advanced cancer a cardiovascular syndrome? (7 min)
Costanza Lisi; Milan / Italy
Author Block: C. Lisi, S. Figliozzi, F. Catapano, R. Levi, L. Monti, M. Francone; Milan/IT

Purpose: Pre-chemotherapy patients with advanced cancer may experience myocardial alterations due to inflammation and neuro-
hormonal activation. These changes could serve as subclinical indicators of latent heart dysfunction, allowing for enhanced
cardiotoxicity assessment and patient management. Our study sought to explore cardiac MR (CMR) potentials in shedding light on the
cancer-associated uncontrolled immune activation hypothesis and to provide early biomarkers of myocardial damage.
Methods or Background: In this prospective single-center study, we examined 70 individuals, including 35 healthy controls and 35
patients with de novo DLBCL or breast cancer. All patients underwent CMR before starting chemotherapy. Univariate and multivariate
analyses were performed to detect differences in myocardial tissue characterisation and functional parameters between the two
groups.
Results or Findings: Significant differences were found in native T1 mapping times (p<0.01), with oncologic T1 at 998 msec vs.
healthy T1 at 979.5 msec. Global longitudinal strain (GLS) values also differed significantly (p<0.01), with oncologic GLS at 15.06 and
healthy GLS at 19.9. No significant differences were found in global biventricular function ad T2 mapping.
Conclusion: Advanced cancer is a multifaceted syndrome with profound involvement of the cardiovascular system. It initiates tissue
inflammation, oxidative stress, and neuro-hormonal activation, resulting in substantial myocardial tissue changes. These changes
manifest as myocardial fibrosis and apoptosis, potentially accounting for the increase in native T1 values and the decrease in GLS.
This preliminary study underscores the importance of early tissue modifications as subclinical biomarkers of cardio-toxicity and
prompts a reconsideration of the significance attributed to chemotherapy-induced myocardial alterations.
Limitations: This study is limited by its monocentric design and the relatively small sample size.
Ethics committee - additional information: No additional info was provided by the submitter.
Atrial and ventricular strain using cardiovascular magnetic resonance in the prediction of outcomes of pericarditis
patients: a pilot study (7 min)
Riccardo Cau; Cagliari / Italy
77
Author Block: R. Cau1, F. Pisu1, G. Muscogiuri2, R. Montisci1, L. Saba1; 1Cagliari/IT, 2Milan/IT
Purpose: The aims of our study were to explore with cardiovascular magnetic resonance (CMR) the impact of left atrial (LA) and
ventricular myocardial strain in patients with acute pericarditis and to investigate their possible prognostic significance for adverse
outcomes.
Methods or Background: In this retrospective study, 33 consecutive patients with acute pericarditis were studied with CMR. The
primary endpoint was the combination of recurrent pericarditis, constrictive pericarditis, and surgery for pericardial diseases defined
as pericardial events. Atrial and ventricular strain functions were performed on conventional cine SSFP sequences.
Results or Findings: After a median follow-up time of 16 months (interquartile range [13-24]), 11 patients (eight males) acute
pericarditis patients reached the primary endpoint. In multivariable Cox regression analysis, the left atrial (LA) reservoir, LA conduit,
and LA conduit strain rate parameters were all independent determinants of adverse pericardial events. Adjusted models of LA
conduit and LA reservoir showed strong predictive performance, achieving significantly higher time-dependent AUCs than the LA
conduit rate-based model (0.848 [95% CI, 0.7 – 1.0] and 0.851 [95% CI, 0.69 – 1.0], P < 0.05), for outcome prediction at 12 months.
Conclusion: LA reservoir and conduit mechanisms on CMR are independent predictor in patients with acute pericarditis. Including
atrial strain parameters in the management of acute pericarditis may improve risk stratification.
Limitations: Retrospective single-center study
Ethics committee - additional information: This study was ethically approved by the IRB.
Atrial and ventricular involvement in acute myocarditis patients with preserved ejection fraction (7 min)
Riccardo Cau; Cagliari / Italy
1 1 2 3 1 1 1 2 3
Author Block: R. Cau , F. Pisu , G. Muscogiuri , J. S. Suri , R. Montisci , L. Saba ; Cagliari/IT, Milan/IT, Roseville, CA/US
Purpose: To evaluate ventricular and atrial strain parameters using cardiovascular magnetic resonance (CMR) in patients with acute
myocarditis (AM) and preserved ejection fraction (EF) in comparison with control subjects.
Methods or Background: This retrospective study collected 126 consecutive patients (99 males, 27 females; mean age of 44.72 ±
18.22 years) with AM that fulfilled the Lake Louise criteria and with preserved EF and 52 age- and sex-matched control subjects. Left
atrial (LA) and left ventricular (LV) strain functions were assessed on conventional cine SSFP sequences.
Results or Findings: AM patients with preserved ejection fraction had impaired global longitudinal strain (LS), global radical strain
(RS), and global circumferential strain (CS) in comparison with control subjects. LA reservoir and conduit were significantly reduced in
the myocarditis group compared to the control group. In cardiovascular risk factors-adjusted multivariable analysis, atrial and
ventricular strain mechanisms remained significantly impaired in patients with AM with preserved EF compared to control subjects.
Most importantly, a combined model of atrial and ventricular functions exhibited superior discriminatory ability when compared to
base models of atrial or ventricular strain alone.
Conclusion: Besides ventricular strain parameters dysfunction, patients with AM and preserved EF demonstrated impaired reservoir
and conduit mechanisms. A combined analysis of both atrial and ventricular function may improve the diagnostic accuracy for
patients with AM and preserved EF.
Ethics committee - additional information: The study was approved by the IRB.
Strain analysis using feature-tracking CMR: a predictor of cardiac involvement in hypereosinophilia patients? (7 min)
Christos Vasileiou Gkizas; Lille / France
78
Author Block: C. V. Gkizas, B. Saal, M. Dubois, A. L. Rodriguez Musso, G. Lefevre, D. Launay, B. Longere, F. Pontana; Lille/FR
Purpose: The aim of this study was to assess the role of cardiac strain analysis using feature-tracking cardiac magnetic resonance
(CMR) in patients with hypereosinophilia (HE).
Methods or Background: Sixty-one consecutive patients (aged 52±18 years, 35 males) diagnosed with HE and referred for CMR
were enrolled. Based on clinical evaluations, patients were classified into a “symptomatic group, Group1” (n=20), identified by
evidence suggestive of potential HE-induced cardiopathy, and a “screening group, Group2” (n=41), where CMR was executed to
exclude cardiac involvement despite the absence of any other cardiac abnormality. The CMR protocol included conventional cine
imaging, T1 and T2 parametric mapping, and late gadolinium enhancement (LGE) imaging. Global longitudinal strain (GLS), global
circumferential strain (GCS) and global radial strain (GRS) of left ventricle (LV) were measured (cvi42, Circle Cardiovascular Imaging).
Results or Findings: Despite preserved LV ejection fraction (LVEF), GLS, GCS and GRS were decreased in both groups but were
more impaired in Group1 compared to Group2 (GLS= -10.2±3.9 vs -14.5±4.2, p=0.0004; GCS= -12.0±4.1 vs -14.9±4.2, p=0.015;
GRS= 17.6±9.1 vs 24.1±9.6, p=0.0165). No differences were found in all strain parameters when comparing the two different
patterns of LGE (ie. myocarditis or subendocardial fibrosis).
Conclusion: In symptomatic patients, strain parameters are diminished even when the LVEF remains within the normal range.
Utilising feature-tracking CMR may offer a valuable tool in the early identification of cardiac abnormalities in asymptomatic HE
individuals.
Limitations: Relatively small population of patients (rare disease). No endomyocardial biopsies
Funding for this study: Authors did not receive a fund for this study
Ethics committee - additional information: This study was approved by the IRB: HP1907
The additive effect of metabolic syndrome on left ventricular deformation and function in patients with obstructive
coronary artery disease assessed by 3.0T cardiac magnetic resonance feature tracking (7 min)
Chenyan Min; Chengdu / China
Author Block: C. Min, Y. Gao, Y. Zhigang, Y. Li; Chengdu/CN

Purpose: Metabolic syndrome (MetS) can increase the risk of morbidity and mortality of cardiovascular disease and obstructive
coronary artery disease (OCAD). This study aimed to explore the impact of MetS on left ventricular (LV) deformation and function in
OCAD patients and investigate the independent factors of impaired LV function and deformation.
Methods or Background: One hundred and twenty one patients with OCAD and 52 sex- and age-matched controls who underwent
cardiac magnetic resonance scanning were enrolled in the study. All OCAD patients were divided into two groups: OCAD with MetS
[OCAD(MetS+), n = 83] and OCAD without MetS [OCAD(MetS−), n = 38]. LV functional and global strain parameters were measured
and compared among the three groups. Multivariable linear regression analyses were constructed to investigate the independent
factors of LV deformation and impaired LV strain in OCAD patients. The logistic regression analysis and receiver operating
characteristic curve (ROC) were performed to test the prediction efficiency of MetS for LV deformation and impaired LV strain.
Results or Findings: From controls to the OCAD(MetS-)group to the OCAD(MetS+) group, LV mass (LVM) increased, and LV global
function index (LVGFI) and LV global longitudinal peak strain (GLPS) decreased (all p < 0.05). Compared with the OCAD(MetS-) group,
the LV GLPS declined significantly (p=0.027), the LVM increased (p=0.006), and the LVGFI decreased (p=0.043) in the OCAD(MetS+)
group. MetS was an independent factor of decreased LV GLPS (β=−0.211, p=0.002) and increased LVM (β=0.221, p=0.003). The
logistic multivariate regression analysis and ROC analysis showed that combined MetS improved the efficiency of predicting LV GLPS
reduction (AUC = 0.88) and LVM (AUC = 0.89) increase.
Conclusion: MetS aggravated the damage of LV deformation and function in OCAD patients and was independently associated with
LV deformation and impaired LV strain.
Biventricular dysfunction and ventricular interdependence in patients with pulmonary hypertension: a 3.0 T cardiac
MRI feature tracking study (7 min)
Han Fang; Chengdu / China
79
Author Block: H. Fang, J. Li, Z-G. Yang; Chengdu/CN
Purpose: The objective of this study was to explore the right ventricular (RV) and left ventricular (LV) dysfunction and the ventricular
interdependence in patients with pulmonary hypertension (PH), using cardiac MRI feature tracking (MRI-FT).
Methods or Background: One hundred and seven PH patients (mean pulmonary artery pressure >20 mm Hg) and 72 age- and sex-
matched controls were recruited. Strains were compared between controls, PH patients with preserved RV ejection fractions (EF) (≥
40%, n = 48) and PH patients with reduced RVEF (<40%, n = 59) and multivariable linear regression analysis were performed to
explore the RV-LV interdependence in patients with PH.
Results or Findings: RV strain decreased sequentially from controls, through PH with preserved RVEF, to PH with reduced RVEF. PH
patients with reduced RVEF had significantly lower LV strain, especially septal strain and LV peak diastolic strain rate compared with
both controls and PH patients with preserved RVEF. Multivariable analyses showed that RVEF was independently correlated with LV
strain, furthermore, independent of RVEF, RV strain was significantly correlated with LV strain(LVGRS: β = 0.416; LVGCS: β = −0.371;
LVGLS: β = 0.283).
Conclusion: Subclinical impairment of RV function was found in PH with preserved RVEF. LV strain was impaired when RV was
dysfunctional, which was associated with worsening RV strain. Therefore, while focusing on improving RV function, LV dysfunction in
PH patients should also be monitored and treated early in order to slow the progression of the disease.
Limitations: First, this was a single-centre and cross-sectional study conducted in a large tertiary care hospital, and there might be
potential selection bias. Second, because our study was a retrospective study, the inherent design limitations hindered our ability to
consider changes in biventricular interdependence at different stages of PH.
Funding for this study: This work was supported by the 1·3·5 project for disciplines of excellence, West China Hospital, Sichuan
University (ZYGD18013).
Ethics committee - additional information: The current study received approval from the Biomedical Research Ethics Committee
of West China Hospital, Sichuan University: 2019-648.
80
ST 3 - Insights into Imaging - Enhancing radiology through critical thinking
Categories: Education, Professional Issues, Research

The Editor-in-Chief of Insights into Imaging, Prof. Luis Marti-Bonmati (Valencia/ES) explains his concept of 'critical thinking', why
evidence-based radiology is crucial, and his vision for the future of the journal.
Moderator:
Interview (30 min)

Luis Marti-Bonmati; Valencia / Spain
81
ST 4 - Be accepted the missing link between radiology and female cancer patients.

Be accepted was launched in cooperation with the ESR during ECR 2022 and has become a useful communication tool for radiologists,
an appreciated guide for female cancer patients as it has lead to an optimization of the whole heathcare cain. It puts the focus on the
important role of radiologists for patients and physisians and supports to increas their visibility. Be accepted is a success story that
has some great news and developements to announce as you will be introduced to the projects in the pipeline. it has a 360 degree
approach and unique compositon with the goal to provide a benefit to all stakeholders involved.
Moderator:
Interview (30 min)

Sonja Wehsely
Caroline Justich; Vienna / Austria
Christian Loewe; Vienna / Austria
82
CUBE 2 - Next-Gen peripheral interventions presented through cases

During the "What would you do?" sessions, an expert leads the audience through past interventions. At critical junctures in each of
the cases, the audience is then asked about the course of action they would take, with the implications of different decisions then
being explored.
Moderator:

Next-Gen peripheral interventions presented through cases (28 min)

Adrián Picado Bermúdez; Valencia / Spain
1. To highlight specific cases that exemplify the advancements in IR.

2. To provide a comprehensive understanding of challenges and solutions in IR.
3. To discuss the clinical benefits of next-gen interventions.
83
EFRS 1 - EFRS European Diploma
Categories: Education, Professional Issues, Radiographers, Research, Students

ETC Level: LEVEL I
This session aims to discuss the concept of shared learning across Europe, the consideration of European credit transfer system
(ECTS) in the context of the European diploma, and to explore of the barriers and enablers of shared learning.
Moderator:
Chairperson's Introduction (2 min)

EFRS Diploma - Background (8 min)

Paul Bezzina; Msida / Malta
EFRS Educational Wing Perspectives (10 min)

Mark F. McEntee; Cork / Ireland
Discussion (35 min)
Closing (5 min)
84
AI-SC 1 - Radiology with AI: a demonstration of (future) workflow using integration standards
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics

Moderator:
Virginia Tsapaki; Vienna / Austria

Radiology with AI: a demonstration of (future) workflow using integration standards (57 min)
Peter Mildenberger; Mainz / Germany
1. To learn about interoperability standards for AI in radiology.

2. To learn about semantic standards for AI in radiology.
3. To present use cases of integration of AI results in clinical workflow.
85
OF 3T - Tackle twisted cases, win a must-have EDiR educational package (part 1)
Categories: Education, Professional Issues, Students

ETC Level: LEVEL II
CME Credits: 0.5
Moderator:
Laura Oleaga Zufiria; Barcelona / Spain

1. To dive into and experience the wonders of general radiology.

2. To prepare thoroughly for the exam while having a good time with other peers.
3. To gain deep knowledge of neuroradiology and have the opportunity to grow.
Let the games begin (20 min)

1. To scan and interpret two cases of today's subspecialty and possible outcomes based on the attendees' decisions.
2. To get to know and team up with peers from all over the world to help as many patients as possible.
3. To solve the quiz in order to win an EDiR simulation place.*
*Please note that there can only be one winner per session.
Pooling of conclusions and perceptions (5 min)

1. To jointly summarise and review what we have learned at today's session.
86
RT 3 - Medical image perception: when we look but fail to see?
Categories: Education, General Radiology, Professional Issues, Students

CME Credits: 1
The session aims to share the latest research and examples from everyday radiology practice on mechanisms underlying errors in
medical image perception. Furthermore, the question of how the industry can be helpful with medical image perception is raised. The
discussion at the end of the session will integrate the perspective of all participants and offer the take-home messages.
Moderator:

A trained eye: tips and tricks from an experienced radiologist (8 min)

Fleur Kilburn-Toppin; Cambridge / United Kingdom
An eye in training: challenges for trainees (8 min)

Emanuele Messina; Rome / Italy
A common mechanism underlying errors in medical image perception (8 min)

Jeremy Michael Wolfe; Boston / United States
How can industry and research help with medical image perception? (8 min)
Ioannis Sechopoulos; Nijmegen / Netherlands
Discussion (23 min)
87
BS 3b - Vascular
Categories: Imaging Methods, Vascular

CME Credits: 1
Moderator:
Valentin Sinitsyn; Moscow / Russia

Valentin Sinitsyn; Moscow / Russia
Imaging of vasculitis (19 min)

Marco Francone; Milan / Italy
1. To present the current imaging techniques to evaluate vasculitis.

2. To demonstrate typical imaging findings.
Imaging of posttraumatic vascular lesions (19 min)

Hatem Alkadhi; Zürich / Switzerland
1. To present the most frequent imaging findings of posttraumatic vascular lesions.

2. To become familiar with the different imaging modalities
Imaging of aneurysmal diseases (19 min)

1. To present the current imaging techniques to evaluate aneurysmal diseases.

2. To demonstrate the typical imaging findings.
88
BS 3a - Musculoskeletal: rheumatic diseases
Categories: Imaging Methods, Musculoskeletal

CME Credits: 1
Moderator:
Magdalena Sylwia Posadzy; Poznan / Poland
Chairperson’s introduction (3 min)

Imaging of rheumatoid arthritis (19 min)

Julien Galley; Villars-Sur-Glâne / Switzerland
1. To learn about the typical imaging findings in rheumatoid arthritis.

2. To understand the impact of the different imaging modalities.
Imaging of seronegative spondyloarthropathies (19 min)

Christian W.A. Pfirrmann; Forch / Switzerland
1. To learn about the typical imaging findings in seronegative spondyloarthropathies.

Imaging of crystal deposition diseases (19 min)

Reto Sutter; Zürich / Switzerland
1. To learn about the typical imaging findings in crystal deposition diseases.

89
RPS 313 - What lies in store for us in CT or MR imaging
Categories: Imaging Methods, Physics in Medical Imaging, Research

CME Credits: 1
Moderator:
Jonas Seth Andersson; Umeå / Sweden
Multi-spot source cone-beam CT toward quantitative image evaluations (7 min)

Antonio Sarno; Naples / Italy
Author Block: A. Sarno1, P. Cardarelli2, P. Mauriello1, A. Minopoli3, M. C. MOLLO1, S. Pardi1, G. Paternò2, M. Pugliese1, R. de Asmundis1;
1
Naples/IT, 2Ferrara/IT, 3Pisa/IT
Purpose: The purpose of this study was to investigate the potential of a cone beam CT (CBCT) scanner equipped with a multi-spot x-
ray source for reduction of scatter and cone-beam artefacts toward quantitative evaluations.
Methods or Background: The simulated study mimicked a CBCT scanner with a multi-spot x-ray source. This configuration consists
in several focal spots arranged in a linear array parallel to the rotation axis and permits to acquire projections in an unconventional
scanning trajectory (ExoCT). A specific beam collimation is employed for scatter reduction and a controller operates the several spots
in order to have a non-conventional trajectory of the source during the gantry rotation for tightening the FOV sampling. In-silico
studies were performed to evaluate the conspicuity improvements of selected details and the impact on accuracy of estimates of
materials’ attenuation coefficients/HU.
Results or Findings: In a test conducted using a Defrise phantom, which alternates PMMA slabs and 1-mm air gaps, ExoCT
configuration demonstrated that the conspicuity of such gaps is kept constant across the reconstructed volume, differently from what
happened in CBCT where it reduces moving along the axial direction. The use of an ExoCT configuration with 10 focal points parallel
to the rotation axis permitted to reduce the scatter-to-primary ratio of 86%. On the other hand, three spots were sufficient to increase
the estimation accuracy of the attenuation coefficients of a PMMA phantom up to 10%.
Conclusion: We investigated an innovative solution for updating the CBCT classical configuration toward quantitative evaluations of
the images. Such a solution showed – unlike the results attainable with CBCT – to preserve image conspicuity over the entire
reconstructed FOV and permitted to improve the accuracy in the estimates of the sample attenuation coefficients by 10%.
Limitations: The study was conducted in-silico making it a limitation.
Funding for this study: This work rose from qCT project funded by the Italian Ministry of University and Research (CUP
E53D23012420006)
Ethics committee - additional information: No ethical approval was required for this study.
Image domain deep-learning low-dose CT simulator (7 min)

Sjoerd A. M. Tunissen; Nijmegen / Netherlands
90
Author Block: S. A. M. Tunissen1, N. Moriakov2, M. Mikerov1, E. J. Smit1, I. Sechopoulos1, J. Teuwen2; 1Nijmegen/NL, 2Amsterdam/NL
Purpose: The main goal of this study was to develop an image-domain method to generate low-dose CT (LDCT) scans from clinical-
dose CT (CDCT) scans reconstructed with non-linear reconstruction methods, e.g., iterative reconstruction, without access to the
reconstruction method.
Methods or Background: The method consists of three convolutional neural networks (CNNs), to (i) denoise the CDCT to obtain a
noiseless estimate, (ii) estimate the standard deviation (σ) in each voxel of the LDCT, and, (iii) estimate the local noise power spectra
(NPS) in the LDCT. White noise is transformed into LDCT noise, using the σ and NPS estimates, and added to the noiseless estimate to
obtain the simulated LDCT. For training, paired brain LDCTs and CDCTs were used, reconstructed with iterative reconstruction
(AIDR3D, Canon Medical Systems), divided into training/validation/test sets (251/25/50 cases). Each CNN was evaluated on the test
set by (i) determining the decrease in standard deviation in the cerebrospinal fluid (CSF), (ii) comparing the estimated and actual σ of
the LDCT noise inside the skull, and, (iii) comparing the NPS of the generated and actual LDCT noise.
Results or Findings: The denoising network decreased the σ by a median (IQR) factor of 1.71 (1.61-1.95). The median (IQR)
difference between estimated and actual σ was +0.1 (-0.2-+0.3) HU. The median (IQR) error of the radially-averaged 2D NPS of the
simulated and actual LDCT noise inside the skull was 13.9% (11.9%-15.9%).
Conclusion: The proposed method allows for generation of LDCT from CDCT scans without access to the reconstruction algorithm
and works fully in the image domain. Making LDCTs more available for testing new applications without patient radiation.
Limitations: The CNNs of the pipeline need to be retrained for different reconstruction methods. The method is tested only on brain
scans, however it is not anatomy specific.
Funding for this study: The funding for this study was funded by the Health Holland, Canon Medical Systems.
Anthropomophic 3D printed lung vessel phantoms combined with lung nodules for thorax CT imaging clinical
performance evaluation (7 min)
Irene Hernandez-Giron; Dublin / Ireland
Author Block: I. Hernandez-Giron1, P. McHale1, A. Gaffney1, B. Snow1, J. Egan1, C. D'Helft1, R. Byrne1, W. Veldkamp2, J. den Harder3;
1
Dublin/IE, 2Leiden/NL, 3Amsterdam/NL
Purpose: This study aimed to design and manufacture 3D-printed low cost customised anthropomorphic phantoms for thorax CT
imaging evaluation, including lung vessels and disease (lung parenchyma and nodules).
Methods or Background: Two identical mirrored models of the lung vessel tree (vessels diameters between 1 cm and 1 mm) were
generated inside elliptical inserts (10x15x6 cm) using in-house code (Matlab). By design, small phantom sections with vessels were
closed to trap the 3D-printed powder in to mimic diseased parenchyma. Lung nodules with different geometries (spiculared, spherical,
ovoids, rugous) were designed (Meshmixer), with 3-15 mm diameter range. A thorax elliptical model (30x20x6 cm) with holes to
insert the lungs and the spine, was manufactured with PMMA using CNC. The lung inserts were 3D printed with TPU (HP, MJF
technique), the spine with allumide (SLS) and the nodules with different 3D printers and materials (PA12-MJF, PA12-BlueSint-SLS,
PRUSA Tough Resin). The thorax phantom was scanned in a Siemens Somaton Edge Plus CT (thorax protocol) for a range of kV and
the attenuation of the materials measured with selected ROIs and histograms .
Results or Findings: The attenuation of the phantom materials was, for 120 kV: thorax-PMMA (118+-4)HU, spine-allumide
(785+-10)HU, lung vessels-TPU (80+-10HU), parenchyma-(raw-TPUpowder) (-680+-30)HU. For the lung nodules it was: PA12-Bluesint
(-90+-20HU), PA12-MJM (-32+-10HU) and Prusa_ThoughResin(123+-10HU). The 3D printed materials showed in general a slight
increase in HU with increasing kV. The total cost of this in-house anthropomorphic phantom was around 500 euros.
Conclusion: A low cost anthropomorphic thorax phantom, containing realistic lung vessel distributions and lung nodules was
designed and manufactured with tissue-equivalent materials using 3D printing, to be used in task-based clinical assessment of image
quality in CT.
Limitations: The accuracy and reproducibility of the 3D-printed lungs and nodules will be stablised in the future.
Funding for this study: The lung models were created with in-house software created during the NWO funded CLUES project (Pr.
Nr. 13592, 2015-2019, Wouter Veldkamp project leader). The in-house code was expanded on during the project Through the eyes of
AI: safe and optimal integration of Artificial Intelligence in the Radiology Workflow (Pr.Nr 17392, 2019-2022), funded by NWO-Veni
Personal grant programma awarded to I. Hernandez-Giron
Ethics committee - additional information: This study did not use any patient data and did not require Ethics committee aproval
Development of a breast perfusion phantom including in-line optical spectroscopy measurements for the validation
and testing of dynamic contrast-enhanced dedicated breast CT (7 min)
Liselot Goris; Hengelo / Netherlands
91
Author Block: L. Goris1, S. Manohar2, I. Sechopoulos3; 1Hengelo/NL, 2Enschede/NL, 3Nijmegen/NL
Purpose: To develop a dynamic perfusion breast phantom to validate the time-dependent quantitative iodine concentration
estimates in dynamic contrast-enhanced dedicated breast CT (DCE-bCT).
Methods or Background: Perfusion functionality is added to a 3D-printed anatomic breast phantom with tumor mimic, by
incorporating a pumping system, including two programmable pumps for input of water and contrast agent, tubing, and an output
reservoir. Potassium iodide is used as the contrast agent since its light absorbance was found to be concentration-dependent. The in-
line spectroscopy setup includes a light source (400-600 nm), to illuminate the contrast inlet and outlet leading into and leaving the
breast phantom respectively, and photodetectors to measure the transmitted light. To evaluate precision, nine different iodine
concentrations (0.5-7.6 mg I/mL) were tested three times and the coefficient of variation (CoV) was calculated. For accuracy,
measurements of six concentrations were used to fit a calibration curve, and the errors of the remaining three were calculated. The
division of concentrations used for fitting and testing was repeated three times to determine the average error. A time-varying iodine
profile, including wash-in and wash-out, was pumped through the phantom and simultaneously measured with the optical and DCE-
bCT systems.
Results or Findings: The in-line spectroscopy measurements had a CoV of 0.29% and an average error of 0.13%. The optical
system detected the time-varying iodine profiles, with the recorded timepoints matching those visible in the transmission curves.
Conclusion: A combination of a 3D-printed phantom, a pumping system, and real-time spectroscopy seems a feasible approach for
physical simulations of DCE processes in a breast for validation of DCE-bCT.
Limitations: Iodine-concentration estimates through the tumor mimic are not yet possible.
Funding for this study: ERC Grant No.864929
Model-based noise correction of CT perfusion blood flow measurements based on digital perfusion phantoms (7 min)
Neha Vats; Heidelberg / Germany
Author Block: N. Vats, P. Mayer, M. Klauß, H-U. Kauczor, W. Stiller, S. Skornitzke; Heidelberg/DE
Purpose: Image noise can negatively influence the accuracy of CT perfusion blood flow (BF) measurements, the aim of this study was
to evaluate model-based noise correction using digital perfusion phantoms.
Methods or Background: Digital perfusion phantoms were simulated by forward convolution of impulse response function with
arterial input function averaged over 59 patients, considering ground-truth BF (GTBF) of 5-420ml/100 ml/min with a temporal
sampling of 1.5s. To investigate the impact of noise on perfusion measurements, 2x576 random samples of Gaussian noise (standard
deviation=25HU) were introduced. These noise-added phantoms were evaluated on a commercially available workstation (syngo.via;
Siemens Healthineers) using deconvolution to measure noise-impacted BF. Phantoms were iteratively generated and evaluated to
measure BF accurately by approximating GTBF from noise-impacted BF. To this end, noise correction was estimated from differences
between measured BF and noise-impacted BF. The absolute difference between BF and GTBF were calculated, as well as the contrast-
to-noise ratio (CNR). For validation, parenchyma and tumors of 14 pancreatic adenocarcinoma patients were evaluated.
Results or Findings: The measured noise-impacted BF and noise-corrected BF were 140.3±111.7 ml/100ml/min, and 131.9±125.9
ml/100ml/min, respectively, whereas GTBF was 131.3±127.7 ml/100ml/min. After correction, the average absolute difference in BF
measurements decreased significantly from 18.8 to 3.6 ml/100ml/min and CNR improved from 2.52 to 2.66. For patients, BF
converged from 148.3±50.8 ml/100ml/min to 155.0±91.5 ml/100ml/min for parenchyma, and 45.8±20.3 ml/100ml/min to 13.3±18.7
ml/100ml/min for tumour tissue following correction.
Conclusion: Convergence of BF in simulated and patient data signifies efficacy of developed algorithm in improving measurement
accuracy by correcting negative influence of image noise. With further refinement, algorithm holds the potential to standardise
perfusion measurements, enhancing comparability across patients, imaging centres, and equipment vendors, thus contributing to
more accurate diagnosis and treatment planning.
Limitations: For patient dataset, ground-truth were unavailable, and BF in some tumours approached to zero with correction.
Funding for this study: This study was funded by the BMBF: grant 031L0163.
Ethics committee - additional information: The research protocol received approval from the ethics committee at University
Hospital Heidelberg and was conducted in accordance with the ethical standards of the World Medical Association (Declaration of
Helsinki).
An update on iMet-MRI: a European project aimed at improving metrology for quantitative MRI (7 min)
Matt G Hall; London / Ukraine
92
Author Block: C. McGrath1, C. Clarkson2, A. McDowell2, E. Cooke2, M. Causevic3, M. Cashmore2, P. Tofts4, A. Manzin5, M. G. Hall2;
1
Belfast/UK, 2London/UK, 3Sarajevo/BA, 4Brighton/UK, 5Torino/IT
Purpose: Conventional clinical MRI is designed to produce single-use images for use by individual human radiologists. Because
conventional MRI does not contain quantitative information, comparisons of images between different scanners and timepoints is
extremely challenging. Quantitative MRI (qMRI), in contrast, aims to makes measurements of physical parameters. As such it offers
huge potential for improved reproducibility and to detect diffusion changes in tissue which are difficult or impossible to detect in
conventional images. Effective measurements require uncertainties to be understood and the performance of individual systems to be
quantified. The iMet-MRI project aims to provide a solid foundation for qMRI to support clinical translation. This is an update on its
progress.
Methods or Background: Metrologically characterised materials suitable for quantitative T1 and T2 imaging, diffusion, fat fraction,
and iron content. Samples with multiple target parameter values have been produced. We have developed two sets of scan protocols:
one maximising measurement quality, another for clinically feasible timescales. Analysis codes have been developed for all
measurands and analysis is in progress. We have also developed a detailed digital phantom which has allowed simulation-based
investigation of sensitivity to different measurement inaccuracies.
Results or Findings: A modular phantom designed and manufactured, phantoms validated, and scan protocols developed. Data has
been received from core MRI systems and is being analysed, with additional acquisition continuing.
Conclusion: The project is now approximately 75% complete, and we are on target to deliver all our objectives. Experience from
protocol implementation is being collated ahead of the production of a good practice guide for MRI QA. The software tools will also be
made available to allow open reproducibility of all our results.
Limitations: The project's focus is the performance of the scanner, not the interaction of scanner and patient.
Funding for this study: This project (iMet-MRI) has received funding from the EMPIR programme co-financed by the Participating
States and from the European Union's Horizon 2020 research and innovation programme.
Ethics committee - additional information: Not Applicable for this study.
Magic angle Magnetic resonance imaging (maMRI): first in vivo study using a unique low field scanner: what can we
see? (7 min)
Dimitri Amiras; London / United Kingdom
Author Block: D. Amiras, K. Chappell, J. McGinley, H. Lanz, C. Gupte, M. Ristic; London/UK

Purpose: Conventional knee MRI (cMRI) imaging is sensitive and specific at identifying acute injuries in soft tissues due the presence
of blood or fluid. However, due to the inherent low MRI signal in collagen, cMRI does not show a significant contrast between the
scarred lax tissues with poor collagen alignment and normal tissues. It is observed that well aligned collagen fibres produce increased
signal, at short TEs, when imaged at the Magic Angle (MA≈55°) to the main magnetic field (B0). Our aim is to exploit this effect.
The aim of our study was to investigate whether a novel low-field magnet rotating about 2 axes could exploit the magic angle
phenomenon to generate images of soft tissues in the knees of healthy volunteers (maMRI).
Methods or Background: A bespoke low-field system was developed to exploit the magic angle effect. This comprises a 0.15T open
magnet with B0 being parallel to the magnet poles. The magnet can be rotated about two orthogonal axis to produce arbitrary
orientation of B0 to the stationary patient.
The overall magnet assembly provides an accessible gap sufficient to comfortably accommodate extremity imaging. The scanner was
also capable of performing cMRI to a standard comparable to other low-field MRI. A pre-clinical study with ethical approval for imaging
healthy volunteers imaged subjects, maMRI sequences were performed on each subject in addition to cMRI.
Results or Findings: Our results highlight the potential benefit of maMRI - healthy ligaments and tendons could potentially be
differentiated from those that are degenerative and those with chronic injuries. This is of considerable interest in informing clinical
decision-making and can provide valuable information for the prevention of injury.
Conclusion: Our results highlight the potential benefit of magic maMRI - healthy ligaments and tendons could potentially be
differentiated from those that are degenerative and those with chronic injuries. This is of considerable interest in informing clinical
decision-making and can provide valuable information for prevention of injury.
Funding for this study: This study was funded with the Wellcome Trust Innovator Award.
Ethics committee - additional information: Local ethics committee approved this study.
Multiple overlapping-echo detachment imaging technology based on deep-learning for head movements (7 min)
Yue Zhang; Zhengzhou / China
93
Author Block: Y. Zhang, J. Cheng, J. Bao, X. Wang, Z. Li; Zhengzhou/CN
Purpose: Some patients frequently experience uncontrollable involuntary movements, such as Huntington's chorea and delirium,
which pose challenge for successful clinical MRI examinations, particularly when utilising magnetic resonance technology that
provides quantitative information. Here, we aim to use a newly self-developed ultra-fast, anti-motion multiple overlapping-echo
detachment (MOLED) quantitative magnetic resonance technology to address this issue.
Methods or Background: To validate the accuracy, reproducibility and movement resistance of MOLED technology, a set of self-
made phantoms and five volunteers underwent 3.0 T MRI, including conventional T2-weighted imaging (T2WI), multi-section multi-
echo spin-echo (MESE), and MOLED, which can obtain T2 mapping for a slice of the brain in approximately 150 ms. A deep learning
network was employed for image reconstruction. In the volunteers, we performed six scans on the same volunteer, five scans with
minimal head motion (REP 1-5), and the final scan was performed under moderate head movements (REP move). The T2 values of
seven ROIs were used for the analysis.
Results or Findings: The results of phantoms revealed a positive correlation between the T2 values obtained by MOLED and MESE.
In addition, there was good consistency between the two sequences: mean difference (Meandiff) = 4.20%, standard deviation of
difference (SDdiff) = 1.71%. In the volunteers, T2 values derived from MOLED have excellent repeatability for all REPs, with all ROIs
showing good consistency across the five scans, with CoVs ranging from 0.47 to 2.74. Moreover, the head movements caused
characteristic artifacts in the T2WI, but did not induce any systematic bias to the T2 measurement.
Conclusion: The ultra-fast and anti-motion properties of MOLED show high accuracy and excellent repeatability, and access to
reliable quantitative results even under head movements, thus showing significant advantages in imaging patients with involuntary
movements.
Ethics committee - additional information: The study received institutional review board approval and written informed consent
was obtained from all participants.
94
HW 3Pb - Transition zone: how to detect typical and not-so-typical tumours

CME Credits: 1

2. To understand how to make the differential diagnosis of benign hyperplastic nodules.
4. To learn about the impact of PSA density in PI-RADS 1 and 2.

Ivo Gerardus Schoots; Rotterdam / Netherlands
95
RC 302 - Pros and Cons: the follow-up of patients with a history of breast cancer should be performed
with contrast-enhanced imaging
Categories: Breast, Imaging Methods

CME Credits: 1
Moderator:
Silvia Perez Rodrigo; Madrid / Spain

Pro (15 min)

Elisabetta Giannotti; Cambridge / United Kingdom
This house believes that follow-up of patients with a history of breast cancer should be performed with contrast-enhanced imaging.
Con (15 min)

Matthew G. Wallis; Cambridge / United Kingdom
This house believes that follow-up of patients with a history of breast cancer should NOT be performed with contrast-enhanced
imaging.
Panel discussion: Is there a role for contrast-enhanced imaging? (25 min)
96
RPS 314 - Research evidence to inform and shape the future
Categories: Evidence-Based Imaging, Radiographers, Research

CME Credits: 1
Moderators:
Jenny Gardling; Lund / Sweden
Deniz Akata; Ankara / Turkey
An investigation into the knowledge, attitudes, and practice of lead shielding utilisation in interventional radiology
departments and cath-labs across Europe (7 min)
Andrew England; Cork / Ireland
Author Block: M. O'Rouke1, N. Moore1, S. Svetlic2, H. L. Bucknall3, M. F. F. McEntee1, A. England1; 1Cork/IE, 2Milan/IT, 3London/UK
Purpose: According to current literature, there is a lack in the knowledge, attitudes and practices (KAP) of radiation protection (RP)
among interventional radiology (IR) and Cath-lab staff. To the best of the authors’ knowledge, there are no studies investigating the
radiation protection KAP of radiology staff within IR departments and cath-labs across Europe. This study aims to determine the RP
KAP of staff within IR and cath-labs across Europe and the associated influencing factors.
Methods or Background: A cross-sectional study in the form of an online questionnaire was developed. Participation was advertised
online via online platforms and through email. Inclusion criteria included qualified healthcare professionals currently working in
Interventional Radiology and cath-labs across Europe. Section 1 of the questionnaire consisted of questions regarding demographic
data. Section 2 comprised of questions regarding RP training and protocols. Section 3 involved questions regarding the use of
different types of RP lead shields, both personal and co-worker use. Section 4 concluded the questionnaire with questions regarding
other methods of minimising radiation dose within the departments.
Results or Findings: A total of 178 responses to the questionnaire were recorded. Most respondents were female (72%),
radiographers (75%) and within the age bracket of 25-34 (46%). Only (53%) had ever received RP specific training and the majority
(63%, 80) of respondents were currently practicing in Ireland.
Conclusion: The KAP of IR and Cath-Lab staff regarding RP within departments across Europe is low. The unavailability of basic
radiation protection tools and RP specific training courses/modules were some of the reasons for sub-optimal self-protection against
ionising radiation exposure among the respondents. To avoid unnecessary exposure to themselves, co-workers, and patients, it is
suggested that medical professionals be equipped with appropriate training and RP tools.
Limitations: Self-reporting questionnaire limits this study.
Ethics committee - additional information: This study was ethically approved by the Medical School SREC - University College
Cork.
Review of diagnostic reference levels (DRLs) in interventional neuroradiology (INR) (7 min)

Marvin Grech; Ghaxaq / Malta
97
Author Block: M. Grech, F. Zarb, R. Grech, P. Bezzina; Msida/MT
Purpose: The objective of this study was to evaluate radiation dose quantities and methodology for establishing diagnostic reference
levels DRLs (3rd quartile/75th percentile) for interventional neuroradiology (INR).
Methods or Background: Four different databases: Scopus, Web of Science, PubMed, and Pro Quest were utilized using four
keywords: Air Kerma-Area Product (PKA), Fluoroscopy Time (FT), Reference Air Kerma (CAK) and DRLs and their synonyms to search
literature in English within the last decade.
Results or Findings: DRLs were recorded from 38 studies for different Neurointerventional procedures extracted from the literature
for PKA (Gy/cm2), FT (Minutes), and RAK (mGy). Procedure - Number of studies / PKA (Gy/cm2) Mean (min/max) / FT (minutes) Mean
(min/max) / RAK (mGy) Mean (min/max), Cerebral Angiography - 29 / 102.64 (41/256.65) / 11.73 (6/28.4) / 657.74 (289/921.1), Stroke
Thrombectomy - 8 / 163.57 (110/225.1) / 41.34 (30/44/7) / 1012.56 / (730/1590), Aneurysm Coiling - 27 / 254.06 (52.1/487.4) / 54.51
(16/90) / 3309.89 (505/4750)
AVM/AVF Embolisation - 9 / 384.63 (206.4/550) / 86.37 (58.57/535) / 4130.20 (2350/6000).
Statistically significant (p<0.05) variations are evident between procedures based on their complexity. Other variations based on
equipment type, operator experience and methods of data collection were also noted. The authors also sought to identify potential
reasons for such discrepancies within the included studies.
Conclusion: To date, few studies have published data regarding DRL’s in INR procedures. The evident variation in DRL quantities,
warrants the need for optimization strategies, and guidelines for quality control and to reduce procedural radiation doses as per the
“ALARA” principle. Establishment and analysis of DRLs in INR procedures is essential for subsequent optimisation of radiation doses in
INR procedures.
Limitations: This is a literature review and is based on the information available within the reviewed papers.
Public awareness and preferences for medical radiation risk information (7 min)
Author Block: N. Mernagh, A. England, N. Moore, R. Young, M. F. F. McEntee; Cork/IE

Purpose: New legislation says that healthcare workers must inform patients of the radiation dose and risks associated with medical
imaging. Literature suggests the public are being poorly informed. There has been little research carried out where comparisons of
risk have been evaluated in terms of patient understanding. The aim of this study is to investigate radiation awareness and
preferences the public may have on radiation risk communication prior to medical imaging examinations.
Methods or Background: A cross-sectional study was used to obtain quantitative and qualitative data from the public. The online
questionnaire was designed using Qualitrics XM. The target participants were members of the public over 18 years, irrespective of
whether they have undergone previous imaging. Data was analysed using Microsoft Excel.
Results or Findings: A total of 413 participants completed the survey. Majority (n=364,88.1%) had previously had imaging. Most
respondents rated their knowledge on radiation risks as poor (n=213/401, 53.1%). Participants were asked who should be
communicating these risks with them, the person carrying out the examination (usually the radiographer) received the most
responses (n=334, 34.7%), while the referrer was the second most common (n=224, 23.3%). For the risk communication methods,
majority stated very helpful for the associated risk is negligible/minimal/very-low/low/moderate (34.2%), followed by the associated
risk of cancer is 1 in 2000 (33.4%).
Conclusion: The public have a poor awareness of radiation risks from medical imaging. Multiple responses to who should
communicate risk suggests a multidisciplinary approach is warranted. Simple language to convey risk is preferred along with phrases
which directly address cancer risk.
Limitations: Self-reporting survey limits the scope of this study.
Ethics committee - additional information: Medical School SREC - University College Cork approved this study.
Correlating breast lesion dimensions: imaging vs pathological results (7 min)

Laura Martins Segura de Jesus; Vila Real de Santo António / Portugal
98
Author Block: L. M. S. d. Jesus, A. F. C. L. Abrantes, S. I. Rodrigues, L. F. Carvalho, L. P. V. Ribeiro; Faro/PT
Purpose: This study aims to establish the correlation between breast neoplasm dimensions as measured via Mammography (MGM),
Ultrasonography (US), and Magnetic Resonance Imaging (MRI), and the real tumour dimensions obtained from anatomopathological
reports of surgical specimens.
Methods or Background: A retrospective study was carried out on 56 women diagnosed with breast cancer, between January 2021
and November 2022, in a Portuguese Public Hospital. The study assessed the size of the primary tumour using MGM, US, and breast
MRI. Subsequently, we compared these measurements with those obtained from the anatomopathological examination.
Results or Findings: Our findings revealed that, in MGM, US, and MRI, there was an overestimation in 34/56 (60.7%), 18/56 (32%),
and 31/56 (55%) cases, respectively, and an underestimation in 18/56 (32%), 34/56 (60.7%), and 19/56 (33%) cases, respectively. We
observed an absolute agreement in measurements for 4/56 cases between MGM and US when compared to pathological anatomy,
and 6/56 cases between MRI and pathological anatomy.
Conclusion: Our study concluded that there were no significant differences in tumour size measurements between MGM and MRI
when compared to pathological anatomy. Additionally, we found that factors such as histological type or breast density did not
significantly influence measurement accuracy. Notably, MRI exhibited the highest accuracy in evaluating lesion dimensions,
demonstrating a superior correlation with tumour measurements obtained through anatomopathological examination.
Limitations: The main limitations of this study are the small sample size due to the number of patients who underwent neoadjuvant
chemotherapy, the lack of measurements of all the lesions in all the available imaging modalities due to the external origin of the
patients, and the issue of measurement error.
Funding for this study: Not applicable for this study.
Ethics committee - additional information: This study was funded by the Uaif: 159-2022.
Patients’ experience to MRI examinations: a systematic qualitative review with metasynthesis (7 min)
Isabel Nieto Alvarez; Waldkirch / Germany
Author Block: I. Nieto Alvarez1, J. Madl1, L. Becker1, O. Amft2; 1Erlangen/DE, 2Freiburg/DE

Purpose: The review informs practitioners, patients, and policymakers of observations for future research, and describes the
experience and characterising factors in MRI through the adult patients’ voice for guidance of future efforts addressing patient needs.
Methods or Background: Patients often mention distress, anxiety or claustrophobia related to MRI, resulting in no-shows,
disturbances of the workflow, and lasting psychological effects. Patients’ experience varies and is moderated by socio-demographic
aspects alongside the clinical condition. While qualitative studies help understand individuals’ experiences, to date a systematic
review and aggregation of MRI individuals’ experience is lacking.
We conducted a systematic search in PubMed, Scopus, Web of Science, and PsycInfo databases according to the PRISMA guidelines to
identify primary studies reporting patients’ responses to MRI. For quality appraisal, the Joanna Briggs Institute (JBI) tools were used.
Metasynthesis, a concept map, and meta-aggregation were used for data synthesis.
Results or Findings: We identified eight papers on patients’ experience description for qualitative meta-summary (294 fulltexts, 46
studies with sufficient quality, 49 quantitative studies), 220 patients in total. Meta-aggregation of 144 patient quotes answered: (1)
experiences before, at the scanning table, during, and after an MRI, (2) differences based on clinical condition, and (3) characterising
factors, including coping strategies, look-and-feel of medical technology, interaction with professionals, and information. Noteworthy
across studies was the difficulty with the confined space, fear of results, need for information and coping strategies.
Conclusion: Our findings provide a foundational description of adult patients’ MRI experience, revealing themes and characterising
factors at each stage of the procedure through the patients' voice.
Limitations: Most publications lack participants' health literacy level, occupation, developmental conditions, ethnicity, or country of
origin. Studies were mostly conducted in university hospitals. Interviews’ raw data unavailability impeded computer-aided analysis.
profit sectors.
Ethics committee - additional information: The study was conducted according to a systmatic literature review with meta-
summary and metasynthesis of published patient quotes.
Reporting radiographers in Europe: an online survey (7 min)

Malene Roland Vils Pedersen; Vejle / Denmark
99
Author Block: M. R. V. Pedersen1, J. Jensen2, C. Senior3, N. Gale4, C. J. Heales4, N. Woznitza5; 1Vejle/DK, 2Odesa/DK, 3Axminster/UK,
4
Exeter/UK, 5London/UK
Purpose: Reporting radiographers undertake an important role in health care and for the radiographer profession in
general. First introduced in the UK, and now in several other European countries. The objective was to investigate the workforce of
reporting radiographers across the European Federation of Radiographer Societies (EFRS) community.
Methods or Background: A voluntary anonymous 34-item electronic survey was distributed online using social media accounts
such as
Twitter, Facebook, and LinkedIn cover a wide range of topics relating to professional roles, advanced practice, education, and
seniority. The questionnaire was distributed during a 12-week period in 2022.
Results or Findings: A total of 345 individual responses were received from 15 countries with majorities of respondents from
United Kingdom (n=245, 71%) and Denmark (n=66, 19%). The mean age was 41.9 (S.D 9.8), similar for females, 42.5 (S.D 9.0), and
men 40.9 years (S.D 9.7). Most reporting radiographers worked in public hospitals (90%). The vast majority of the respondents
(n=270, n=94%) authored and signed their own clinical reports while a minority (n=18, 6%) stated that their reports were checked by
radiologists.
Conclusion: The survey highlights the scope of practice of reporting radiographers working in Europe. Reporting is becoming a
career path for an increasing number of radiographers across Europe.
Limitations: A limitation is that the survey was published in English, so language may have formed a barrier to
participation within non-English speaking countries.
Ethics committee - additional information: Research Ethics Committee of the University of Southern Denmark approved this
study.
Density bone evaluation by ECHOS: a pilot study (7 min)

Rute Santos; Coimbra / Portugal
Author Block: R. Santos, J. Nobrega; Coimbra/PT

Purpose: The purpose of this study was to correlate the BMD values obtained through ECHOS with the individual characteristics of
the participants and to compare the BMD values (T-Score and Z-Score) between DEXA and ECHOS.
Methods or Background: Two hundred and nineteen participants were submitted to an ECHOS evaluation and 16 of 219
participants were evaluated by DEXA. All of the participants were evaluated on lumbar spine. All participants answered a
sociodemographic questionnaire with factors influencing bone mineral density (eating habits, smoking, physical activity) and signed
informed consent.
Results or Findings: Of the 219 participants, 68% were female, 54.8% practiced physical activity, 5.9% had hormonal pathologies,
53% were between 22 and 59 years and 11.4% smoked regularly in, however, after comparing the BMD there was a significant
correlation with weight, consumption of tobacco and foods rich in calcium but not with the other variables this in the first study,
however the second study revealed the existence of a significant correlation between the values of Z-Score but despite this in the
values of T-Score was not verified.
Conclusion: In short, this study showed that some factors such as smoking, weight and calcium intake may influence the values
obtained in ECHOS and also the existence of agreement between the two techniques used to assess BMD. ECHOS, having the
advantages of ultrasound, can be an imaging method used to screen for osteoporosis or osteopenia. However future studies must be
developed.
Limitations: Some questionnaires were not completed by participants, the equipments were not always available, not allowing the
evaluation of same number of subjects.
Ethics committee - additional information: The study in question obtained authorisation from the ethics of the Polytechnic
Institute of Coimbra (IPC) with number 161_CEIPC_2022.
Liver fibrosis and steatosis ultrasound screening: a FibroScan study (7 min)

Rute Santos; Coimbra / Portugal
100
Author Block: R. Santos, B. Fonseca; Coimbra/PT
Purpose: The aim of this study was to correlate the values of hepatic fibrosis and steatosis obtained through FibroScan with the
individual characteristics of the participants, aiming to identify risk factors associated with the presence or absence of these
conditions.
Methods or Background: Data collection was conducted through a screening initiative. The exclusion criteria were used: people
weighing more than 30 kg/m2, people under 18 years of age and participants who have already undergone kidney or liver
transplants. The population was invited to participate in the evaluation, and the entire procedure was explained to them. Volunteers
underwent a FibroScan examination and answered sociodemographic questions.
Results or Findings: Out of 169 participants, 29% were men, 31.5% were overweight and obese, 15.9% had high cholesterol, 2.9%
had diabetes, and 13% had hypertension (HTN). After analysis, 92 participants were selected for fibrosis evaluation and 107 for
hepatic steatosis evaluation. The majority of participants did not have fibrosis or hepatic steatosis (76.1% and 64.4%, respectively).
7.6% of participants had moderate to severe fibrosis, and 3.3% had cirrhosis. Regarding hepatic steatosis, 6.6% had the more severe
grades. Hepatic fibrosis was not associated with diabetes, cholesterol, HTN, sex, or age. Hepatic steatosis was associated with
diabetes, cholesterol, HTN, and age but was not related to the participants' sex.
Conclusion: Hepatic fibrosis and steatosis may be related to risk factors. It is important to screen the population so that an early
diagnosis of liver disease can be made and possible worsening and complications can be avoided.
Fibroscan is a technique that allows a rapid assessment of the liver and its changes, making it ideal for broad population screening.
Limitations: Some of the participants did not answer all the questions. The fibroscan examination was performed by more than two
radiographers.
Ethics committee - additional information: It was a student initiative, where only the data was used afterwards, so no ethical
approval was sought.
101
RPS 304 - MRI in thoracic imaging
Categories: Chest, Imaging Methods, Oncologic Imaging

CME Credits: 1
Moderator:
Jürgen Biederer; Seeheim-Jugenheim / Germany
Elexacaftor/Tezacaftor/Ivacaftor improves bronchial artery dilatation detected by magnetic resonance imaging in

patients with cystic fibrosis (7 min)
Lena Wucherpfennig; Heidelberg / Germany
Author Block: L. Wucherpfennig1, S. Wege1, H-U. Kauczor1, C. P. Heussel1, O. Sommerburg1, M. Stahl2, M. Mall2, M. Eichinger1, M.
Wielpütz1; 1Heidelberg/DE, 2Berlin/DE
Purpose: It was previously shown that magnetic resonance imaging (MRI) detects improvements in mucus plugging and bronchial
wall thickening, but not lung perfusion in patients with cystic fibrosis (CF) treated with elexacaftor/tezacaftor/ivacaftor (ETI). It is still
unclear, whether bronchial artery dilatation (BAD), a key feature of advanced lung disease, indicates irreversibility of perfusion
abnormalities and whether BAD could be reversed in CF patients treated with ETI.
Methods or Background: 59 adults with CF underwent longitudinal chest MRI including MR angiography (MRA) twice, 35 CF patients
(mean age 31±7 years) before (MRI1) and after (MRI2) at least one month (mean duration 8±4 months) on ETI therapy and 24 control
CF patients (mean age 31±7 years) without ETI. MRI was assessed using the validated chest MRI score, and presence and total lumen
area of BAD were assessed with commercial software.
Results or Findings: The MRI global score was stable in the control group from MRI1 to MRI 2 (mean difference: 1.1±3.4, P=0.054),
and was reduced in the ETI group (-10.1±4.2, P<0.001). In the control as well as in the ETI group, BAD was present in almost all
patients at baseline (95% and 94%, respectively) and presence did not change at MRI2. The BAD total lumen area was stable in the
control group from MRI1 to MRI2 (mean difference: 1.0±3.0 mm2, P=0.099), but decreased in the ETI group (-6.6±5.8 mm2,
P<0.001). This decrease correlated with improvements in the MRI global score (r=0.540, P<0.001).
Conclusion: Our data show that BAD may be partially reversible under ETI therapy in adult CF patients with established disease.
Limitations: The number of patients treated with ETI was only moderate. Moreover, our study was conducted in adults with CF,
excluding children with milder lung disease.
Funding for this study: This study was supported by grants from the German Federal Ministry of Education and Research
(82DZL00401, 82DZL004A1, 82DZL009B1), the German Research Foundation (STA 1685/1-1 to M.S.) and the Mukoviszidose e.V.
(S02/09, C-H-P 1504). Funders had no involvement in the collection, analysis and interpretation of data, in the writing of the report
and in the decision to submit the article for publication.
Ethics committee - additional information: This study was approved by the ethics committee of Heidelberg Medical Faculty.
Magnetic resonance imaging detects delayed perfusion inside perfusion defects in patients with cystic fibrosis (7 min)
Patricia Leutz-Schmidt; Heidelberg / Germany
102
Author Block: P. Leutz-Schmidt1, J. Grolig1, S. Triphan1, M. Mall1, O. Sommerburg1, A. Koeppe1, B. Nestler2, H-U. Kauczor1, M.
Wielpütz1; 1Heidelberg/DE, 2Karlsruhe/DE
Purpose: Dynamic contrast-enhanced perfusion MRI (DCE-MRI) detects perfusion abnormalities in patients with cystic fibrosis (CF)
from preschool age related to a reduction in inflow through pulmonary arteries. However, little is known about the alterations of
bronchial arterial inflow from the systemic circulation. We hypothesised that in perfusion defect areas a delayed perfusion from
bronchial arteries may be observed related to bronchial artery dilatation (BAD).
Methods or Background: Morpho-functional MRI incl. DCE-MRI from 75 patients with CF (mean age 15.3±5.5 years, range 6-29
years) were included. The lungs were segmented on coronal 3D T1 and registered onto DCE-MRI. The arterial input function (AIF) was
calculated automatically, in order to quantify pulmonary blood flow (PBF) and mean transit time (MTT). Perfusion defects were
classified in percent (QDP). Quantitative perfusion parameters were compared in areas with normal perfusion vs perfusion defects.
Results or Findings: Mean QDP was 35.4±23.0%. Mean PBF was 165.2±90.4 ml/100 ml per minute in normal vs 56.9±24.2
ml/100ml per minute in perfusion defect areas (P<0.001). Corresponding MTT was 5.4±1.7 seconds in normal vs 6.9±2.2 seconds in
defect areas (P<0.001). 48 CF patients had BAD, and 27 did not show BAD on DCE-MRI. Patients with BAD had higher QDP (46.2±21.3
vs 16.1±9.0, P<0.001), lower PBF (91.9±54.8 vs 178.3±77.4, P<0.001) and tended to have shorter MTT (5.5±1.2 vs 6.2±2.2, P<0.06)
for the whole lung. Considering defect areas only, MTT was shorter in patients with BAD vs without (6.3s ±1.8s vs 8.1s ± 2.6s,
P<0.001).
Conclusion: Lung areas with perfusion defects in the pulmonary arterial phase show delayed perfusion in the systemic arterial phase
in patients with CF. Patients with BAD have more severe perfusion defects and reduced PBF. Increased inflow through BAD can
potentially be quantified by MTT.
Limitations: Ground truth is missing for the validation of quantitative perfusion results.
Funding for this study: This study was funded by the German Center for Lung research (DZL)
This study was supported by grants from the German Federal Ministry of Education and Research (BMBF) (82DZL004A1).
Ethics committee - additional information: This study was approved by an ethics committee; the vote number is S-211/2011.
Delta-radiomics features of ADC maps as early predictors of treatment response in lung cancer (7 min)
Christian Marcel Heidt; Heidelberg / Germany
1 1 1 1 2 3 1
Author Block: C. M. Heidt , J. Bohn , O. von Stackelberg , O. Weinheimer , J. Vogel-Claussen , F. L. Giesel , H-U. Kauczor , C. P.
1 1 1 2 3
Heußel , G. Heußel ; Heidelberg/DE, Hannover/DE, Düsseldorf/DE
Purpose: The purpose of this study was to investigate the feasibility of detecting very early treatment-induced tumour tissue
changes in patients with advanced lung adenocarcinoma using diffusion-weighted MRI-derived radiomics features.
Methods or Background: This prospective study included 144 patients receiving TKI (n=64) or chemotherapy (n=80) for lung
adenocarcinoma. Patients underwent diffusion-weighted MRI at baseline as well as +1, +7 and +14 days after treatment initiation.
Radiomics features quantifying shape, textures and intensities were extracted from tumour ROI segmentations and feature changes
over time were analysed for correlation with treatment response (TR) according to RECIST and progression-free survival (PFS).
Results or Findings: Out of 14 selected delta-radiomics features, 6 showed significant correlations with PFS or TR. Most significant
correlations were found after 14 days. Features quantifying ROI heterogeneity, such as short run emphasis (P = 0.037 (pfs) / 0.006
(tr)), gradient short run emphasis (P = 0.058 (pfs) / 0.012 (tr)) and zone percentage (P = 0.019 (pfs) / 0.013 (tr)) increased in patients
with overall better TR. Patients with worse overall response and survival instead showed an increase in features quantifying ROI
homogeneity, such as normalised inverse difference (P = 0.012 (pfs)/ 0.043 (tr)). Clustering of these imaging features allows
stratification of patients into groups of longer and shorter survival.
Conclusion: As early as 2 weeks after initiation of treatment, diffusion MRI of lung adenocarcinoma reveals quantifiable tissue-level
insights that correlate well with future treatment (non-)response. Diffusion MRI derived radiomics thus shows promise as an early,
radiation-free decision-support to assess efficacy and potentially alter treatment course early.
Limitations: The study was performed on a small, single-centre data set, pending validation on external data.
Funding for this study: Funding was provided by the German Center for Lung Research (FKZ 82DZLF14A2).
Ethics committee - additional information: Approval was granted by Heidelberg Medical School Ethics committee Ethics Vote
S-445/2015.
3.0T MRI with functional sequences and respiratory motion-resolved compressed sensing reconstruction of free-
breathing radial acquisition for immediate response assessment of lung tumour postablation (7 min)
Danyang Zhao; Shanghai / China
103
Author Block: D. Zhao, L. Liu, X. Ye, k. liu, C. Yiheng, S. Yuanxin, F. Wang, H. Sun; Shanghai/CN
Purpose: The purpose of this study was to use serial magnetic resonance imaging (MRI) examinations to observe changes in
malignant lung tumours within a short time postablation.
Methods or Background: Patients with primary or metastatic lung tumours eligible for ablation were included in this study. Ablation
was performed according to standard procedures. Unenhanced and dynamic contrast-enhanced MRI scans were performed
preablation and in one week after ablation. Dynamic imaging were undergone with free-breathing imaging technique combining
compressed sensing (CS) and parallel imaging with golden-angle radial sampling instead of conventional breath-hold volumetric
interpolated breath-hold examination (VIBE) or liver acquisition volume acceleration (LAVA). At each time point, the signal intensity
and parameters of the ablation zone in both perfusion imaging and DCE imaging were examined and analysed.
Results or Findings: Using CS technique we obtained T1-weighted imaging with high spatial resolution without requiring patients to
hold their breath. A total of 19 nodules in 17 patients were included in the study. Among them, 16 nodules examined parameters of
IVIM. The ablation zone in MRI imaging has been shown to have a central area with complete tissue necrosis showed no enhancement
(19/19) on ceT1. The value of Ktrans (18/19), D*(10/16), iAUC90(19/19), CER(19/19) were significantly lower in the postablation zone
compared to the preablation tumour, which represents decreased blood flow perfusion. The increase in D value (11/16) and ADC value
(13/16) indicates that the diffusion of water molecules is less restricted compared to before.
Conclusion: Patients who cannot hold their breath autonomously can get high-quality images and stable parameter data by CS
technique. The ablative margin and the changes of tumour activity and microvascular perfusion can be analysed via functional MRI
images, which can be used to assess immediate response with technical success and ablative margin.
Limitations: The limitations of the study are the lack of cases and long-term follow-up data.
Ethics committee - additional information: The study is retrospective.
3.0T MRI for immediate response assessment of lung tumour postablation: a pilot study (7 min)
Danyang Zhao; Shanghai / China
Author Block: D. Zhao, L. Liu, X. Ye, k. liu, C. Yiheng, S. Yuanxin, F. Wang, H. Sun; Shanghai/CN
Purpose: The purpose of this study was to use serial magnetic resonance imaging (MRI) examinations to observe changes in
malignant lung tumours a short time postablation.
Methods or Background: Patients with primary or metastatic lung tumours eligible for ablation were included in this study. Ablation
was performed according to standard procedures. Unenhanced and dynamic contrast-enhanced MRI scans were performed
preablation and one week postablation. At each time point, the signal intensity of the ablated zone on both T1WI and T2WI images,
and parameters of the ablation zone in both perfusion imaging and dynamic contrast-enhanced imaging were examined, and changes
post-procedure were analysed.
Results or Findings: A total of 19 nodules in 17 patients were included in the study. Among them, 16 nodules examed parameters
of IVIM. The ablation zone in MRI imaging has been shown to have a central area with complete tissue necrosis and a peripheral rim
(PR). The PR appears on T2 sequences as strongly hyperintense and on ceT1 sequences as an enhancing ring. The inner zone was
isointense and showed no enhancement (19/19). The values of Ktrans (18/19), D*(10/16), iAUC90(19/19), CER(19/19) were
significantly lower in the postablation zone compared to the preablation tumour, which represents decreased blood flow perfusion.
The increase in D value (11/16) and ADC value (13/16) indicates that the diffusion of water molecules is less restricted compared to
before.
Conclusion: The ablative margin of the ablated zone was clearly displayed on T1WI imaging, and changes in parameters in functional
as well as diffusion imaging showed the decrease of tumour activity and microvascular perfusion, indicating that the target was
ablated completely. MRI can be used to assess immediate response with technical success and ablative margin.
Limitations: The limitations of this study are the insufficient number of cases examined and the lack of long-term follow-up data.
Feasibility of pulmonary MRI for lung nodule detection and quantification using AI-accelerated 3D gradient echo
imaging (7 min)
Alexander Wolfgang Marka; Munich / Germany
104
Author Block: A. W. Marka1, M. Steinhardt1, M. Graf1, A. Sauter1, K. Weiss2, M. Makowski1, D. C. Karampinos1, J. Gawlitza1, S.
Ziegelmayer1; 1Munich/DE, 2Hamburg/DE
Purpose: Pulmonary MR imaging has made substantial progress in detecting lung nodules through optimisation and developing new
sequences in recent years. Nevertheless, the widespread integration of MRI as an alternative method in lung cancer screening is
limited either by long scan times or insufficient signal-to-noise ratio.
Methods or Background: In this prospective trial, patients with benign and malignant lung nodules admitted between December
2021 and December 2022 underwent low-dose chest CT and pulmonary MRI using a 3D gradient echo sequence, accelerated with a
combination of parallel imaging, compressed sensing, and deep learning. 3D segmentation masks of 192 lung nodules were created
in both modalities, and quantitative morphological features were extracted from the segmentation masks. Two readers evaluated all
cases in a blinded setting and measured the most extensive lesion. The agreement and relationship between all features in both
modalities were assessed.
Results or Findings: A total of 38 patients (mean [SD] age, 65±13 years; 19 women [50%]) with 192 pulmonary nodules were
included and analysed. The average scan time was 2.5 minutes. The quantitative features showed a strong positive relationship
between both modalities (maximum nodule diameter (r = 0.99 95% CI [0.95, 1.02]), surface-area (r = 1.00 95% CI [0.96, 1.04]),
nodule volume (r = 1.00 CI [0.96, 1.04]). Mean differences in maximum diameter were 0.1 mm with a 95% CI of 2.09 mm and -1.89
mm. Nodule diameter measurements for the MR sequence showed excellent interrater agreement.
Conclusion: Pulmonary MRI with an accelerated 3D gradient echo sequence showed comparable quantitative measurements to low-
dose CT. The average examination time of 2.5 minutes is a relevant factor for applying MRI in lung cancer screening as an alternative
screening method.
Limitations: Our study included only 38 subjects with a consequently limited range of pathologies.
Ethics committee - additional information: This study was approved by the ethics committee of Technical University of Munich
School of Medicine.
Conjugate gradient reconstruction for pulmonary thin-section MRI with ultra-short TE: capability for acquisition time
reduction while maintaining image quality and nodule detection (7 min)
Yoshiharu Ohno; Toyoake / Japan
1 2 2 2 2 2 1 1 1 1
Author Block: Y. Ohno , K. Yamamoto , M. Ikedo , M. Yui , M. Shinohara , Y. Sano , H. Nagata , T. Ueda , Y. Ozawa ; Toyoake/JP,
2
Otawara/JP
Purpose: The purpose of this study was to compare the capability of conjugate gradient reconstruction (CG-Recon) for acquisition
time reduction, image quality and nodule detection performance with conventional grid reconstruction (Grid-Recon) on pulmonary
thin-section MRI with ultra-short TE (UTE-MRI) at in vitro and in vivo studies.
Methods or Background: Firstly, a commercially available NEMA phantom was scanned by UTE-MRI sequence under reduction of
sampling spoke numbers as follows: original (n=1: UTEoriginal), 1/2 (UTE1/2), 1/4 (UTE1/4) and 1/6 (UTE1/6). Then, each piece UTE-
MRI data was reconstructed using both methods. Secondly, 40 patients with lung nodule underwent thin-section CT and UTE-MRI by
same sequences, and all UTE-MRI data were also reconstructed with each method. To determine the influence of spoke number
reduction and reconstruction method on quantitative image quality at in vitro study, full width at half maximum (FWHM) of phantoms
was assessed. At in vivo study, lung signal-to-noise ratio (SNR), overall image quality, artifact and probability for nodule presence
were assessed by ROI measurement or 5-point scales. FWHM and SNR were compared among all UTE-MR data by Tukey's HSD test.
All qualitative indexes were compared among all UTE-MR data by Wilcoxon's signed rank test.
Results or Findings: FWHM, SNR overall image quality and nodule detection probability of each UTEoriginal and UTE1/2 with CG-
Recon were significantly better than those of others (P<0.05). Artifact of each UTEoriginal and UTE1/2 with CG-Recon were
significantly lower than those of others (P<0.05).
Conclusion: Conjugate gradient reconstruction is useful for reducing acquisition time while maintaining image quality and nodule
detection performance on pulmonary thin-section MRI with UTE.
Limitations: This was a single centre study with a limited study population.
Funding for this study: A research grant was received from Canon Medical Systems Corporation.
Ethics committee - additional information: This study was approved by the IRB of Fujita Health University Hospital.
Longitudinal morpho-functional MRI for the management of incidental pulmonary nodules in patients with COPD: a
nationwide multicentre trial (7 min)
Lin Zhu; Shanghai / China
105
Author Block: L. Zhu1, L. Qian2, H-U. Kauczor2, M. O. Wielpütz2; 1Shanghai/CN, 2Heidelberg/DE
Purpose: The purpose of this study was to evaluate the capability to accurately diagnose and predict the long-term outcome of
incidental pulmonary nodules in heavy smokers by MRI in multicentre trial research.
Methods or Background: Patients who participated in two rounds of same-day MRI and LDCT scans or who had histopathological
proofs were included from 16 centres nationwide. The accuracy of MRI in predicting nodule growth and lung cancer was verified by
LDCT and histopathological results, and the consistency of MRI and LDCT in judging the outcome of nodules was evaluated. MRI
parsimonious and fuller multivariable logistic regression models were constructed.
Results or Findings: 248 patients were analysed in the present study out of 567 patients after their first round exam, with 11
patients having had histopathological proofs after the imaging examination. MRI has similar detection accuracy to LDCT for patients
with malignant nodules and revealed a substantial intermethod agreement with LDCT on the long-term outcome judgment of
incidental nodules based on two rounds of imaging assessment (κ=0.78-0.80). Our MRI full model showed significant improvement in
the predictive accuracy of incidental nodule progression over the parsimonious model (AUC 0.91 vs 0.81, P＜0.05), and shows no
difference when compared to the LDCT model (P=0.45). This suggests that the MRI full model, which associated sociodemographic,
clinical and characteristics based on MRI imaging, has similar potential to LDCT in diagnosing incidental pulmonary nodules as well as
in progression prediction in high risk patients.
Conclusion: MRI plays a promising role in evaluating and predicting the outcome of incidental nodules in the at-risk population,
which may contribute to radiation management and public health improvement.
Limitations: Most of the included incidental nodules were small and without surgery or biopsy, so LDCT served as the standard of
reference for this part.
Funding for this study: This work was supported in part by grants from the German Federal Ministry of Education and Research
BMBF to the German Center for Lung research (82DZL004A1, 82DZL009B1).
Ethics committee - additional information: This research was approved by the ethics committee of the university of Heidelberg.
106
RPS 308 - Temporal bone and dental imaging
Categories: EuroSafe Imaging/Radiation Protection, Head and Neck, Imaging Methods

CME Credits: 1
Moderator:
Beatrix Kovacsovics; Linkoping / Sweden
Radiologic prevalence and distribution of otic capsule dehiscence in patients with third window syndrome symptoms
and a new classification (7 min)
Yağmur Başak Polat; Istanbul / Turkey
Author Block: A. Yenigun, Y. B. Polat, E. Polat, S. Balsak, M. Basoz, S. Tugrul, O. Ozturan; Istanbul/TR
Purpose: The most well-known type of otic capsule dehiscence is Superior Semicircular Canal Dehiscence (SSCD). However, several
other types of otic capsule dehiscence affect the semicircular canals, cochlea, and vestibule. Our research aimed to examine the
frequency, distribution, and correlation between radiologic otic capsule dehiscence in patients exhibiting symptoms of third window
syndrome. Additionally, we introduced a new classification system for semicircular canal dehiscence (SCD).
Methods or Background: In this retrospective study, we included cases who applied to the ENT department of our university
hospital between January 2015 and September 2023 and underwent standard reformations and Pöschl plane CT scans due to
symptoms suggestive of third window syndrome. A head and neck radiologist and a general radiologist jointly assessed each CT and
decided on measurements and classifications.
Results or Findings: In the study, 219 patients (438 temporal bones) were examined. Semicircular canal dehiscences (SCD) were
categorised into four types: type 0 (no SCD), type 1 (unilateral single canal dehiscence), type 2 (bilateral single canal dehiscence),
and type 3 (unilateral multiple localisation dehiscence), and type 4 (bilateral multiple localisation dehiscence). SCD was observed in
69/219 (31.5%) patients; 150 were type 0 (68.5%), 29 type 1 (13.2%), 21 type 2 (9.6%), 15 type 3 (6.9%) and 4 were type 4 (1.8%).
Cochlear-fascial dehiscence (CFD) and Vestibular aquaduct-jugular bulbus dehiscence (VA-JBD) were seen in 56/219 (25.6%) and
22/219 (10%) patients, respectively. Type 2 and Type 4 were significantly more frequent in cases with CFD than other types. VA-JBD
was seen significantly more in Type 0 cases than in other types.
Conclusion: When we examine the otic capsule, we see that the possibility of CFD increases in bilateral SCD cases. The radiologist
should evaluate the otic capsule as a whole. Particular attention should be paid to multiple channels, bilateral localisation as well as
cochlear and vestibular dehiscences.
Ethics committee - additional information: This study received institutional review board approval and the reference number is
E-54022451-050.05.04-125457.
Diagnostic value of cochlear nerve characteristics for occupational noise-induced hearing loss (7 min)
Nan Wang; Tianjin / China
107
Author Block: N. Wang, L. Liu, L. Zhang, Y. Liu, L. Liu, J. Shi; Tianjin/CN
Purpose: The aim of this study was to explore the diagnostic value of cochlear nerve characteristics in the diagnosis of occupational
noise-induced hearing loss (NIHL).
Methods or Background: This prospectively observational study included patients with NIHL, patients with noise exposure, and
healthy volunteers at our hospital between January 2021 and September 2022. Multiplanar reconstruction of bilateral cochlear nerves
was performed to extract the cochlear nerves characteristics, including the long diameter, short diameter, average diameter, length-
to-diameter ratio, cross-sectional area, and perimeter of the cochlear nerves at three different planes, and the diagnostic value for
NIHL were evaluated by area under the receiver operating characteristic curve (AUC).
Results or Findings: The NIHL (n=27, 24 males, aged 47.9±6.2 years), noise-exposed (n=41, 34 males, aged 46.9±6.7 years), and
healthy control (n=27, 23 males, aged 48.0±6.1 years) groups were included. In the bottom plane, the short diameter (0.55±0.15 vs
0.75±0.16 mm), long diameter (0.83±0.18 vs 1.07±0.23 mm), average diameter (0.69±0.15 vs 0.91±0.18 mm), area (0.39±0.15 vs
0.72±0.21 mm2), and perimeter (2.33±0.48 vs 3.08±0.58 mm) of the NIHL group were significantly lower than those of the noise-
exposed group (all P<0.05). The bottom plane area showed the best diagnostic value, with AUC of 0.921 (95% CI: 0.855-0.987).
Conclusion: Cochlear nerve characteristics might be helpful in the diagnosis of NIHL, and the bottom plane area showed the best
diagnostic value.
Limitations: The sample of NIHL patients was relatively small; more patients need to be included in the future to evaluate the
diagnostic efficacy of quantitative parameters of the cochlear nerve. Some patients' nerves were too slender, and so were smaller
than the resolution of MR 3D-SPACE T2WI sequences. Some volume effects resulted in some layers of the cochlear nerves not being
displayed, so those results could not be included in this study.
Funding for this study: MR 3D-SPACE T2WI sequences can measure the quantitative parameters of the cochlear nerve in NIHL
patients in non-invasive and radiation-free conditions. These parameters can help with clinical differential diagnosis, distinguishing
between NIHL patients and noise-exposed patients. The bottom plane had the highest diagnostic efficiency.
Ethics committee - additional information: This study was approved by the clinical application project ethics committee of Tianjin
occupational diseases precaution and therapeutic hospital.
Temporal bone fractures and related complications in patients with cranio-facial trauma: assessment with MDCT in the
acute emergency setting (7 min)
Marcella Pucci; Geneva / Switzerland
Author Block: M. Pucci, M. Becker, P. Senn, P. Scolozzi, P-A. A. Poletti, A. Platon; Geneva/CH
Purpose: The purpose of this study was to analyse the prevalence and complications of temporal bone fractures in adult and
paediatric patients evaluated for cranio-facial trauma in the emergency setting.
Methods or Background: A retrospective blinded analysis was performed based on the CT scans of a series of 294 consecutive
adult and paediatric patients with cranio-facial trauma investigated in the emergency setting. Findings between the two populations
were compared. The preliminary reports done by the on-call residents were compared with the retrospective analysis by an
experienced reader, which served as a reference standard.
Results or Findings: CT revealed 126 fractures in 116/294 (39.5%) of patients although fractures were suspected clinically only in
70/294 (23.8%), P<0.05. Fractures were longitudinal, transverse and complex in 69.5%, 10.3 % and 19.8 % of cases, respectively.
Involvement of the tympanic cavity, external auditory canal, ossicular chain, facial nerve canal, and otic capsule was present in 54%,
72.2%, 8.7%, 6.3% and 4.8%, respectively. Injuries of the venous sinuses and carotid canal were seen in 18.3% and 17.5% of
temporal bone fractures. Vascular injuries were more common in children than in adults (31.8% vs 15.7%, P<0.05). 79.5% of patients
with temporal bone fractures had both brain injuries and fractures of the facial bones and cranial vault. Compared to the experienced
reader, the sensitivity of the on-call residents was 103/116 (88.7%).
Conclusion: Temporal bone fractures and related complications in particular vascular injuries are common in patients with cranio-
facial trauma and need to be thoroughly looked for, especially in children.
Limitations: This is a retrospective study. Possible underestimation of the true prevalence of temporal bone trauma due to the
selection of patients based on the PACS records.
There was a lack of analysis of the initial and long-term clinical assessment.
Ethics committee - additional information: The institutional Ethics Committee approved this study.
Deep learning reconstruction algorithm with ultra-high resolution CT of the temporal bone: initial experience (7 min)
Romain Gillet; Nancy / France
108
Author Block: A. Beysang1, N. Villani1, F. Boubaker1, U. Puel1, A. Blum-Moyse1, P. A. Teixeira1, K. Haioun2, C. Parietti1, R. Gillet1;
1
Nancy/FR, 2Otawara/JP
Purpose: The purpose of this study was to evaluate the image quality and clinical acceptance of a deep learning reconstruction
(DLR) algorithm compared to traditional iterative reconstruction (IR) and simulated normal resolution (NR) reconstruction algorithms.
Methods or Background: CT acquisitions were performed using an ultra-high resolution CT device with a Mercury 4 phantom at
three dose levels (25, 40, 51 mGy) and on 13 temporal bones at our routine dose level (about 500 mGy.cm). Images were
reconstructed with two IR algorithms (model-based (MBIR) and hybrid (HIR)), one DLR algorithm with 0.25 mm slice thickness, and
one simulated NR algorithm. In phantom acquisition, detectability was compared. Three radiologists performed quantitative (stapes
footplate and superstructure thicknesses and lengths) and qualitative (posterior canaliculus, tympanic chorda tympani, stapes
footplate and superstructure depiction) measurements in human temporal bones. Mean noise values in the external auditory canal air
and lateral semi-circular canal bony island were compared.
Results or Findings: Using DLR, detectability was superior to other reconstruction algorithms at the three dose levels. All
thicknesses were significantly superior using NR compared to DLR and IR algorithms, whereas lengths did not differ. DLR and MBIR
were superior to NR and HIR for posterior canaliculus and stapes superstructure and footplate depiction. DLR was superior to all other
algorithms for tympanic chorda tympani depiction. Compared to NR, there was twice as much noise in the air using HIR, but DLR and
MBIR yielded significantly lower noise values. Bony noise values were lower using DLR and MBIR than with NR and HIR, which
provided equivalent values.
Conclusion: DLR yielded a gain in objective measurement, closer to anatomical data, and in noise with the best clinical acceptance
among the evaluated reconstruction algorithms.
Limitations: There was no surgical confirmation and the number of patients was small.
Ethics committee - additional information: This study was approved by the ethics committee of CHRU de Nancy.
Synthetic CT images of temporal, sinonasal and facial bones from MRI (7 min)
Marlise Daniëlla Van der Veen; Amsterdam / Netherlands
Author Block: M. D. Van der Veen1, T. Van der Velden2, P. R. Seevinck2, J. Kuijer1, B. Schulten1, G. Adriaensen1, T. Goderie1, P.
Merkus1, B. Jasperse1; 1Amsterdam/NL, 2Utrecht/NL
Purpose: Diagnostic evaluation and surgical planning of the head frequently require visualisation of both soft and bony tissues.
Currently, this means a CT and MRI have to be obtained separately and then registered to combine the information from the two
images.
The objective of this study is to train and evaluate a machine learning algorithm to generate synthetic bone CT images from MRI of
the head.
Methods or Background: Paired 3T MRI (Philips, Siemens, GE) and CT scans of the head were used to train a vendor agnostic
machine learning algorithm to generate synthetic CT images from the MRI data.
MRI scans of patients not included in algorithm training were used to create synthetic CT images, which could be compared to their
corresponding true CT images.
A technical voxelwise comparison between synthetic CT and true CT was performed to quantify morphological and radiodensity
accuracy.
To clinically evaluate the algorithm, six surgeons and two radiologists will evaluate the visibility of clinically relevant landmarks for
three areas (temporal, sinonasal and facial bones) on both synthetic CT and true CT on a 4-point Likert scale.
Results or Findings: Voxelwise comparison showed a surface distance error of 0.38 ± 0.37 mm for the skull and a mean
radiodensity error of 4 ± 44 HU, with a correlation coefficient of 0.79 ± 0.08.
In-depth clinical analysis of the images is currently ongoing and will be presented during the ECR 2024.
Conclusion: Generating synthetic CT images of the head from MRI is feasible, allowing inherently registered and efficient
visualisation of soft and bony tissues using a single, radiation-free imaging modality.
Limitations: A limitation of the study is suboptimal blinding during evaluation of the scans, as synthetic CT and true CT can be
visually distinguished by experienced physicians.
Funding for this study: Funding was provided by Amsterdam UMC, innovation impulse and by the research institute Amsterdam
Public Health, within their programme Quality of Care.
Ethics committee - additional information: The study was approved by the institutional review board of Amsterdam UMC,
location VUMC (study number: 2022.0234). Written informed consent was obtained from all participants.
The value of dual-layer spectral CT in the detection of the temporal bone cholesteatoma (7 min)
Xiaoxue Fan; Shenyang / China
109
Author Block: X. Fan, X. Lu, C. Ding; Shenyang/CN
Purpose: The aim of this study was to evaluate the diagnostic accuracy of quantitative parameters from dual-layer spectral CT
(DLSCT) to identify cholesteatoma.
Methods or Background: In the prospective study, patients with clinically suspected cholesteatoma who underwent presurgical
DLSCT and subsequently underwent surgical treatment and pathology were enrolled between August 2023 and October 2023. The
subjects were divided into cholesteatoma and non-cholesteatoma groups. Quantitative parameters (virtual monoenergetic images at
40 to 70keV, effective atomic number (Zeff) and 40-70 spectral attenuation curve slope) derived from DLSCT were measured for both
groups. Three experienced radiologists independently performed the quantitative parameters to evaluate reproducibility.
Interobserver and intraobserver agreement were calculated using Cohen's coeffient (κ). Diagnostic accuracy of DLSCT quantitative
parameters was calculated using receiver operator curves.
Results or Findings: 24 patients were included (14 patients with cholesteatoma and 10 patients with non-cholesteatoma). There
was significantly higher 40keV and Zeff in cholesteatoma than that in non-cholesteatoma (P < 0.001 each). The area under the
receiver operating characteristics curve was slightly higher for 40keV than that of Zeff, however, there was no statistically significant
difference (0.921 vs 0.829; P= 0.2465, respectively). The optimal energy level for detecting cholesteatoma was 40 keV, and the
optimal threshold of 24 HU gave sensitivity of 78.6%, specificity of 100.0%, with an area under the curve of 0.921 (P <0.001).
Conclusion: Quantitative DLSCT parameters can help differentiate cholesteatoma from other inflammatory lesions of the ear, with
40kev images may provide an accurate demonstration of temporal bone cholesteatoma.
Limitations: This study included a small sample size and did not compare the diagnostic efficacy of cholesteatoma with that of DWI.
Funding for this study: The authors state that this work has not received any funding.
Ethics committee - additional information: This prospective study was approved by the local hospital's Institutional Review Board
(approval number: 2022PS898K).
Dental imaging in clinical photon-counting CT at a quarter of DVT dose (7 min)

Stefan Sawall; Heidelberg / Germany
Author Block: S. Sawall1, J. Maier1, S. Sen2, H. Gehrig1, T-S. Kim1, H-P. Schlemmer1, S. O. Schönberg3, M. Kachelrieß1, M. Rütters1;
1
Heidelberg/DE, 2Kiel/DE, 3Mannheim/DE
Purpose: The main objective of this study was to investigate the image quality of a low-dose dental imaging acquisition protocol in
the first clinical photon-counting computed tomography (PCCT) system in comparison to a normal-dose acquisition in a digital volume
tomography (DVT) system.
Methods or Background: Clinical PCCT systems offer an increased spatial resolution compared to previous generations of clinical
systems. Its spatial resolution is in the order of dental DVT systems. Resolution-matched acquisitions of ten porcine jaws were
performed in a PCCT (Naeotom Alpha, Siemens Healthineers) and in a DVT (Orthophos XL, Dentsply Sirona). PCCT images were
acquired with 90 kV at a dose of 1 mGy CTDI16 cm. DVT used 85 kV at 4 mGy. Image reconstruction was performed using the
standard algorithms of each system to a voxel size of 160×160×200 μm. The dose-normalised contrast-to-noise ratio (CNRD) was
measured between dentine and enamel as well as dentine and bone. Two experienced readers qualitatively evaluated overall
diagnostic quality of images and quality of relevant anatomical structures such as root channels and dentine. Reproducibility was
assessed using the intraclass correlation coefficient (ICC).
Results or Findings: CNRD is significantly higher in all PCCT acquisitions. In particular, CNRD is 37% higher for the contrast dentine-
enamel and 31% higher for the dentine-bone contrast (p<0.05). Overall diagnostic image quality was higher for PCCT over DVT
(p<0.02 and p<0.04 for readers 1 and 2). Quality scores for considered anatomical structures were higher in PCCT compared to DVT
(all p<0.05). Inter- and intrareader reproducibility were acceptable (all ICC>0.75).
Conclusion: PCCT provides an increased image quality over DVT even at far lower dose levels.
Limitations: The study is limited to porcine ex-vivo samples since multiple measurements are required. The resulting intersection
lengths might not accurately reflect the clinical case of a human patient.
Ethics committee - additional information: No approval was required for this study since the used porcine jaws are not
considered an animal experiment.
Assessment of dental pulp by T2 mapping in vivo is influenced by age (7 min)

Ksenija Cankar; Ljubljana / Slovenia
110
Author Block: K. Cankar, A. Golez, A. Tenyi, K. Romaric, T. Robida, J. Vidmar; Ljubljana/SI
Purpose: The cone beam computed tomography (CBCT) provide only information on hard dental tissues. This in vivo study tested
whether standard 3T clinical MRI system can be used to quantify the dental pulp state using the T2 mapping method.
Methods or Background: Multi-echo-spin-echo (MESE) T2-weighted sequence in a single sagittal slice and with a field of view that
covered the pulp chamber was employed. The teeth of the 15 subjects were divided in two groups according to their age. In the first
group, children with the age between 11 and 15 were included (mean age 12.6±1.3 years) and in the second group were adults with
the age between 24 and 48 (mean age 33.8±8.3 years). All T2 maps were calculated from the images of single-rooted premolar teeth.
In the present study, only intact teeth without caries were included.
Results or Findings: In the group of adult’s teeth, there were statistically significantly higher T2 relaxation times in dental pulp
(125.6±19.1 ms) compared to T2 relaxation times obtained in children’ teeth (111.9±11.2 ms) (p=0.022). In contrast, in the group of
adult’s teeth there was lower variability of dental pulp tissue relaxation times (35.0±4.9 ms) compared to the pulp tissue in children’s
teeth (44.6±8.4 ms) (p=0.0006).
Conclusion: An increase in the T2 relaxation times and a decrease in their variability in the adults could be attributed to lower cell to
water ratio and increased homogeneity in the pulp tissue. The study confirmed that in vivo T2 mapping of dental pulp can be used to
reliably quantify the dental pulp state. In healthy subjects, dental pulp assessment is strongly influenced by age.
Limitations: The teeth were not all from different subjects to maximize the number of teeth in the same field of view.
Funding for this study: This study was funded by the Ministry of Higher Education, Science and Technology, Slovenia: Grant
No.P3-0019.
Ethics committee - additional information: The study has been independently reviewed and approved by the Ethical Committee
of the National Ministry of Health (Approval number 0120-659/2016/6).
111
RPS 312 - Foetal and paediatric thoraco-abdominal imaging
Categories: Cardiac, Chest, GI Tract, Oncologic Imaging, Paediatric

CME Credits: 1
Moderator:
Maria Rita Cabrita Carneiro; Lisboa / Portugal
Fractal analysis of perfusion in neuroblastoma using MRI predicts volumetric tumour response and outcome after
induction chemotherapy (7 min)
Florian Michallek; Berlin / Germany
Author Block: F. Michallek, M. Dewey, A. Eggert, T. M. Thole-Kliesch; Berlin/DE

Purpose: Angiogenesis is a hallmark of cancer, however, evidence on the role of tumour perfusion assessment in paediatric patients
with neuroblastoma is sparse, although contrast-enhanced magnetic resonance imaging (MRI) is routinely used in neuroblastoma
patients. Fractal analysis allows to characterise perfusion patterns on routine contrast-enhanced MRI, which has potential as a non-
invasive imaging biomarker for predicting treatment response and outcome by providing information on vascular changes during
induction chemotherapy.
Methods or Background: In a national multi-centre setting (30 centres), we retrospectively performed fractal analysis of contrast-
enhanced, fat-saturated, T1-weighted MRI. Patients underwent MRI at the timepoint of diagnosis (TP1) and prior to surgical tumour
resection (TP2). At both timepoints, fractal analysis was performed to quantify complexity of the tumour perfusion pattern by fractal
dimension (FD). At TP1, we investigated, whether fractal analysis allows to predict local tumour response. At TP2, we evaluated
prognostic implications of fractal analysis for predicting event-free-survival (EFS). Additionally, we applied fractal analysis to CD34-
immunohistochemistry in a subset of patients to correlate imaging findings with underlying vascular structure. We adhered to the
STARD guideline.
Results or Findings: We included n=73 patients in separate discovery (n=36; single-centre) and validation cohorts (n=37; 29
centres). At TP1, fractal analysis predicted relative volumetric tumour response with mean bias -0.05 (limits of agreement: -0.35–0.25)
using Bland-Altman analysis. Areas with low FD at TP1 showed high tumour response, while high FD-areas tended to persist. At TP2,
fractal analysis separated patients with and without favourable EFS in Kaplan-Meier-Analysis (p=0.0019) in tumours lacking MYCN-
amplification. In CD34-immunohistochemistry, FD increased significantly after induction chemotherapy (FD-TP1: 1.23±0.09, FD-TP2:
1.44±0.07, p<0.001).
Conclusion: In a retrospective multi-centre setting, fractal analysis of perfusion predicted volumetric response and outcome after
induction chemotherapy in patients with neuroblastoma.
Limitations: Prospective validation is lacking making this a limitation of the study.
Ethics committee - additional information: Neuroblastoma Registry 2016 German Clinical Trial Register with the DRKS-ID:
DRKS00023442; Ethics committee number: 16-432.
Feasibility and diagnostic accuracy of early postoperative MRI after resection of neuroblastic tumours (7 min)
Maryanna Chaika; Tuebingen / Germany
112
Author Block: M. Chaika, M. Esser, P. Krumm, R. Guglielmi, S. Gassenmaier, C. Urla, S. Warmann, J. F. Schäfer; Tübingen/DE
Purpose: In standard MRI, it may be problematic to differentiate residual tumour (RT) from scar, reactive changes, or recurrence
after three months. The aim of this study was to evaluate a short MRI protocol performed in the early postoperative period.
Methods or Background: The inclusion criteria were: histologically confirmed neuroblastic tumour, resection by the reference
surgery in our centre, early postoperative MRI and MRI preoperatively, and standardised MRI Protocol. The MRI protocol included the
following sequences: T1w vibe Dixon before and after contrast, T2w with fat saturation and DWI with the calculation of the ADC map.
The analysis was performed by 3 independent readers (resident/adults radiologist/board certified pediatric radiologist) using a 4-
points Likert-scale . A multimodal reference standard was determined by: F/U imaging, consensus tumour-board, and consensus
between senior radiologist and surgeon.
Results or Findings: Thirty-nine patients with a median of 46 age months (5-177). MRI was performed in the mean 8(+/- 5) days
after surgery. RT was found in 13 patients by MRI and confirmed by the reference standard, with a typical location at the mesenteric
root and retrocrural. 4 RTs were expected by the surgeons with a median volume of 19 ml (1-34 ml), and nine tumours were
unexpected with a median volume of 1 ml (0.25-7). Sensitivity, specificity, and accuracy (reader) were 77,54, and 70 %(Resident),
81,85, and 82 %(Adult Radiologist), 92,92, and 92 %(Pediatric Radiologist). Reading the postoperative MRI alone, the diagnostic
performance of the paediatric radiologist were 88,74, and 84%.
Conclusion: Early MRI protocol is feasible for determination of residual tumour. Experience in pediatric imaging is crucial to
achieving high diagnostic precision. Reading the preoperative MRI improves diagnostic accuracy.
Limitations: Small sample size. The results are preliminary and need to be confirmed by follow-up studies.
Funding for this study: No funding from any specific grant was received for this study.
Ethics committee - additional information: The local institutional review board (Ethics Committee at the Medical Faculty of the
Eberhard Karls University, Tuebingen, Germany) approved this retrospective, monocentric study.
Neonatal pulmonary MRI at term-equivalent age for defining bronchopulmonary dysplasia severity and its relationship
with short-term respiratory outcomes (7 min)
Gianluca Folco; Imperia / Italy
Author Block: A. M. Munari1, G. Folco1, C. B. Monti1, F. Rizzetto1, N. A. R. Panarisi2, S. Zirpoli1; 1Milan/IT, 2Assago/IT
Purpose: This prospective study aims to demonstrate that neonatal pulmonary magnetic resonance imaging (MRI), using
commercially available sequences, can assess lung parenchymal injury associated with bronchopulmonary dysplasia (BPD) and may
predict short-term respiratory outcomes.
Methods or Background: Pulmonary MRI, using turbo spin echo and gradient echo sequences during natural sleep, was performed
in premature infants who underwent routine brain MRI at term-equivalent age. Lung MRIs were scored using the modified Ochiai
scoring system and a new proposed score, the Bronchopulmonary Dysplasia Magnetic Resonance Index (BPDMRI), based on four
hyperaeration (hyperexpansion, emphysema, cysts, mosaic pattern) and four parenchymal variables (triangular subpleural opacities,
fibrous stripes and bands, edema and atelectasis, distortion of bronchovascular bundles). Both scores were correlated with two short-
term respiratory outcomes: the length of mechanical ventilation and the length of any respiratory support until discharge.
Results or Findings: The study included 25 premature infants (9 with BPD, 36%). Pulmonary MRI showed a wide variety in
appearance of pulmonary parenchyma among BPD patients and revealed significant structural differences across the range of MRI
scores. Both the modified Ochiai score and the BPDMRI resulted positively correlated to the length of mechanical ventilation (r=0.828,
p<0.001; r=0.900, p<0.001) and the length of any respiratory support until discharge (r=0,897, p<0.001; r=0.953, p<0.001). In the
severe disease group, the length of respiratory support showed a smaller IQR when related with the BPDMRI rather than when related
with the clinical definition.
Conclusion: Pulmonary MRI identified lung parenchymal abnormalities associated with BPD, showing a strong correlation with the
length of mechanical ventilation and the length of any respiratory support until discharge.
Limitations: Single centre, retrospective study is the limitation of this study.
A prediction model based on CT radiomics for the tri-classification of mycoplasma pneumonia, bacterial pneumonia,
and viral pneumonia (7 min)
Lulin Bi; Taiyuan / China
113
Author Block: L. Bi, S. Xu, L. Wei, M. Guo, S. Quan; Taiyuan/CN
Purpose: This study aimed to investigate the diagnostic efficacy of machine learning-based CT radiomics models established by
different machine learning algorithms in identifying mycoplasma pneumonia, bacterial pneumonia, and viral pneumonia in children.
Methods or Background: A retrospective analysis was performed for 835 children with pneumonia confirmed by etiology in Shanxi
Children's Hospital from November 2019 to February 2022, including 297 cases of mycoplasma pneumonia, 336 cases of bacterial
pneumonia and 202 cases of viral pneumonia. All cases were randomly divided into a training set (n=584) and a testing set (n=251)
at a ratio of 7:3. The CT images of each child were automatically segmented using the VB-Net model to generate the region of
interest (VOI) of the whole lung region, and the radiomics first-order features, texture features and filtering features were extracted.
Univariate feature selection (UFS) and mutual information feature selection (MI) were used for feature screening. Six machine
algorithms were used to establish a classification prediction model by logistic regression (LR), support vector (SVM), random forest
(RF), K-nearest neighbor (KNN), decision tree (DT) and Bayes. Model performance was evaluated by area under the ROC curve (AUC).
Results or Findings: A total of 386 radiomic features were extracted from the CT images of each child. Five superior features were
screened out by UFS and MI. The six machine learning models had discrimination ability, among which the DT model had the best
predictive ability, with AUC of 0.918, 0.806 and 0.823 for predicting mycoplasma pneumonia, bacterial pneumonia and viral
pneumonia in the training set and 0.816, 0.741 and 0.876 in the validation set.
Conclusion: CT radiomics based on machine learning can effectively distinguish between the three types of pneumonia, and the DT
model is better than other machine learning models.
Funding for this study: This study was supported by the application of machine learning-based fusion models in the development
of a diagnosis and treatment system for novel coronavirus-infected pneumonia in children (the Shanxi Provincial Health Commission
No. 2023019).
Ethics committee - additional information: This retrospective study was approved by the Ethics Committee of Shanxi Children's
Hospital, and the requirement to obtain informed consent was waived (IKB-KYYN-2023-007).
CMR Assessment of cardiac remodelling in children with chronic kidney disease (7 min)
Si Si Song; Deyang / China
Author Block: S. S. Song1, L. Xie2, H. Xu2, Y-K. Guo2; 1Deyang/CN, 2Chengdu/CN

Purpose: The aim of this study was to investigate the characteristics of cardiac remodelling in children with CKD and to explore the
clinical factors that may contribute to cardiac remodelling. To analyse whether CMR parameters of cardiac remodelling can predict
clinical adverse events.
Methods or Background: This study prospectively enrolled 138 children with CKD (93 with CKD stage 1–2 and 45 with CKD stage
3–5) and 50 healthy controls. The main cardiac parameters measured included left ventricular mass index(LVMI), left ventricular
remodelling index (LVRI). Clinical adverse events included CKD progression and all-cause mortality.
Results or Findings: About 36.2%(50/138) of children with CKD had cardiac remodelling. LVMI and LVRI tended to increase with
increasing stage of CKD (all P< 0.05). 23.7% of children with CKD stage 1–2 had cardiac remodeling, which was mainly concentric
remodeling (20.4%). The proportion of remodeling increased to 62.2% in CKD stage 3–5.
Multiple linear regression analysis showed that systolic blood pressure was independently correlated with LVMI (β = 0.391), and LVRI
(β = 0.003) (all p < 0.05). Haemoglobin was independently correlated with LVMI (β = -0.176, p < 0.001). eGFR and proteinuria grade
were independently correlated with LVMI (eGFR: β = -0.047, proteinuria grade: β = 1.741) and LVRI (eGFR: β = 0.000, proteinuria
grade: β = 0.033) (all p < 0.05).
Multifactorial Cox regression analysis showed that LVMI was an independent predictor of adverse events in CKD (HR: 1.091; 95% CI:
1.015–1.173; p = 0.018).
Conclusion: Cardiac remodelling was prevalent in children with CKD, even in the early stages. Possible influencing factors for cardiac
remodelling included proteinuria, hypertension, and anaemia. LVMI was an independent predictor of clinical adverse events in CKD.
Limitations: This is a single-center study with a limited sample size.
Funding for this study: This work was supported by National Natural Science Foundation of China (82120108015,
82102020，82071874, 81971586) and Sichuan Science and Technology Program (2020YJ0029, 2017TD0005).
Ethics committee - additional information: This study protocol was approved by the Institutional Review Board of West China
Second University Hospital of Sichuan University (K2019061).
Applicability of the O-RADS scoring system in children and adolescents: results of a preliminary retrospective
monocentric study (7 min)
Elisa Mercanzin; Bergamo / Italy
114
Author Block: E. Mercanzin1, P. Clauser2, V. Rizzo3, S. Sironi1, A. Hojreh2; 1Bergamo/IT, 2Vienna/AT, 3Maglie/IT
Purpose: The study aims to assess the accuracy of the Ovarian-Adnexal Reporting & Data System (O-RADS) classification on
ultrasound (US) and magnetic resonance imaging (MRI) in paediatric patients.
Methods or Background: In this single-centre retrospective study, all US and MRI studies performed between 2005 and 2022 for
suspected ovarian lesions in patients < 18 years were included. Histopathology was considered as standard of reference. Two readers
in consensus evaluated imaging characteristics as defined by the O-RADS (features, size, wall thickness, ascites). Each lesion was
then assigned an O-RADS category. Chi-square test was used, sensitivity, specificity, and accuracy were calculated using a cutoff >O-
RADS 3, and compared.
Results or Findings: Included in this study were 95 patients (mean age 12,6, range 0-18). Histology found 45 non-neoplastic cystic
masses (47.4%), 47 benign neoplasms (49.5%) and 3 malignant neoplasms (3.2%). MRI was available in 80 cases and US in 72 cases.
No significant difference in the US features was found between malignant and benign lesions (p>0.2). On MRI a significant difference
was found between benign and malignant masses in terms of lesion diameter (p=0.02) and enhancement (p=0.025). MRI correctly
classified all malignant cases (3/3, 100%), while US only 1/2 (50%). Specificity was comparable between modalities (respectively 89%
vs 88%, p<0.0001). Accuracy was higher for MRI (87% vs 90%, p<0.0001).
Conclusion: Our results showed that the MRI O-RADS demonstrated a high diagnostic accuracy and can be applied in paediatric
patients. In this population, a higher specificity would be desirable to reduce false positives and unnecessary follow up or surgery. A
lower accuracy was found for US.
Limitations: This was a retrospective study, its limitations are the retrospective evaluation of the US images, low number of
malignant lesions.
Funding for this study: No research funding was received for this study.
Ethics committee - additional information: Ethics committee: Medical University of Vienna 2057/2017 approved this study.
The distal colostogram for anorectal malformation: considerations in low- and middle-income countries (7 min)
Miriam Leiderer; Passau / Germany
Author Block: M. Leiderer; Passau/DE

Purpose: The purpose of this study was to evaluate imaging challenges of anorectal malformations (ARM) in older male children and
those with prior loop colostomy.
Methods or Background: Treatment of ARM in low- and middle-income countries suffers from limited resources in paediatric
surgery. Children undergo colostomy in the newborn period, and definitive treatment is delayed. Consequently, children are older
than the recommended age for definitive reconstruction at time of surgery. A loop colostomy adds additional technical challenges. We
reviewed all colostograms obtained for patients with ARM presenting for definitive repair at a specialised hospital in Uganda, between
April and August 2023.
Results or Findings: Twenty-three patients were included, twelve male, of which seven with suspicion of rectourinary fistula. Three
of these were up to 12 m (5 m, 9 m, 12 m), in all of which the fistula tract was visualised. No fistula tract was visualised in older
patients (22 m, 3 y, 4y , 7 y), however, in three of these the fistula origin was seen as a small anterior tapering of the rectal pouch. In
one patient with no sign of fistula the absence of a fistula was confirmed during surgery. Fourteen patients (61%) had a fecaloma in
the distal pouch.
Conclusion: For all the patients with suspected genitourinary fistula older than 12 m the fistula did not opacify, an age group never
expressly reported previously. Possible explanations are occlusion by accumulated feces and mucus or through colonic distention;
and ability of older children to react to abdominal pressure by tensing the striated perirectal muscular complex. Fecaloma because of
loop colostomy was common, leading to chronic rectal distension associated with a poorer functional outcome and increased infection
risk.
Limitations: The small number of cases warrants further research with a larger case number. Also, no follow-up is available yet to
assess functional outcome, making these the limitations of this study.
Funding for this study: No funding was received.
Ethics committee - additional information: The requirement for informed consent was waived for this retrospective study.
Evaluation of perfusion impairment in Kawasaki disease using fully quantitative cardiovascular magnetic resonance
myocardial perfusion: correlation with left ventricular remodelling (7 min)
Zhongqin Zhou; Chengdu / China
115
Author Block: Z. Zhou, L-y. Wen, S. Azhe, L. Hu, Y-K. Guo; Chengdu/CN
Purpose: This study aims to assess perfusion impairment in Kawasaki disease (KD) using fully quantitative cardiovascular magnetic
resonance (CMR) myocardial perfusion and correlates it with left ventricular (LV) remodeling.
Methods or Background: Eight-seven children (60 males, 7.49±2.15 years) diagnosed with KD and 33 controls (22 males,
8.22±2.81 years) were enrolled to complete CMR. A dual-bolus protocol was used to acquire perfusion images. Myocardial blood flow
(MBF) was acquired through post processing and corrected for the heart rate-blood pressure product (MBF corrected [MBFcor]). The
diameter of coronary artery in patients with KD was standardized by Z score. Z score≥5 represented mid-giant coronary aneurysm.
Results or Findings: MBFcor was lower in patients with KD than that of controls (2.23±0.50 vs 2.48±0.67 mL/g/min, p=0.031),
especially in patients at acute phase (2.08±0.54 vs 2.48±0.67 mL/g/min, p=0.026). The MBFcor in patients with Z score≥5 decreased
when compared with the other patients with KD and controls (2.04±0.54 vs 2.34±0.44 vs 2.48±0.57 mL/g/min, p=0.005). MBFcor was
correlated with Z score in patients with KD (r=-0.305, p=0.004). Multivariate analysis revealed that age, acute phase and Z score was
correlated with MBFcor (β=-0.238, p=0.019; β=-0.217, p=0.031; β=-0.290, p=0.005, respectively). Radial peak strain (PS), the
absolute value of circumference PS, LV mass index and LV remodeling index was correlated with MBFcor (r=0.233, p=0.035; r=0.259,
p=0.019; r=-0.268, p=0.012; r=-0.391, p<0.001, respectively). Multivariate analysis revealed that acute phase and MBFcor was
associated with LV remodeling index (β=0.290, p=0.003; β=-0.235, p=0.025, respectively)
Conclusion: Fully quantitative CMR myocardial perfusion can evaluate myocardial perfusion impairment in children with KD. The
perfusion impairment in KD is related to acute disease and coronary artery dilation and may cause abnormalities in LV function and
LV remodeling.
Ethics committee - additional information: This study was approved by the Medical Ethics Committee of our hospital.
116
EF - Artificial intelligence (AI): what you need to know, do, and say?
Categories: Artificial Intelligence & Machine Learning, EuroSafe Imaging/Radiation Protection, General Radiology, Imaging Methods,
Physics in Medical Imaging
CME Credits: 1
Moderators:
Paddy Gilligan; Dublin / Ireland
Lucie Sukupova; Prague / Czechia

Introduction (5 min)
Peter Joseph Macmahon; Dublin / Ireland
What you need to know: setting up a syllabus for AI training (15 min)
Irene Hernandez-Giron; Dublin / Ireland
1. To learn the principles EFOMP AI syllabus.

2. To learn about the content of an AI syllabus.
3. To become familiar with how its implemented in training.
What you need to do: setting up a multidisciplinary AI evaluation service (15 min)
1. To learn about the rationale behind such a service.

2. To understand the structure and elements involved.
3. To review learning from the initial set-up in Finland.
What you need to say: ethical, patient, and legal perspective (15 min)
Mary Kirwan; Dublin / Ireland
1. To describe informed consent, legal obligations, patient autonomy and GDPR in AI technology.
2. To understand how this can be delivered in practice.
3. To look toward future trends in this area.
Panel discussion: AI: do we know, do or say enough? (8 min)
117
RC 301 - CEUS applications in abdominal and GI radiology: let the bubbles burst!
Categories: Abdominal Viscera, GI Tract, Imaging Methods

CME Credits: 1
Moderator:
Paul S. Sidhu; London / United Kingdom

CEUS in GI tract (15 min)

Polina Rudenko; Valencia / Spain
1. To define the main indications for CEUS implementation in small bowel.

2. To summarise the technique of CEUS in GI.
3. To learn about the visual and quantitative analysis of the bowel wall enhancement.
4. To exemplify (by cases from daily routine) the role of CEUS in IBD (inflammatory bowel disease) diagnostics and follow-up.
CEUS in liver (15 min)

Maija Radzina; Riga / Latvia
CEUS for interventional treatment guidance and follow-up in abdominal organs (15 min)
Dirk-André Clevert; Munich / Germany
1. To learn how to manage intervention and follow-up in abdominal organs.

2. To understand new technical developments in ultrasound intervention.
3. To appreciate the advantages of ultrasound.
4. To become familiar with the developments of interventions and follow-up in abdominal organs in the future.
Panel discussion: How can CEUS be integrated into daily practice? (10 min)
118
HW 3Sa - The imaging dilemma in acute stroke: CT vs MR

CME Credits: 1

1. To become familiar with stroke and therapeutic decisions by investigating the role of MRI and CT in guiding therapeutic decisions
related to strokes.
2. To learn about imaging techniques to differentiate stroke mimics from genuine stroke cases.
3. To develop practical skills in identifying stroke mimics by exploring conditions that resemble strokes but have
4. To become familiar with identifying stroke mimics by exploring conditions that resemble strokes but have
5. To understand the interventional-diagnostic correlation with the aim of a better prognostic outcome.

Marios-Nikos Psychogios; Basel / Switzerland
Charlotte S. Weyland; Aachen / Germany
119
OF 3R - Maximising the potential of CT technologies for enhanced patient safety
Categories: Education, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods, Radiographers
CME Credits: 1
In an era of rapidly evolving medical imaging technologies, ensuring patient safety remains a paramount concern. This session delves
into the realm of Computed Tomography (CT) and its potential to enhance patient safety through advanced techniques, optimised
radiation doses, and new skills and competencies. The aim of the session is to provide an update on the latest advancements and
practices in CT imaging so as to better ensure that patients receive the highest quality care with minimal radiation exposure. The
esteemed speakers will share their expertise and provide valuable insights for radiographers, radiologists and healthcare
professionals committed to enhancing patient safety in diagnostic radiology.
Moderator:
Andrea Bellizzi; Mosta / Malta

Andrea Bellizzi; Mosta / Malta
New skills and competencies for radiographers working with photon-counting CT (16 min)
Marcel L. Dijkshoorn; Rotterdam / Netherlands
Optimising radiation dose and image quality in cardiac CT (16 min)

Svea Deppe Moerup; Middelfart / Denmark
Opportunities for CT dose optimisation in paediatrics (16 min)

Shane J Foley; Dublin / Ireland
120
RC 307 - Prostate imaging-guided procedures
Categories: Education, Genitourinary, Imaging Methods, Interventional Oncologic Radiology, Oncologic Imaging
CME Credits: 1
Moderator:
Raphaele Marie Renard Penna; Paris / France

Raphaele Marie Renard Penna; Paris / France
Prostate "in-bore" MRI-guided biopsy (15 min)

Joan C. Vilanova; Girona / Spain
1. To define the technique and how it is performed.

2. To identify patients most suitable for this procedure.
3. To learn possible complications of the procedure.
New generation focal therapies for prostate cancer (15 min)

Jurgen Fütterer; Nijmegen / Netherlands
1. To define the technique and how it is performed.

2. To identify men suitable for this procedure.
3. To learn possible complications of the procedure.
MRI scoring for focal therapy response assessment (PI-FAB) (15 min)
1. To learn about the new scoring for response assessment to focal therapy for prostate cancer.
2. To learn how the scoring system works and when it can be applied.
3. To understand the potential clinical implications.
Panel discussion: How to avoid surgery for prostate disease? (10 min)
121
RC 308 - Imaging Menière disease
Categories: Education, Head and Neck, Imaging Methods

CME Credits: 1
Moderator:
Bert De Foer; Antwerp / Belgium

Basics of Menière's imaging: when and how to do it (15 min)

Steve Connor; London / United Kingdom
1. To present the challenges in Menière's diagnosis and to discuss current clinical indications for Menière's imaging.
2. To review anatomy and imaging appearance of endolymphatic and perilymphatic spaces.
3. To discuss available methods of Menière's imaging.
Imaging of Menière disease: how to read it (15 min)

Anja Bernaerts; Antwerp / Belgium
1. To present imaging findings of Menière's disease.

2. To define current classification systems/basic grading of endolymphatic hydrops.
3. To highlight implications for treatment and follow-up.
Imaging of secondary hydrops and non-hydropic temporal bone diseases (15 min)
Anne Renée Juliette Péporté; Frauenfeld / Switzerland
1. To explain the differences between primary and secondary hydropic ear disease.
2. To understand the specificities of the most common aetiologies of secondary hydropic ear disease.
3. To highlight other temporal bone pathologies that might be detected on hydrops imaging.
Panel discussion: Recommendations for the radiological routine (10 min)
122
CUBE 3 - Preoperative embolization

The "Special Topic" sessions address rarer, more challenging interventions or topics of importance to daily practice.
Moderator:

Preoperative embolization (28 min)

1. To understand the importance of preoperative embolization.

2. To define the indications of performing preoperative embolization.
3. To familiarize with the technique of embolization and its complications.
123
ST 5 - European Radiology - Leading clinical research in medical imaging

The new Editor-in-Chief of European Radiology, Prof. Bernd Hamm (Berlin/DE) explains his vision and aims for the ESR's flagship
journal and the ESR Journal Family: which articles will be published, challenges and opportunities in an ever-changing landscape of
radiology.
Moderator:
Interview (30 min)

Bernd Hamm; Berlin / Germany
124
RPS 415 - Examining portal venous thrombosis: diagnosis and management strategies
Categories: Abdominal Viscera, Imaging Methods, Interventional Radiology, Vascular

CME Credits: 1.5
Moderator:
Andrea Mazzaro; Treviso / Italy
Visualisation of the entire portal and hepatic venous systems in a single acquisition with ferumoxytol-enhanced MRA (7
min)
Amir Imanzadeh; Boston / United States
Author Block: A. Imanzadeh1, M. Jalili2, T. Yoshida1, C. Hassani1, P. Finn1, A. Bedayat1; 1Los Angeles, CA/US, 2Bridgeport, CT/US
Purpose: Capturing premium images, particularly of minute terminal vessels, is intricate but vital in diagnosing portal hypertension.
It aids in identifying portal and hepatic venous thromboses, porto-systemic shunts, and vascular anomalies. Given that hepatic and
portal veins have differing fill times during dynamic contrast-enhanced imaging, a multi-stage test is done. This often doesn't
perfectly assess these venous systems. Our research aimed at gauging clarity, diagnostic certainty, and artefact presence in these
systems for those having Ferumoxytol-enhanced steady-state MRA (FE-MRA).
Methods or Background: In this IRB approved HIPAA compliant study, we reviewed 20 patients (average age: 62.5 years) subjected
to breath-held 3D MRA while ferumoxytol was distributed. An anonymous radiologist, unaware of patient data, rated the images.
Criteria included overall quality, hepatic system visibility, and portal system visibility, rated on a 5-point Likert scale. Emphasis was on
secondary and tertiary vessels. A 5 signified top-notch contrast with clear vessel borders, while 1 indicated it was non-diagnostic.
Confidence in determining vessel state was gauged on a 3-point scale (3 signifying high confidence, even with smaller vessels).
Artifact grading, arising from movements or devices, utilized a 5-point Likert scale (1 being none, 5 being acute).
Results or Findings: Image grades were stellar (4.5 ± 0.7). Scores for hepatic and portal systems stood at 4.6 ± 0.5 & 4.6 ± 1.3,
respectively. Confidence in vessel diagnosis was robust (2.9 ± 0.3), covering tinier vessels. Artefact due to external factors scored low
(1.5 ± 0.6).
Conclusion: Both venous systems can be studied in one breath-held 3D session with ferumoxytol. FE-MRA consistently depicted tiny
vessels. Upcoming studies should juxtapose with gadolinium-based agents. Given its properties, ferumoxytol might become the go-to
for visceral venous imaging.
Limitations: Limited number of patients and retrospective nature of study limited the scope of this study.
Ethics committee - additional information: Study was approved by our ethics committee.
Vascular complications of liver transplantation in children, early diagnosis and management (7 min)
Andrés Felipe Mejía León; Bogotá / Colombia
125
Author Block: A. F. Mejía León, L. M. Acosta, L. Acosta, J. E. Chaustre Soledad, S. Trujillo, J. M. Perez, G. Caviedes; Bogotá/CO
Purpose: The aims of this study were to: (1)Describe the most common early vascular complications of liver transplant in the
paediatric population. (2) Show the main imaging findings through different case examples of our institution. (3) Compare the
radiological findings found in our cohort of patients with those described in the literature.
Methods or Background: The number of paediatric patients undergoing transplantation has grown exponentially in recent years,
enhancing the importance of early detection of the different complications related to early and late postoperative stage in the clinical
scenario. The radiologist and radiologist-in-training must be aware of these complications and early imaging
manifestations.Understanding the findings in different imaging modalities, is essential to a proper approach, with Doppler ultrasound
being one of the most important modalities for early detection of complications. Through a multimodality approach we intend to show
the importance of each of these modalities for early diagnosis, and likewise, to describe the type of management according to the
postoperative time. The management of these complications is not limited to surgical management. The role of interventional
radiology is increasingly recognised, depending on the type of complication and the postoperative time being a fundamental pillar in
the management of these patients.
Results or Findings: Regarding the main vascular complications in the postoperative period of liver transplants in paediatrics, we
found greater prevalence of arterial thrombosis, follow by stenosis of the hepatic artery and hepatic veins. Some of these cases were
managed by interventional radiology, obtaining favorable results.
Conclusion: Findings do not differ from those described in literature presenting a similar frequency of appearance. When
complications are detected soon in the early postoperative period, this group of patients benefit from earlier interventions, which
translates into better clinical outcomes.
Limitations: Only Latin American patients were included limiting the scope of this study.
Ethics committee - additional information: Low risk research hence no ethical approval was required.
Reliability and accuracy of tomographic 3D ultrasound for grading vessel stenosis: a phantom study (7 min)
Adel Ahmad Alzahrani; Jeddah / Saudi Arabia
Author Block: A. A. Alzahrani1, M. Aslam2, S. R. Sultan3; 1Makkah/SA, 2London/UK, 3Jeddah/SA

Purpose: The aim of this phantom study was to assess the accuracy of 3-D tomographic ultrasound (t3DUS) for grading stenosis,
using the manufacturer's measurements as the gold standard. The percentage of maximum stenosis was obtained using 2-D
ultrasound (2DUS) and t3DUS imaging techniques on a peripheral vascular phantom, including channels with 50%, 75% and 90%
stenosis.
Methods or Background: A high-resolution premium Philips Elite ultrasound imaging system (Philips Healthcare, Bothell, WA, USA)
was used to obtain maximum percentage of stenosis from a peripheral vascular phantom using 2DUS and t3DUS (PIUR Imaging
GmbH, Vienna, Austria) imaging techniques. Inter-operator reproducibility and accuracy of DR and AR 2DUS and t3DUS in measuring
maximum stenosis percentage were assessed.
Results or Findings: The inter-operator reproducibility of DR and AR 2DUS and t3DUS maximum stenosis measurements between
operator A and B (n = 90 for each operator per technique) was excellent (DR 2DUS, ICC = 0.96, 95% confidence interval [CI]:
0.900.98, p < 0.001
Conclusion: Free-hand t3DUS is a reproducible and accurate imaging method for grading stenosis. The inter-operator reproducibility
of t3DUS for grading stenosis was excellent with a low coefficient of variation. The mean difference in stenosis measurements from
manufacturer reference values for all channels was lower in t3DUS than in 2DUS. There was significant under- and overestimation of
2-D DR and AR, respectively, compared with Tomographic 3-DUS for grading vessel stenosis.
Funding for this study: This study was funded by the King Abdullah Medical City KAMC.
Reducing the radiation doses and contrast volume in CT portal venography for patients with sinusoidal obstruction
syndrome on a second-generation dual-layer spectral CT (7 min)
Yiran Wang; Zhengzhou / China
126
Author Block: Y. Wang; Zhengzhou/CN
Purpose: The aim of this study was to investigate the feasibility of 60keV virtual monoenergetic images (VMIs) of spectral CT
imaging, combined with low contrast dose for patients with sinusoidal obstruction syndrome (SOS) by comparing with polyenergetic
CT images.
Methods or Background: In the prospective study, 28 patients who underwent dual-layer spectral CT scanning were divided into
two group, group A (control group,120kVp) and group B (experimental group, 100 kVp). 95 mL of 350 mgI/mL iodixanol was
administered to the control group, and personalised injection protocol (25 mgI/kg/s with a duration of 17 seconds) was used for group
B, respectively. CT numbers of the portal veins (PVs), liver parenchyma, and subcutaneous fat tissue in the abdomen were measured
by the region of interest (ROI). The standard deviation of the images was interpreted as the objective image noise (IN). The diagnostic
acceptability (DA) and sharpness of PV margins were obtained using a 5-point score.
Results or Findings: Hepatic heterogeneity, puddle-like or micronodular appearance, peripheral distribution of heterogeneity,
clover-like sign, splenomegaly, as well as the subjective impression of the observer were significantly associated with SOS diagnosis.
The observers' confidence in the diagnosis of SOS increased significantly in group B. The CT value of liver in group B was higher than
group A, conversely, SNR and CNR were lower (P＜0.05). Compared with group A, the effective radiation dose of group B decreased by
26.6%. The average iodine loads were 33.3 g and 28.3±3.6 g, respectively.
Conclusion: Compared to conventional polychromatic CT images, 60 keV VMIs of 100 kVp improved diagnostic performance of
sinusoidal obstruction syndrome in portal venography imaging with reduced contrast media and lower radiation dose.
Limitations: Our sample is small in spite of a long period of study limiting the scope of this study.
Ethics committee - additional information: This study was approved by the First affiliated hospital of Zhengzhou university Ethics
committee.
Development and validation of a new score for selecting the best vascular access for liver cancer intra-arterial
procedures (trans-femoral or trans-radial access score) (7 min)
Alessandro Maresca; Rome / Italy
Author Block: A. Maresca, L. Tenore, A. Posa, A. Contegiacomo, M. Lippi, L. Natale, R. Iezzi; Rome/IT
Purpose: Intra-arterial procedures play an important role in the treatment of liver neoplastic lesions. Trans-radial artery access has
been considered for IR procedures; the choice of a femoral or radial approach is based on the operator’s experience or preference.
This study aims to develop a CT-based scoring system for selecting the best candidates for trans-radial approach.
Methods or Background: All the patients who underwent TACE using a trans-radial approach were included. Patients’ clinical data,
pre-procedural CT-image and lesion location were registered and a CT scoring system was developed. Procedural complexity, based
on angiographic and procedural variables, was evaluated. The relationship between procedural difficulty and the categorical variables
were statistically tested.
Results or Findings: One hundred and eight-two trans-radial TACE procedures were analysed; only 93 patients were retrospectively
included in our study due to the adequacy of the data. The parameters which resulted significant (p<0.0001) in determining the
procedural difficulty were represented by the aortic arch diameter, the suprarenal aortic diameter, the celiac trunk takeoff angle and
the anatomical variants of celiac trunk anatomy. Multivariate regression analysis identified four variables like predictors of procedural
complexity: radial caliber < 2.6 mm, left subclavian artery angle < 65 degrees, suprarenal aorta diameter > 33 mm, celiac trunk
takeoff angle > 81 degrees. Using these four variables, a 4-point risk score was developed (1-2: easy – 3-4: complex).
Conclusion: TAS (transradial access) score seems to be useful to predict complexity of endovascular interventions through a
transradial approach.
Limitations: The most important limitation of the study is that it is focused on the learning curve of a single operator on a single
procedure (lobar chemoembolisation). Data are not applicable to all operators and/or other procedures and do not take into account
individual operator variabilities and differences in procedure complexity.
Funding for this study: No funding was obtained for this study
Ethics committee - additional information: The study was led under the approval of the local ethics committee and the
institutional review board (IRB).
Direct Percutaneous Thrombolysis (DPT): an effective method of salvaging thrombosed native arteriovenous fistula (7
min)
Vignesh Selvamurugan; Palakkad / India
127
Author Block: V. Selvamurugan1, R. Prasad2, R. R. Yadav3, P. Hasani4, A. Israr3, H. Lal3, S. Sharma3; 1Madurai, Ut/IN, 2Delhi/IN,
3
Lucknow/IN, 4Vadodara/IN
Purpose: Thrombosed arteriovenous fistulas (AVFs) are either treated by thrombectomy or pharmaco-mechanical thrombolysis with
or without percutaneous balloon angioplasty. In this study, we have described an effective and economical technique of salvaging
these fistulas using a 20–22-gauge spinal needle and urokinase - direct percutaneous thrombolysis (DPT).
Methods or Background: This prospective study comprised of 148 patients out of which 120 patients presented with AVF
thrombosis and were divided into two groups; those with no obvious stenosis on ultrasound (n=38) and second with venous stenosis
(n=82). Remaining 28 patients developed thrombosis post angioplasty for venous stenosis. Percutaneous injection of urokinase into
the thrombus was done under ultrasound guidance, followed by balloon angioplasty if there was associated stenosis.
Results or Findings: In 38 patients who didn’t have any stenosis, 32 AVFs were successfully thrombolysed by DPT, with technical
success of 84.2%. Remaining six patients required angioplasty because of chronic nature of clot. In 82 patients who had venous
stenosis, 80 cases were treated successfully by DPT followed by angioplasty with technical success of 97.5%. In third group (n=28),
who developed thrombosis post angioplasty, 100% success rate was noted. The mean length of thrombus was 31.4 ± 4.6 mm and
mean diameter of thrombosed vein was 10.5 ± 1.2 mm. There were no major complications encountered during the procedure.
Conclusion: Ultrasound guided DPT with urokinase is a safe and economical option for salvaging thrombosed AVF without vascular
stenosis that does not need angioplasty.
Limitations: The main limitation was inherent selection bias, as we did not enroll patients with very long thrombosed segment and
grossly dilated thrombosed venous segment. There was also lack of comparison between the various groups enrolled.
Ethics committee - additional information: All procedures performed in the studies involving human participants were in
accordance with the ethical standards of the institutional research committee. Retrospective analysis so ethical approval not taken.
Relevance of preoperative and early postoperative USG vascular assessment in predicting haemodialysis AV fistula
failure in chronic kidney disease patients (7 min)
Shikhar Gupta; Delhi / India
Author Block: S. Gupta, M. Ahmad; Aligarh/IN

Purpose: The increasing prevalence of chronic kidney disease (CKD), coupled with advancements in the diagnosis and treatment of
renal diseases and improvements in life expectancy, has led to a greater number of patients requiring hemodialysis. The preferred
method of vascular access for haemodialysis is AV fistula formation; however, it is associated with a high rate of failure.
Methods or Background: In this prospective study, 40 CKD patients planned for initiation of maintainence hemodialysis were
selected and preoperative ultrasound vessel mapping and early postoperative ultrasound assessment on day seven were utilised to
establish criteria for predicting early fistula failure. Preoperative ultrasound mapping was employed to assess various factors such as
cephalic vein diameter, compressibility, and colour flow, radial and brachial artery diameter, peak systolic velocity, and intimal wall
calcification. Postoperatively, ultrasound examinations were conducted on day seven and at six weeks to evaluate fistula blood
volume and detect any complications.
Results or Findings: A significant association between fistula failure and factors such as cephalic vein diameter, brachial artery
diameter, intimal vessel wall calcification, and comorbid conditions like diabetes mellitus was observed. Furthermore, the blood flow
at day seven was notably lower in the failure group compared to those with a functioning fistula and any fistula with blood flow <154
ml/min on day 7 may be predictive of early fistula failure.
Conclusion: Preoperative vessel mapping and early postoperative ultrasonography is indispensable for patients who require AV
fistula formation for hemodialysis and provide valuable information for selecting suitable vessels in successful fistula creation and
enable early intervention to salvage a failing fistula after the surgery. By utilising these, healthcare professionals can make informed
decisions and take necessary steps to optimise the outcomes of AV fistula formation in patients undergoing haemodialysis.
Limitations: No limitations were identified in this study.
Funding for this study: No fundings were received for this study.
Ethics committee - additional information: Institutional Ethics Committee approved this study.
The vanishing veins (7 min)

Mahmoud Ali Sarhan; Eastbourne / United Kingdom
128
Author Block: M. A. Sarhan; Eastbourne/UK
Purpose: The aim of this study was to help radiologists to understand the practical way for lower limb venous examination and to
detect the blood clot within the veins even in the early stages of deep venous thrombosis.
Methods or Background: Important clinical information coming from extensive and long experience in examining veins to provide
young doctors with the correct path to examine such cases and achieve reliable and sound results.
Results or Findings: Vanishing vein is a magic technique for the early detection of blood clot. Thrombosis is one of the emergency
conditions that necessitates early and accurate diagnosis for the patients. Simple techniques could add a lot to the patients.
Conclusion: The vanishing veins are the happy normal veins.
Limitations: These are the limitations of this study: (1)Obese patients (2) Marked lower limb oedema (3) Dermal and sub-dermal
infection (4) Machine capabilities
Funding for this study: This study was self-funded.
Imaging of abdominal arteries: intra-individual comparison of Relaxation-Enhanced Angiography without contrast and
triggering (REACT) with 4D contrast-enhanced MR angiography at 3T (7 min)
Jan Paul Janssen; Cologne / Germany
1 1 1 2 1 3 1 1 1
Author Block: J. P. Janssen , L. Goertz , K. Kaya , J-P. Grunz , T. Persigehl , K. Weiss , L. Pennig , C. H. Gietzen ; Cologne/DE,
2 3
Würzburg/DE, Hamburg/DE
Purpose: The main aim of this study was to compare a Novel Relaxation-Enhanced Angiography without contrast and Triggering
(REACT) sequence with 4D contrast-enhanced magnetic resonance angiography (4D CE-MRA) for imaging of the abdominal arteries.
Methods or Background: Thirty patients (35.7±16.8 years; 10 males) who received abdominal vessel imaging using a standardised
protocol at 3T were included in this retrospective, single centre study. The protocol comprised both 4D CE-MRA and flow-independent
REACT (Compressed SENSE factor 10, reconstructed voxel size 0.8x0.8x0.9 mm3) sequences. Two radiologists independently
evaluated abdominal arteries for the presence of stenosis, variants, and other vascular findings (e.g., dissection). Subjective image
quality of arteries was assessed using a 4-point Likert scale (1=non-diagnostic, 4=excellent). Vessels were classified based on size:
(1) aorta (supra- and infrarenal segments), (2) large (celiac trunk, superior mesenteric artery, renal arteries), (3) medium (splenic
artery, common and proper hepatic artery) and (4) small (gastric arteries, hepatic arteries, inferior mesenteric artery) arteries.
Results or Findings: REACT yielded a median acquisition time of 304 s. Considering 4D CE-MRA as the standard of reference, REACT
achieved a sensitivity of 87.5% and a specificity of 100% for relevant (≥50%) stenosis while detecting 89.3% of vascular variants and
100% of other findings. Vessel quality was comparable between both sequences at the aorta (4D CE-MRA: 3.94±0.24, REACT:
3.88±0.44; p<0.044) as well as for medium (4D CE-MRA: 2.93±0.96, REACT: 2.77±0.90; p=0.028) and small arteries (4D CE-MRA:
2.15±0.85, REACT: 2.04±0.86; p=0.032). In contrast, 4D CE-MRA yielded slightly higher scores for large arteries (4D CE-MRA:
3.61±0.61, REACT: 3.35±0.72; p<0.001).
Conclusion: REACT provides a good diagnostic performance for the detection of relevant stenosis, variants, and other findings of
abdominal arteries while yielding to 4D CE-MRA comparable image quality, underlining its use for non-contrast evaluation of the
abdominal vasculature.
Limitations: No limitations were identified in this study.
Ethics committee - additional information: The study is retrospective so no ethical approval was sought.
Vascular access devices registry as a working tool to improve infection control and thromboembolism management in
cancer patients (7 min)
Mikhail Cherkashin; Saint Petersburg / Russia
129
Author Block: M. Cherkashin, S. Alexandrov, F. Valieva, A. Nikolaev, T. Bolshakova, E. Ilyukhin, N. Berezina; Saint-Petersburg/RU
Purpose: The aim of this study was to discuss permanent vascular access devices registry development and implementation.
Methods or Background: Vascular access devices (port-systems, peripherally implanted central catheters etc) characterised by
thrombotic and infectious complications. After device implantation oncology patient can receive different types of anti-tumour therapy
in different hospitals and sometimes it's very hard to obtain clear information - which device was implanted, were there any
complications etc. The applicable way to exchange clinically important information between hospitals is creation of electronic
registers with access for healthcare providers, involved in patient management. We decided to use online platform "Russian Registry
of Treatment of Venous Thromboembolism" (NCT03881345) with some customisation, related to central venous devices,
concominante blood stream infections, device malfunctions etc.
Results or Findings: Registry was started with 3 hospitals operating with paediatric oncology in St Petersburg. System development
includes relevant steps: level of access (in each hospital should be dedicated owner with total access and each healthcare provider,
involved in device management should have rights to upload information); data fields customisation (type of device, place of
insertion, technical features during implantation, complications, malfunctions, thrombosis/occlusion, infection, therapy etc). In first
month since process was started, 24 cases were uploaded (14 port-a-caths, 8 PICCs, 2 midlines) based on retrospective data we
estimate 400-600 cases per year.
Conclusion: Registry is the extremely helpful tool for patient tracking during all stages of cancer treatment. Based on our results,
such types of initiatives should be spreaded on city level with all paediatric oncology departments involvement.
Limitations: No information was provided by the submitter.
Funding for this study: This study was funded by the hospital.
Ethics committee - additional information: This study was funded approved by Local ethics committee.
Diagnostic performance of angio-MRI in the evaluation of large vessel vasculitis: preliminary results (7 min)
Ejona Duka; Varese / Italy
Author Block: E. Duka1, L. Di Meglio1, E. Tombetti1, M. Papa2, G. M. Roda'1, G. Carrafiello1; 1Milan/IT, 2Gragnano Trebbiense/IT
Purpose: The study aims to evaluate diagnostic performance and reproducibility of angio-MRI in the evaluation of patients affected
by large vessel vasculitis (Takayasu arteritis and Giant cell arteritis).
Methods or Background: Retrospective analysis of 22 patients affected by large vessel vasculitis that performed angio-MRI in our
centre being referred from the Immunology and Rheumatology colleagues. Two readers with five- and two-years’ experience in angio-
MRI retrospectively analysed the images. The evaluation regarded the involvement of aorta, supra-aortic trunks, splanchnic vessels
and iliofemoral vessels. The inter-observer agreement and reliability was established with Cohen’s kappa test. The diagnostic findings
were stated in percentage. The diagnostic performance was expressed in terms of percentage with 95% of interval confidence in
terms of sensitivity and specificity.
Results or Findings: Aortic involvement (wall thickening with enhancement) was the commonest finding in 63% of patients. The
diagnostic performance was 100% specificity, 80% sensitivity and the inter-reader agreement showed moderate accuracy (κ 0.88).
Conclusion: Angio-MRI represents a valid and reliable tool in the evaluation of patients affected by large cell arteritis, in the
diagnosis, management and during follow-up.
Limitations: There is a small number of patients presenting with the pathology since it is a rare disease. More patients could be
enrolled in order to develop a more accurate description of MRI signs of this disease.
Ethics committee - additional information: No ethics committee approval necessary.
Usefulness of low-energy virtual monochromatic CT imaging with deep-learning image reconstruction in the
delineation of endoleaks after endovascular stent‑graft placement (7 min)
Takatoshi Higashigawa; Tsu / Japan
130
Author Block: T. Higashigawa1, Y. Ichikawa1, K. Nakajima2, T. Kobayashi1, K. Domae1, A. Yamazaki1, N. Kato1, H. Sakuma1; 1Tsu/JP,
2
Ise/JP
Purpose: The aim of this study was to investigate the usefulness of low-energy virtual monochromatic imaging combined with deep
learning image reconstruction (DLIR) in improving the delineation of endoleak after endovascular stent‑graft placement in contrast-
enhanced dual-energy CT (DECT).
Methods or Background: Sixty-one consecutive patients (median age, 79 years; 46 men) after endovascular stent-graft placement
who underwent contrast-enhanced DECT between December 2021 and February 2023 were studied. Virtual monochromatic 40- and
70-keV images were reconstructed using DLIR (TrueFidelity-H) and conventional hybrid iterative reconstruction (IR) (ASIR-V50%).
Contrast-to-noise ratio (CNR) of endoleak on the venous phase CT were calculated. Four different reconstructed image series (hybrid
IR and DLIR at two energy levels, 40- and 70-keV) were displayed side-by-side and visually evaluated for endoleak conspicuity on a 5-
point comparative scale from 0 (best) to -4 (significantly inferior).
Results or Findings: A total of 30 out of 61 patients had endoleak (type II, 27; type III, 2; type I, 1). CNR of the endoleak were
significantly higher in DLIR than in hybrid IR on both 40- and 70-keV images (40-keV, 14.5 ± 7.3 vs 8.6 ± 4.2, p < 0.001; 70-keV, 8.7
± 4.5 vs 5.5 ± 2.6, p < 0.001). The endoleak conspicuity score for 40-keV DLIR images (Reviewer 1, -0.2 ± 0.4; Reviewer 2, 0.0 ± 0.0)
was significantly higher than 40-keV hybrid IR (Reviewer 1, -0.5 ± 0.5; Reviewer 2, -1.0 ± 0.0; p < 0.05), 70-keV DLIR (Reviewer 1,
-1.8 ± 0.4; Reviewer 2, -2.0 ± 0.0; p < 0.001) and 70-keV hybrid IR images (Reviewer 1, -1.8 ± 0.4; Reviewer 2, -2.4 ± 0.5; p <
0.001), respectively.
Conclusion: The utilisation of low-energy virtual monochromatic imaging combined with DLIR method improves the delineation of
endoleak after endovascular stent-graft placement.
Limitations: Retrospective study limits the scope of this study.
Funding for this study: No funding was was received for this study.
Ethics committee - additional information: The current study was conducted with the approval of our institutional review board,
and written informed consent was waived because existing clinical CT data were used for this study. The opportunity to opt out of
participating in this study was provided by a notice posted on the hospital’s website. No patient indicated an intention to be excluded
from this study.
131
S 4 - Students Session 1
Categories: Artificial Intelligence & Machine Learning, Chest, Genitourinary, Head and Neck, Students, Vascular
CME Credits: 1.5
Moderator:
Ioana Andreea Gheonea; Craiova / Romania
Artificial intelligence in breast cancer screening: a retrospective analysis of breast cancer detection, breast cancer
characteristics and workload reduction (8 min)
Viktor Lu; Malmö / Sweden
Author Block: V. Lu, V. Dahlblom, M. Dustler, S. Zackrisson, K. Johnson; Malmö/SE

Purpose: This study aimed to evaluate the diagnostic accuracy of an AI system in digital mammography (DM), analyse the detected
breast cancer characteristics, and explore workload reduction.
Methods or Background: Screening with DM faces challenges due to labour intensity. AI has been proposed as a solution. Double-
read DM images from women who underwent screening between 2010 and 2015 in Malmö, Sweden, were analysed retrospectively
and assigned an AI score between 1 and 10 by a commercial AI system (Transpara, ScreenPoint Medical). Scenario 1 assessed the
diagnostic accuracy of the AI system, recalling examinations with an AI score of 10. In scenario 2, women with an AI score of 10 were
recalled alongside women recalled by radiologists. Proportions of various breast cancer characteristics were compared. Workload
reduction was explored by excluding AI scores 1 to 9 from the radiologists' reading stream. Non-inferiority was concluded if the lower
limit of the 95% CI of the difference was >5%.
Results or Findings: One screening occasion per 26758 women was included, of which 753 were excluded due to failed AI analysis.
Radiologists achieved a sensitivity of 70.3% (95%CI: 64.6; 76.0) and a specificity of 97.7% (95%CI: 97.6; 97.9). Scenario 1 achieved a
sensitivity of 70.7% (95%CI: 65.0; 76.4) and a specificity of 91.6% (95% CI 91.3; 91.9). Scenario 2 achieved a sensitivity of 79.3%
(95%CI: 74.2; 84.3) and a specificity of 90.3% (95%CI: 90.0; 90.7). The AI system and radiologists detected similar breast cancer
characteristics, although the AI system missed four tubular carcinomas. The AI system could reduce the workload by 50.3% while
maintaining a non-inferior sensitivity of 69.1%.
Conclusion: The AI system achieved a non-inferior sensitivity but lower specificity in both scenarios, detected various breast cancer
characteristics, and significantly reduced workload. Specificity could improve with consensus meetings at the expense of a minor
increase in workload.
Limitations: It is a retrospective study design, and prior mammograms were not analysed.
Funding for this study: Funding was received from the Swedish governmental funding of clinical research (ALF).
Ethics committee - additional information: This study and the data collection have been approved by the Regional Ethical
Review Board in Lund University (Event No 2009/770 and 2017/326), which includes opt-out consent for the population.
Impact of an artificial intelligence (AI) software on the diagnostic performance of a junior radiologist in detecting
prostate cancer on MRI (8 min)
Ludwig Meinsohn; Rennes / France
132
Author Block: L. Meinsohn1, L. Alberge2, C. Adam2, G. Herpe2, G. D'Assignies2, L. Beuzit1; 1Rennes/FR, 2Paris/FR
Purpose: The purpose of this study was to evaluate the impact of an artificial intelligence (AI) software on the diagnostic
performance of a junior reader (resident) in detecting clinically significant prostate cancer (csPCa) on MRI.
Methods or Background: A dataset comprising 204 mpMRI cases from the PROSTATEx Challenge was employed. Each targeted
lesion had a known histology. A resident read each case without AI and then after a 4-week interval with AI assistance. In contrast, an
experienced radiologist reviewed the cases without AI. Their tasks encompassed lesion detection and classification per PI-RADS v2.1
standards. The readers’ performances and the standalone AI tool were evaluated using sensitivity/specificity/accuracy metrics
comparing csPCa to PI-RADS scores. PI-RADS ≥ 3 were considered as MRI positive. Differences were statistically compared using
McNemmar tests. Interobserver variability was reported using Cohen’s κ on the PI-RADS scores. The reading times for all cases were
also assessed and compared without and with AI.
Results or Findings: The accuracy of the junior radiologist in detecting csPCa increased from 0.63 to 0.75 (p= 3.8 x10-5) with the
support of AI. Sensibility and specificity also increased, respectively, from 0.84 to 0.91 (p=0.12) and from 0.53 to 0.67 (p=3x10-4).
Accuracy, sensibility and specificity of the experienced radiologist were 0.78, 0.96 and 0.69, respectively, and no statistically
significant difference was observed with the AI-assisted junior (p=0.5, p=0.25, p=0.86). Cohen κ between the junior and the
experienced radiologist increased from 0.44 to 0.54 with the support of AI. The standalone AI accuracy, sensibility, and specificity
were 0.77, 0.96, and 0.66, respectively. The annotation time of the junior radiologist was reduced by 16% using the AI tool.
Conclusion: Leveraging AI notably enhanced the junior radiologist's diagnostic precision, sensitivity, and specificity in csPCa
detection, mirroring the proficiency of an experienced colleague and curbing interreader discrepancies.
Funding for this study: Funding was provided by Eurostars E114567 - ProstAID.
Ethics committee - additional information: The study was approved by the CHU Rennes (France) ethics committee N°20.06.
Comparison of two software versions of a novel deep-learning algorithm for haemorrhage detection in non-contrast
cranial CT in a real-world patient collective (8 min)
Franziska Katharina Tombach; Würzburg / Germany
Author Block: F. K. Tombach, T. A. Bley, B. M. W. Petritsch, J. F. Heidenreich, P. Gruschwitz; Würzburg/DE

Purpose: The purpose of this study was to evaluate the performance of an improved deep-learning algorithm (version 2, 2022) for
intracranial haemorrhage (ICH) detection in non-contrast cranial computed tomography (cCT) scans and the comparison of its
performance to the previous version (version 1, 2020).
Methods or Background: The deep-learning pipeline based on a three-dimensional neural network was used to automatically
process cCTs. In software version 2, additional training cycles were used in an attempt to better detect subarachnoid haemorrhages
in particular. A sum of n=1700 cCT, created in the period from April 2020 to April 2022, was retrospectively processed by the
algorithm (version 2, 2022) and compared to the written report ("ground truth"). Discrepant results were again supervised in terms of
a consensus vote by a resident with six years of experience. 519 CT scans were analysed and compared using both deep-learning
software versions regarding diagnostic accuracy parameters.
Results or Findings: In the clinical collective (n=1700) with a prevalence of 12.6%, the software version 2 detected ICH with a
sensitivity of 94.4%, a specificity of 96.1%, an overall accuracy of 95.9% and a negative predictive value of 99.2%. In the second
dataset (n=519), version 2 detected ICH with a sensitivity of 87.3%, an overall accuracy of 94.2% and a negative predictive value of
98.2%. Compared to version 1 (sensitivity=84.1%, overall accuracy=96.1%, negative predictive value=97.8%), sensitivity and
negative predictive value in version 2 were increased, resulting in two fewer false negative findings.
Conclusion: The improvements in the deep-learning software version 2 lead to a significant increase in sensitivity and negative
predictive value and, thus, fewer false negative findings when used as a triage tool.
Limitations: Limitations included the retrospective study design and the single-centre, single-vendor approach. We clinically
indicated cCT without preselection. A supervised (resident with six years experience) radiology report was used as ground truth.
Funding for this study: Philipp Gruschwitz [Grant No Z-02CSP/18] was funded by the Interdisciplinary Center of Clinical Research
Würzburg, Germany. The Department of Diagnostic and Interventional Radiology received a Siemens research grant. The other
authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject
matter of the article.
Ethics committee - additional information: The study received permission from the local institutional review board (Ethic
Committee of the University of Würzburg; protocol number: 20230919 02).
A retrospective multicentre research: development and validation of deep learning classifiers for detecting inner ear
malformation on temporal bone CT (8 min)
Abdulrahman Alkojak Almasri; Pécs / Hungary
133
Author Block: A. A. Almasri1, A. Dhanasingh2, S. Sugarova3, F. Almuhawas4, L. Hofmeyr5, F. Wagner6, E. Wappl-Kornherr2, E. Kedves7,
A. Alsanosi4, K. Sriperumbudur2, A. Kedves2; 1Pécs/HU, 2Innsbruck/AT, 3St. Petersburg/RU, 4Riyadh/SA, 5Stellenbosch/ZA, 6Bern/CH,
7
Sopron/HU
Purpose: The ability to identify inner ear malformations (IEM) has been demonstrated by deep learning (DL) and artificial intelligence
(AI). Based on patients' computed tomography (CT), we created an automated system to identify a specific IEM.
Methods or Background: While developing the deep learning model for inner ear CTs used in this retrospective, multicentre study,
we included 2053 patients who had been imaged between 2016 and 2021. Three nations - Saudi
Arabia, South Africa, and Russia - provided temporal CT datasets. Deep convolutional neural networks were used to create supervised
learning models, and all of the data were categorised as incomplete partition
type III or other. 25 professional experts with or without training from Austria, the United Kingdom, South Africa, and Egypt evaluated
24 patients for interobserver validity by covering the variability of observers.
Results or Findings: The specificity and sensitivity of supervised learning models were 80.1%, 88.4%, 80.6%, and 88.1%,
respectively. The performance of the two-stage DL algorithm was better than the one-stage algorithm (AUC
0.86, 95% CI 0.82-0.90; AUC 0.80, 95% CI 0.74-0.86). Interobserver analysis using Kruskal Wallis ANOVA and one sample Wilcoxon
test revealed that the profession (including AI) had an impact on correctly identifying present or absent malformations but not
training (p=0.0674). The analyses even showed that the correct assignment by AI was superior to professionals (p=0.0403).
Conclusion: We outline the development and verification of a potential fully automated workflow for IEM detection. The decision-
maker must supervise the tool, even though it may have good diagnostic accuracy when risk stratification is being done.
Limitations: The limitation of this study is the possible imbalance, which may occur and could cause overfitting.
Ethics committee - additional information: This study was approved by the independent ethics committee of 3 Hospitals IRB Nos.
22/0084/IRB, 23_001/IRB, and S_23_001/IRB, respectively.
Emergency stroke imaging: current challenges and potential solutions offered by artificial intelligence (8 min)
Ioana Erica Stanescu; Ploiesti / Romania
Author Block: L. Plamadeala, I. E. Stanescu; Iasi/RO

Purpose: The study aims to provide a comprehensive overview of recent advancements in artificial intelligence (AI), machine
learning (ML), and advanced imaging techniques in the context of stroke care. It synthesizes findings from the most recent 55 studies,
sourced from PubMed and conducted between 2016 and 2023, to elucidate the transformative potential of these technologies across
various facets of stroke management.
Methods or Background: A systematic review of these studies explores AI and ML applications in stroke care, including diagnostic
accuracy, treatment optimisation, imaging enhancement, prognosis prediction and lesion segmentation. Diverse methodologies, such
as convolutional neural networks, support vector machines, deep learning models and motion correction algorithms, are employed.
Data from these studies are analysed to assess the impact and effectiveness of these technologies in stroke management.
Results or Findings: The findings collectively reveal the profound impact of AI and ML technologies on stroke care. They enable
rapid and precise diagnosis, efficient treatment selection, enhanced imaging interpretation, accurate prognosis prediction, and
sensitive lesion segmentation. Convolutional neural networks and support vector machines exhibit remarkable efficiency in stroke
subtype identification and large vessel occlusion detection. Furthermore, motion correction algorithms improve image quality and
lesion detectability in cerebral CT, while deep learning models predict stroke using raw EEG data with exceptional accuracy.
Automation platforms for intracranial large vessel occlusion detection expedite diagnostic work-ups, and multimodal deep learning
frameworks like DeepStroke outperform traditional triage methods.
Conclusion: The fusion of AI, ML, and advanced imaging transforms stroke care and enhances diagnosis, treatment, imaging, and
prognosis. To fully benefit, we must tackle research gaps in treatment studies and address data privacy and integration challenges.
Limitations: When implementing these innovative stroke care approaches, we must address some key limitations, including a focus
on diagnosis, modality-specific strategies, and data-related challenges.
Ethics committee - additional information: This study is educational.
CBCT imaging of intraosseous jaw bone lesions in children and adolescents (8 min)
Sara Alexia Roman; Cluj Napoca / Romania
134
Author Block: S. A. Roman, E. Crasnean, A. Ban, M. Hedesiu, R. A. Roman; Cluj Napoca/RO
Purpose: To present in a Cone Beam Computer Tomography (CBCT) pictorial the imaging features of the jaw bone space-occupying
lesions in children and adolescents, both benign and malignant ones.
Methods or Background: Intraosseous lesions of the jaw in children are not very common, mosly being asymptomatic. Several of
the benign ones may present locally aggressive features and need proper recognition and management. Lesions are found often
incidentally on panoramic radiography or CBCT, indicated by dentists or orthodontists for abnormalities in teeth eruption.
We present retrospective CBCT examinations of under 18 years old patients treated for jaw bone masses in the last 5 years in the
Maxillofacial Surgery Clinic, ages ranging from 5 to 18. By using individualized CBCT reconstructions, imaging characteristics were
analyzed and correlated with histopathology. Several features were assessed: density, internal structure, contour, locularity, location,
the relationship with nearby teeth and structures, effect on corticals, e.g.
Results or Findings: Differences and similarities between imaging characteristics are presented, structured for odontogenic and
non-odontogenic, benign and malignant. Images for different cystic lesions, being the majority of the cases treated, are presented,
from cysts like naso-palatal, inflammatory radicular, dentigerous, the typical and atypical odontogenic keratocyst, to central giant cell
granuloma, cherubism or ameloblastoma, followed by the denser lesions, such as cementoblastoma, cementifying or ossifying
fibromas and odontomas. Malignancy was present in the group, mostly represented by osteosarcoma.
Conclusion: Most of the jawbone lesions in children are benign entities, with typical CBCT features, the mandible being the most
affected site. Atypical presentation may pose problems in reaching a diagnosis, and needs attention, since malignant lesions are not
uncommon. The option of individualized sections in CBCT helps narrowing the differential diagnosis, allowing a proper morphological
lesion evaluation.
CT-angiography (CTA) of deep inferior epigastric artery (DIEA) and its perforators in preoperative mapping for
harvesting of deep inferior epigastric perforator (DIEP) flap (8 min)
Evgeniia Zhigailova; Moscow / Russia
Author Block: Z. Evgeniia, E. Mershina, V. Sinitsyn; Moscow/RU

Purpose: The aim of the study was to define the capabilities of CTА in the preoperative mapping of DIEA and its perforators for
harvesting DIEP flap.
Methods or Background: DIEP flaps have become the method of choice for autologous breast reconstruction. The study of variable
anatomy and preoperative mapping of DIEA and its perforators using CTA are vital for optimal flap harvesting. We retrospectively
analysed CTA of the abdominal aorta and its branches of 20 patients (m/f - 7/13, mean age 53±12 yrs). CTA was performed with dual-
source CT. We used the Mann-Whitney test for continuous outcomes and Spearman’s correlation for binary outcomes with p≤0.05 for
significance. 3D models with cinematic-rendering techniques were created.
Results or Findings: CTA datasets from 20 patients with 104 perforators were analysed. The DIEA diameters varied from 0.76 to
1.93 mm (mean 1.27±0.24 mm), the distance from the vessel to the linea alba varied from 1.25 to 8.74 cm (4.46±1.52 cm), and the
distance to the umbilicus varied from 2.81 to 9.73 cm (5.58±1.37 cm). A statistically significant correlation was found between the
diameter of DIEA and the quantity of perforators (p=0.575). There was no statistically significant correlation between DIEA’s diameter
and its perforators’ diameters (p=0.233). No correlations were found between DIEA’s diameter and the diameter of iliac arteries
(p1=0.091; p2 =-0.051; p3=0.049) and between the presence of calcinosis in the common iliac artery (U=43, p>0.05).
Conclusion: This is one of a few studies to analyse the use of preoperative mapping of DIEA for harvesting DIEP flaps with CTA. The
results demonstrated the usefulness of CTA and 3D reconstruction in studying DIEA and DIEP variable anatomy for preoperative
mapping.
Limitations: The study's limitations are its retrospective nature and the insufficient number of patients.
"Radiological Cases: Quiz Time!": a new approach for the education of radiology (8 min)
Antonio Vitor Nascimento Martinelli Braga; Salvador / Brazil
135
Author Block: A. V. N. Martinelli Braga, A. L. Costa, C. F. Lins; Salvador/BR
Purpose: In this study, we aim to analyse medical students' perception regarding the "Radiological Cases: Quiz Time!" gamification
tool as a novel approach to radiology education.
Methods or Background: The gamification tool "Radiological Cases: Quiz Time!" was elaborated by radiology monitors guided by a
radiologist using the "Kahoot" online platform. The class was divided into groups, and multiple questions regarding radiological cases
were asked. Then, students were invited to answer a questionnaire that assessed their sociodemographic profile, self-assessment of
learning radiological anatomy and the student's opinion about the radiological workshop "Radiological Cases: Quiz Time!". The
questions were based on the Likert-modified scale, totalling 35 questions. Incomplete questionnaires were excluded. The Cronbach's
alpha analysis was performed, and values above 0,7 were considered acceptable.
Results or Findings: Of the 150 invited, 59 completed the questionnaire, averaging 20±5 years old and from those, 34 (57,62%)
were women. Cronbach's alpha was 0,94, attesting to the internal consistency of the study. In total, 44 (74,57%) considered that the
gamification tool contributed substantially to learning; 48 (81,35%) considered the workshop to have high educational value; 44
(74,57%) judged the material as clinically significant for their future clinical practice/experience; and 53 (89,83%) considered that the
"Radiological Cases: Quiz Time!" contributed to form friendship bonds.
Conclusion: The workshop "Radiological Cases: Quiz Time!" is an interactive, creative and innovative tool for teaching radiology,
promoting clinically meaningful learning and the formation of new bonds of friendship.
Limitations: The study's main limitation was the low response to the questionnaire as it was administered at the end of the
semester, close to the holidays.
Ethics committee - additional information: The study received institutional review board approval, and written informed consent
was obtained from all participants. CAAE: 66828923.2.0000.5544.
Imaging in pulmonary embolism: a medical student's approach (8 min)

Violetta Benedek; Debrecen / Hungary
Author Block: V. Benedek; Debrecen/HU

Purpose: Pulmonary embolism is a potentially fatal status, and diagnosing and treatment needs an institutional environment. For a
medical student, it is important to be familiar with the most common pathologies and available diagnostic and therapeutic
techniques. We should also be aware of limitations and potential harms. Proper referrals and awareness of the ALARA principle are
essential for every physician. My goal was to evaluate chest CT referrals in pulmonary embolism.
Methods or Background: Chest CT angiography studies were selected during August and September 2022 in two emergency
centres at the University of Debrecen, Hungary. CT scans, D-dimer levels and other clinical findings (including pregnancy, known
malignancy or inflammatory disease) were collected in a database. The data of 380 patients (171 male and 209 female) was
evaluated. Additionally, 905 chest CT studies using different imaging protocols were evaluated to compare the applied doses in the
same period.
Results or Findings: Elevated D-dimer (>0,5 μg/ml) values were measured in 344 (90,52%) cases, and 32 patients (9,3%) were
diagnosed with pulmonary embolism. Only three patients had pulmonary embolism without elevated D-dimer value. A comparison of
applied doses among different chest CT protocols showed that pulmonary embolism protocol study doses are significantly lower than
in other protocols.
Conclusion: Pulmonary embolism is a commonly referred question by emergency departments, while positivity or the ratio of other
pathology findings is low during imaging. When requesting these examinations, it is crucial to consider the patient's clinical status,
various scoring systems, and additional diagnostic tests. The role of radiologists and radiographers is to design appropriate imaging
protocols and workflows supporting referring physicians in finding the most appropriate diagnostic tool for their patient's clinical
condition. The role of proper referrals should be emphasised in undergraduate training.
Limitations: Only limited clinical conditions were considered during the patient status evaluation.
Ethics committee - additional information: The study was approved with the registration number: DE KK RKEB.IKEB. 6291-2023.
Volume assessment of affected lung parenchyma in COVID-19 (8 min)

Peder Jørgensen Bruun; Bergen / Norway
136
Author Block: P. J. Bruun, A. P. Parkar; Bergen/NO
Purpose: This study aims to investigate if quantitative CT volume assessment as a stand-alone criterion of affected lung parenchyma
is helpful in distinguishing non-critically ill and critically ill patients with COVID-19.
Methods or Background: Patients admitted between March 2020 and December 2021 with an RT-PCR-confirmed COVID-19
diagnosis and a chest CT were included retrospectively. Patients were divided into two groups: critically ill (patients who died or were
admitted to the ICU) and non-critically ill for the others. The percentage of affected lung parenchyma was analysed by two observers
using a semi-quantitative method. The volumes and the basic demographic data were collected, and reliability between the observers
was assessed. Statistical analyses were done in SPSS.
Results or Findings: 67 patients (41 males and 26 females) were included. Eleven patients were admitted to the ICU, and five died
without being admitted to the ICU. The mean volume of affected lungs was 35% in males and 31% in females. Lung volume affection
over 60% led to ICU admittance, but lower values did not exclude ICU admittance. Depending on which cut-off value was used to
distinguish the groups (20%-40%-60%), the sensitivity was 88%-79%-80%, but the specificity was low at 26%-24%-32%. The PPV
increased 24%-75%-100%, but the NPV dropped 88%-38%-19%. A cut-off to distinguish the two groups could not be determined.
Reliability between the two observers was good (ICC=0.8).
Conclusion: Semi-quantitative CT-volume assessment of affected lung parenchyma as a stand-alone criterion is not helpful in
distinguishing between non-critically ill and critically ill patients with COVID-19. The use of lung CT for evaluating the severity of the
disease is not advisable.
Limitations: Analysers were not completely blinded to all patient outcomes. The time between admittance and CT was not
standardised.
Ethics committee - additional information: The study was approved by the Norwegian regional ethics committee (REK).
137
RPS 416 - Radiomics and artificial intelligence
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Oncologic Imaging
CME Credits: 1.5
Moderator:
Doris Leithner; New York / United States
Handcrafted radiomics, deep radiomics and transcriptomics data provide complementary and potentiating prognostic
information in soft-tissue sarcoma patients (7 min)
Amandine Crombé; Talence / France
Author Block: A. Crombé, C. Lucchesi, F. Bertolo, M. Kind, A. Michot, R. Perret, F. Le Loarer, A. Bourdon, A. Italiano; Bordeaux/FR
Purpose: The purpose of this study was to identify subgroups of soft-tissue sarcoma (STS) patients using handcrafted and deep
radiomics, to understand them, and investigate their impact on metastatic relapse-free survival (MFS).
Methods or Background: We included all consecutive adults with newly diagnosed locally-advanced STS managed at our sarcoma
centre between 2008 and 2020, with contrast-enhanced (CE) baseline MRI. After MRI post-processing, segmentation and
reproducibility assessment, 175 handcrafted radiomics features (h-RFs) from T1-weighted imaging (WI), T2-WI and fat-suppressed CE-
T1-WI were calculated. Convolutional autoencoder neural network (CAE) and half-supervised CAE (HSCAE) were trained in repeated
cross-validation on CE-T1-WI from one training cohort (n=200 patients) and validated on a testing cohort (n=25 patients), to extract
1024 deep radiomics features (d-RFs) per model. Following RNAseq of 110 samples, gene expression levels were calculated.
Unsupervised classifications based on h-RFs, CAE, HSCAE and RNAseq were built with hierarchical clustering and explained according
to histological features, radiological features, gene expression, pathway and survival analyses.
Results or Findings: 225 patients were included (120 men [53.3%], median age: 62 years). Three radiomics classifications were
obtained (h-RF, CAE and HSCAE groups), which were not associated with the transcriptomics groups, but with prognostic radiological
features known to correlate with higher grade (all P-values<0.001), and Sarculator groups (all P-values<0.001). HSCAE and h-RF
groups were also associated with MFS in multivariable Cox regressions (P =0.0146 and 0.0043, respectively). Combining these groups
improved the prognostic performances of the transcriptomics groups alone (c-index=0.603, increasing to 0.666 with h-RF [P=0.0380]
and 0.709 with HSCAE [P=0.0110]). Fifteen genes were dysregulated and two pathways were up-regulated in the h-RF groups, which
were linked to tumorigenesis and immune response.
Conclusion: Radiophenotypes of STS on pre-treatment MRI obtained with handcrafted and deep radiomics were explainable by
radiologists, independently associated with MFS and strengthened transcriptomics signature.
Limitations: This is a retrospective single-centre study.
Ethics committee - additional information: The study was approved by the Institutional Review Board of Bergonié Institute,
comprehensive cancer centre of Bordeaux, France.
CT lung cancer screening: pricing and cost-saving potential for deep learning computer-aided lung nodule detection
software (7 min)
Mathias Prokop; Nijmegen / Netherlands
138
Author Block: Y. Du1, M. Greuter2, M. Prokop3, G. De Bock2; 1Hangzhou/CN, 2Groningen/NL, 3Nijmegen/NL
Purpose: This study aimed to explore appropriate pricing for commercial deep learning computer aided detection (DL-CAD) systems
in different modes of use to maximise cost savings and identify the most cost-effective mode for lung cancer screening.
Methods or Background: We evaluated DL-CAD as a concurrent, prescreening, and second reader in three representative
countries. A scoping review was conducted to estimate the radiologist reading time with and without DL-CAD. The hourly cost of
radiologist time was collected for the US, UK and Poland, and the monetary equivalent of saved time was calculated. The minimum
number of screening CTs needed to reach break-even for a one-time investment for a DL-CAD was calculated.
Results or Findings: The mean reading time per case without DL-CAD was 2.5 minutes. It decreased by more than one minute when
using DL-CAD as a concurrent and prescreening reader, respectively. It increased by about a half minute for DL-CAD as second
reader. These reading times translated into costs of one to four euros per case for concurrent reading and one to six euro for
prescreening reading. To reach break-even with a one-time investment for a DL-CAD, the minimum number of CT scans was about
12,000-54,000 for concurrent reader and 9,000-65,000 for prescreening reader.
Conclusion: Based on the current pricing, it is necessary for the per case cost to be significantly below €6 or for DL-CAD to be used
in a high-workload setting to reach break-even in lung cancer screening. The use of DL-CAD as a prescreening reader has the greatest
potential for cost savings.
Limitations: This study focused on the costs associated with DL-CAD, as it was beyond the study's scope to consider downstream
costs related to diagnosis and treatment.
Funding for this study: This work is a part of NELCIN-B3 project. The NELCIN-B3 project is funded by The Royal Netherlands
Academy of Arts and Sciences (Grant No. PSA_SA_BD_01) and Ministry of Science and Technology of the People's Republic of China,
National Key R&D Program of China (Grant No. 2016YFE0103000).
Ethics committee - additional information: Because this article does not contain any studies on human or animal subjects, ethics
committee approval was not sought.
A fully automated deep learning model based on multiparametric imaging for predicting tumour recurrence of locally
advanced rectal cancer after neoadjuvant chemoradiotherapy: a multicentre study (7 min)
Zonglin Liu; Shanghai / China
Author Block: Z. Liu, Y. Sun, T. Tong; Shanghai/CN

Purpose: The purpose of this study was to develop and validate a fully automated deep learning model based on multimodal MRI for
DFS prediction of locally advanced rectal cancer (LARC) patients treated with neoadjuvant radiotherapy and chemotherapy (nCRT).
Methods or Background: A total of 462 rectal cancer patients treated with nCRT from three centres were retrospectively enrolled,
including clinical information, baseline multimodal MRI images (T2, ADC, Dapp, Kapp), and DFS data. The data from centres I and II
were combined (373 cases) and randomly assigned to training, validation, and internal testing sets. The data from centre III were
used as the external testing (89 cases). We developed a multitask joint survival model that simultaneously performed segmentation,
risk classification, and survival prediction. These multitasks collaborated with each other, contributing to fully exploiting the key
features of both imaging data and clinical data. Considering the multimodal input, an attention mechanism was introduced to
efficiently capture useful information within and between modalities to minimise the impact of noise on experimental results. In
addition, we experimented with conventional singletask models and compared the performance with our model.
Results or Findings: For risk classification, our model achieved significantly better performance than the singletask model, with an
AUC of 0.960 vs 0.782 in the internal testing set, and remained at an AUC of 0.767 in the external testing set; for segmentation and
survival prediction tasks, our model achieved a dice of 0.748 and a C-index of 0.731 in internal testing set, respectively.
Conclusion: Multitask deep learning model are expected to provide fully automated one-step prediction, contributing to optimising
personalised treatment for LARC patients.
Limitations: Sample size were relatively small. Predictions were only made for DFS and not implemented for OS.
Funding for this study: Funding for this study was provided by the National Natural Science Foundation of China
(81971687,82001776).
Prognostic value of the consensus molecular subtype 4 (CMS4) predicted by multiparametric radiomics-based machine
learning in colorectal cancer: a multicentre retrospective study (7 min)
Zonglin Liu; Shanghai / China
139
Author Block: Z. Liu, Y. Sun, T. Tong; Shanghai/CN
Purpose: The consensus molecular subtype (CMS) is a novel classification system that reflects the genetic characteristics of the
tumour. Among the four subtypes, CMS4 is associated with the worst prognosis. This study aimed to investigate whether a radiomics-
based machine learning approach could predict CMS4 status in CRC patients.
Methods or Background: A total of 228 CRC cases from three centres were retrospectively included. Cases from centre I were
divided into training (138 cases) and validation sets (33 cases) in an 8:2 ratio; cases from centre II and III were combined as the
external testing set (57 cases). Sequencing data and baseline MRI images, including T2-weighted (T2WI) and contrast-enhanced (CE)
sequences, were available for each case. The sequencing data was input into the CMS classification system to generate CMS subtype
outcomes. Radiomics features from the two sets were extracted with the same parameter settings. Several machine learning
algorithms were applied in sample balance, feature normalisation, feature filters, and classifier construction to explore the best-
performing and most robust model for CMS4 prediction. The rad-score for each patient was calculated by the T2WI and CE models
separately. The combined model was established by applying logistic regression on the results of the above two models.
Results or Findings: We found that the CE model achieved better performance than the T2 model in both the test set (0.815 vs
0.790) and external validation set (0.741 vs 0.702). After merging the two models, the predictive performance of the Merged model
was further improved, with the AUCs of 0.855 and 0.759 in the test set and external validation set.
Conclusion: Multiparametric radiomics-based machine learning shows promising potential in distinguishing CMS4 from other
subtypes of CRC.
Limitations: The study's relatively small sample size as well as the manual delineation of the lesions' contours were identified as
limitations.
Funding for this study: Funding for this study was provided by the National Natural Science Foundation of China
(81971687,82001776).
Interlesional morphological heterogeneity as a novel noninvasive prognostic biomarker (7 min)

Zuhir Elkarghali; Amsterdam / Netherlands
Author Block: Z. Elkarghali, M. Hattink, O. Maxouri, S. Rostami, D. I. Rodríguez Sánchez , S. Trebeschi, R. G. H. Beets-Tan;
Amsterdam/NL
Purpose: Interlesional genetic heterogeneity is an established fact in tumour biology; however, performing a biopsy on every lesion
is not feasible. For the first time in the literature, we explore the use of medical image analysis techniques to quantify morphological
heterogeneity between lesions as a proxy for biological heterogeneity and analyse its value in stratifying patients based on prognosis.
Methods or Background: We collected a diverse pancancer cohort of 1692 CE-CT scans with genetically proven diagnoses and
performed complete 3D tumour segmentation. From each delineation (n=11,268 lesions), we derived radiomic features that fully
characterise each lesion's morphology. We utilised seven similarity distance metrics (Euclidean, Chebyshev, City-Block, Minkowski,
Correlation, Bray-Curtis, and Cosine) to measure the median morphological dissimilarity of lesions in a patient. Survival analysis (log-
rank test) was performed to compare patients with high or low interlesional morphological heterogeneity (relative to the data set level
median distance).
Results or Findings: We computed seven distance metrics for every combination of lesions within a patient and calculated the
median as a patient level metric of morphological interlesional heterogeneity. Chebshev (Χ2=12.49, P =0.000408), City-Block
(Χ2=12.08, P =0.000508), Euclidean (Χ2=11.64, P =0.000646), and Minkowski (Χ2=11.64, P=0.000646) distance measures could all
strongly stratify patients into high- and low-risk groups. Cosine (Χ2=4.13, P=0.042) and correlation (Χ2=3.99, P=0.046) similarity
metrics were also predictive to a lesser extent. Bray-Curtis distance measures could not significantly stratify patients (Χ2=3.35,
P=0.067).
Conclusion: Interlesional morphological heterogeneity, as measured by radiomics and similarity distance metrics, strongly predicted
overall survival.
Limitations: External validation has yet to be performed as a proof-of-concept study despite including over 1500 real-world cases.
Ethics committee - additional information: IRB approval was granted: IRBd19-147.
Clinical assessment of deep learning reconstruction-based accelerated rectal MRI (7 min)

Wenjing Peng; Beijing / China
140
Author Block: W. Peng, L. Wan, X. Tong, F. Yang, S. Wang, L. Li, H. Zhang; Beijing/CN
Purpose: The purpose of this study was to conduct a clinical assessment of deep learning reconstruction (DLR)-based rectal MRI in
comparison to standard MRI.
Methods or Background: Patients with biopsy-proven rectal adenocarcinoma from November/2022 to May/2023 were prospectively
enrolled in the study to undertake rectal MRI, including protocols using standard fast spin-echo (FSEstandard) and DLR-based
accelerated FSE (FSEDL). Imaging quality including signal-noise ratio (SNR), contrast-noise ratio (CNR), as well as subjective scoring
based on Likert scale were assessed by two radiologists. Diagnostic performance including T-staging, N-staging, EMVI, and MRF was
further evaluated by five radiologists. The time consumed in the application of each diagnostic metric was documented for reading
efficiency analysis.
Results or Findings: In total, 117 patients (77 males; age range 21 – 77 years) were enrolled in the study; 60 patients undertook
radical surgery. DLR enabled a reduction of 65% in acquisition time. Moderate to excellent intra- and interreader agreement was
achieved for all assessment metrics. FSEDL exhibited higher SNR, CNR, and subjective scores in noise, tumour margin clarity,
visualisation of bowel wall layering and rectal mesorectal fascia, overall image quality, and diagnostic confidence (P < 0.05). FSEDL
was rated higher T-staging accuracy by junior readers (reader 1, 58% vs 70%, P = 0.016; reader 3, 60% vs 76%, P = 0.021), with
comparable performance in evaluating N-staging, EMVI, and MRF. No difference was found concerning diagnostic performance by
senior readers (P > 0.05). FSEDL exhibited shorter diagnostic time in T-staging and overall evaluation by all readers, as well as in
EMVI and MRF by junior readers (P < 0.05).
Conclusion: FSEDL is clinically feasible for rectal MRI, which could facilitate improved image quality and reading efficiency than FSE
standard, while reducing 65% acquisition time. Moreover, it has potential in helping junior radiologists improve the accuracy of T-
staging.
Limitations: This was a single-centre study.
Funding for this study: Funding was received from the CAMS Innovation Fund for Medical Sciences (CIFMS) [grant number 2021-
I2M-C&T-A-017], Capital's Funds for Health Improvement and Research (CFH) [grant number 2022-2-4024], the National Natural
Science Foundation of China [grant number 81971589], 2020 SKY Imaging Research Fund [grant number Z-2014-07-2003-01], and
CAMS Innovation Fund for Medical Sciences (CIFMS) [grant number 2022-I2M-C&T-B-077].
Ethics committee - additional information: This prospective study received approval from our institutional review board, and
written informed consent was obtained from all participants.
Radiomic analysis of PI-RADS 4 and 5 lesions detected on 3T mpMRI: role in the diagnosis of clinically significant
prostate cancer (7 min)
Pietro Andrea Bonaffini; Monza / Italy
Author Block: P. A. Bonaffini1, A. Corsi2, R. Muglia1, G. Perugini3, M. Roscigno3, L. F. Da Pozzo3, P. Marra1, S. Sironi1; 1Monza/IT,
2
Liege/BE, 3Bergamo/IT
Purpose: The purpose of this study was to identify radiomic features potentially supporting the detection of clinically significant
prostate cancer (csPC) in PI-RADS 4/5 lesions detected on 3T multiparametric MRI (mpMRI) studies.
Methods or Background: We retrospectively enrolled patients who underwent a 3T mpMRI (June 2016-March 2021) and with at
least one PI-RADS 4-5 lesion (PI-RADS v2.1). Final pathological findings from fusion MRI-targeted biopsies served as ground truth.
Clinical (age, PSA, PSA density) and MRI conventional parameters (prostate volume, mean ADC in circular 2D ROI) were collected.
Included lesions were manually contoured on itk-SNAP on ADC maps, and axial T2 images; volumes of interest were also obtained.
Radiomic features were extracted using Pyradiomics. Clinical and radiomic features best correlating with final histological results were
selected. All models were assessed through 100 repetitions using 5-fold cross-validation. Sensitivity and specificity were assessed on
validation samples.
Results or Findings: Among 945 patients who had undergone prostate mpMRI within the study period, 99 patients (median age 69
years) with 111 PI-RADS 4-5 lesions met the inclusion criteria. At the end of the histopathological analysis, 79 lesions (71%) were
found to be csPC (GS≥7). The best predicting clinical (PSA density) and radiomic (ADC-wavelet-
LLL_glrlm_LongRunHighGrayLevelEmphasis/texture feature, T2-wavelett-HHH_glszm_GrayLevelVariance and T2-wavelet-
LLL_glszm_GrayLevelVariance/heterogenicity feature) multivariate model for PI-RADS 4-5 lesions obtained 79% sensitivity, 80%
specificity, 91% PPV, 63% NPV and 79% accuracy. When considering only peripheral zone lesions, a multivariate model with only the
same radiomic features gained 86% sensitivity, 80% specificity, 93% PPV, 70% NPV and 84% accuracy.
Conclusion: Texture and heterogeneity features extracted from 3T mpMRI T2 and ADC sequences may improve the detection of
csPC in PI-RAD 4 and 5 lesions, demonstrating a better performance when considering only peripheral zone lesions and correlating
also with PSA density in the zone-ignorant model.
Limitations: This was a monocentric retrospective study. The sample size was limited.
Ethics committee - additional information: This study was conducted following the Declaration of Helsinki. Patients' anonymity
was granted.
Imaging derived biomarkers integrated with clinical and laboratory values predict recurrence of hepatocellular
carcinoma after liver transplantation (7 min)
141
Thi Phuong Thao Hoang; Ho Chi Minh / Vietnam
Author Block: T. P. T. Hoang1, P. Schindler2, N. Börner1, M. Masthoff2, M. Seidensticker1, J. Ricke1, M. Ingrisch1, O. Öcal1, M.
Wildgruber1; 1Munich/DE, 2Münster/DE
Purpose: The purpose of this study was to investigate the prognostic value of computed tomography (CT) derived imaging
biomarkers in hepatocellular carcinoma (HCC) recurrence after liver transplantation (LT) and develop a predictive nomogram model.
Methods or Background: This retrospective study included 178 patients with histopathologically confirmed HCC who underwent
liver transplantation between 2007 and 2021 at the two academic liver centres. We evaluated dedicated imaging features from
baseline multiphase contrast-enhanced CT supplemented by several clinical findings and laboratory parameters. Time-to-recurrence
(TTR) was estimated by Kaplan–Meier analysis. Univariable Cox proportional hazard regression and multivariable least absolute
shrinkage and selection operator (LASSO) regression were used to identify independent prognostic factors for recurrence. A
nomogram model was then built based on the independent factors selected through LASSO regression, to predict the probabilities of
HCC recurrence at one, three, and five years.
Results or Findings: The rate of HCC recurrence after LT was 17.4% (31 of 178). The LASSO analysis revealed six independent
predictors associated with an elevated risk of tumour recurrence. These predictors included the presence of peritumoural
enhancement, the presence of over three tumour lesions, the largest tumour diameter exceeding 3 cm, serum alpha-fetoprotein (AFP)
levels surpassing 400 ng/mL, and the presence of a tumour capsule. Conversely, a history of bridging therapies was found to be
correlated with a reduced risk of HCC recurrence. In addition, Kaplan-Meier analysis with log-rank test showed patients with irregular
margins, satellite nodules, or small lesions displayed significantly shorter time-to-recurrence. Our nomogram demonstrated good
performance, yielding a C-index of 0.835 and AUC values of 0.86, 0.88, and 0.85 for the predictions of 1-year, 3-year, and 5-year TTR,
respectively.
Conclusion: Imaging parameters derived from baseline contrast-enhanced CT showing malignant characteristics and aggressive
growth patterns, along with serum AFP and a history of bridging therapies, can serve as biomarkers for predicting HCC recurrence
after transplantation.
Limitations: First, this study has a limited sample size. Second, patients with various types of bridging therapies have been included.
Although this situation causes inhomogeneity, it reflects the daily clinical routine of large transplantation centres.
Ethics committee - additional information: This study was approved by an ethics committee; the notification can be found under
the number 22-0110.
CT- and MRI-based multiparametric radiomics improve the preoperative prediction of lymph node metastasis in
patients with rectal cancer (7 min)
Yue Niu; Changsha / China
Author Block: Y. Niu1, X. Yu1, Y. Wang2; 1Changsha/CN, 2Shanghai/CN

Purpose: The purpose of this study was to compare the performance of computed tomography (CT)- and magnetic resonance
imaging (MRI)-based multiparametric radiomics models for individual preoperative prediction of lymph node metastasis (LNM) in
rectal cancer (RC) patients.
Methods or Background: 234 rectal adenocarcinoma patients were retrospectively selected in this study and were randomly
divided into training (n = 164) and testing (n = 70) cohorts by 7:3. The radiomics features of the primary tumour were extracted from
non-contrast-enhanced CT (NCE-CT), contrast-enhanced CT (CE-CT), T2-weighted MRI (T2WI), and gadolinium contrast-enhanced T1-
weighted MRI (CE-T1WI) of each patient to build different radiomics models, respectively. Clinical-radiomics models were established
by combing radiomics features with clinical information. A clinical-IMG model was built by combining all the clinical and radiomics
features via binary least absolute shrinkage and selection operators. A clinical-radiomics nomogram was developed using
multivariable logistic regression analysis. Model performance was evaluated by using the area under the receiver operating
characteristic curve (AUC) and decision curve analysis (DCA).
Results or Findings: The MRI radiomics model in the validation cohort significantly outperformed the CT radiomics model (AUC=
0.785 vs 0.721, P<0.05). The Clinical-IMG nomogram showed the highest prediction efficiency compared to all other models (all
P<0.05), with an AUC of 0.828, sensitivity of 0.759 and specificity of 0.780 in the validation cohort.
Conclusion: The MRI radiomics model performed better than both CT radiomics model and clinical model in predicting LNM in RC.
The clinical-radiomics nomogram that combines radiomics features obtained from both CT and MRI, along with preoperative clinical
characteristics, exhibited the best diagnostic performance.
Limitations: This was a single-centre study and there was no external validation.
CT texture analysis as a predictor for the genetic profile of mass-forming intrahepatic cholangiocarcinoma (7 min)
Angela Ammirabile; Milan / Italy
142
1 1 1 2 1 1 1 1 1 2
Author Block: A. Ammirabile , L. Viganò , V. Zanuso , F. Fiz , M. E. Laino , M. Francone , M. Sollini , L. Rimassa ; Milan/IT, Genoa/IT
Purpose: Intrahepatic cholangiocarcinoma (ICC) is an aggressive disease with increasing incidence. Comprehensive molecular
profiling has shown genetic alterations that could be the target of systemic therapies. Texture analysis of imaging has led to a reliable
prediction of pathology data. This study investigates whether CT-based radiomics can non-invasively predict genetic alterations in
ICC.
Methods or Background: All consecutive patients eligible for systemic therapy for a mass-forming ICC (01/2016-06/2022) were
considered. Inclusion criteria were: the availability of a contrast-enhanced CT at diagnosis before any treatment with an adequate
quality of the portal phase for textural analyses, complete molecular profiling by NGS or FISH evaluation for FGFR2 gene
fusion/rearrangement. Genetic analyses were performed on surgical specimen or biopsy. The tumour was manually segmented and
radiomic features were automatically extracted using the LifeX software. Predictive models were built considering clinical and
radiomic data.
Results or Findings: 90 patients were enrolled (58 NGS,32 FISH) with a median age of 65 years. The most common genetic
alterations were FGFR2 (20/90), IDH1-2 (12/58), KRAS (9/58). The performances of the predictive models for FGFR2 and IDH1-2
improved by adding radiomic features to clinical data, reaching a C-index of 0.892 (vs 0.800 of the clinical model) and 0.811 (vs
0.670), respectively, at internal validation. The pure radiomic model for the prediction of KRAS mutations achieved a C-index of 0.862
at internal validation (vs 0.660 of the pure clinical model) without further improvements with the addition of clinical features.
Conclusion: The radiomic features extracted from CT at ICC diagnosis can potentially provide a reliable noninvasive prediction of its
genetic status with a major impact on therapeutic strategies.
Limitations: The limitations of the study are its retrospective nature, lack of external validation as well as the commonest mutations
being the subject of analysis.
Funding for this study: Funding was provided by the AIRC grant #2019−23822.
Ethics committee - additional information: The local review board of each centre approved the study protocol (coordinating
centre approval: protocol number 142/21, date of approval 17/03/2021). Because of the retrospective nature of the study, the need
for specific informed consents was waived.
CT-based radiomics of cholangiocarcinoma and peritumoural tissue improves survival prediction: development of a
clinical-radiomic model (7 min)
Author Block: A. Ammirabile1, F. Fiz2, S. Langella3, M. Serenari4, M. Sollini1, A. Chiti1, G. Torzilli1, F. Leva1, L. Viganò1; 1Milan/IT,
2
Genoa/IT, 3Turin/IT, 4Verona/IT
Purpose: In patients affected by intrahepatic cholangiocarcinoma (ICC), the prediction of survival based on morphological and clinical
parameters has limited reliability. The present study aims to elucidate if the textural features of ICC and its peritumoural tissue
extracted from preoperative computed tomography (CT) improve the prediction of survival after resection.
Methods or Background: All consecutive patients undergoing resection for ICC at six high-volume centres (2009-2019) were
considered. The arterial and portal phases of CT performed <60 days before surgery were analysed. The tumour was manually
segmented (tumour-VOI), a 5-mm automatic volume expansion was applied to encompass the peritumoural tissue (margin-VOI). The
radiomic features were automatically extracted by the LifeX software. For overall and progression-free survival (OS/PFS), we
considered pre- and post-operative predictive models, based on clinical data and radiomic features from portal and arterial phases.
Results or Findings: 215 patients were included (median age 67.5 years). The three-year OS/PFS rates were 57.0% and 34.9%
(median follow-up 28 months). The predictive model of OS based on clinical variables had a C-index of 0.681. The performance
progressively improved by adding the radiomic features: C-index =0.710 including portal tumor-VOI, C-index =0.752 including portal
tumour-Margin-VOI; C-index=0.764 including all arterial and portal VOIs. The latter model retained clinical variables (CA19-9, tumour
pattern), tumour features (density, homogeneity, GLRLM indices), and margin data (kurtosis, compacity, shape). The model had a
performance equivalent to the post-operative clinical model including the pathology data (C-index =0.765). The same results were
observed for PFS.
Conclusion: CT-based radiomics of ICC and peritumoural tissue improves prediction of survival, and, in combination with clinical
data, leads to a preoperative estimation of outcome equivalent to the post-operative one.
Limitations: The limitations of the study are its retrospective nature, lack of external validation as well as the absence of radiomic
features from CT late-phase.
Funding for this study: Funding was provided by the AIRC grant #2019−23822.
Ethics committee - additional information: The local review board of each centre approved the study protocol (coordinating
centre approval: protocol number 142/21, date of approval 17/03/2021). Because of the retrospective nature of the study, the need
for specific informed consent was waived.
Radiomics in colon cancer: how to identify high-risk patients (7 min)

Michela Polici; Rome / Italy
143
Author Block: M. Polici, D. Valanzuolo, D. Pugliese, G. Tremamunno, F. Palmeri, M. Zerunian, D. De Santis, D. Caruso, A. Laghi;
Rome/IT
Purpose: The study aimed to develop a radiomic model able to identify high-risk colon cancer by analysing properative CT scans.
Methods or Background: The study population included 300 patients with nonmetastatic colon cancer were retrospectively enrolled
from January 2015 to June 2020. The population was divided into two groups, high-risk and no-risk, following the presence of at least
one high-risk clinical factor between staging T4, LVI, PNI, budding, and nodal metastases. All patients had baseline CT scans, and 3D
cancer segmentation was performed on the portal phase by two expert radiologists using open-source software (3DSlicer v4.10.2).
Among the 107 radiomic features extracted, stable features were selected to evaluate the interclass correlation (ICC) (cut-off ICC >
0.8). Stable features were compared between the two groups (T-test or Mann–Whitney), and the significant features were selected for
univariate and multivariate logistic regression to build a predictive radiomic model. Furthermore, survival analyses were performed
with Kaplan-Meier curves, with progression within 24 months considered as end-point.
Results or Findings: In total, 210/300 were classified as high-risk and 90/300 as no-risk. A total of 27 radiomic features were stable
(0.80 ≤ ICC < 0.92). Among these, 15 features were significantly different between the two groups (P < 0.05), and only eight features
were selected to build the radiomic model. The radiomic model yielded an AUC of 0.73. Three radiomic features demonstrated
correlation with progressive disease in Kaplan-Meier curves.
Conclusion: In conclusion, the radiomic model could be seen as a performant, non-invasive imaging tool to properly stratify colon
cancers with high-risk disease especially in preoperative setting.
Limitations: The retrospective nature of the study was identified as a limitation; the study populations were unbalanced.
approval was obtained.
144
RPS 407 - Kidney and bladder imaging
Categories: Genitourinary, Imaging Methods, Oncologic Imaging

CME Credits: 1.5
Moderator:
Rossano Girometti; Udine / Italy
Validation of the bladder cancer “MRI Pathway”: results from a multicentre prospective study (7 min)
Emanuele Messina; Rome / Italy
Author Block: E. Messina, A. Dehghanpour, A. Borrelli, S. Lucciola, C. Catalano, V. Panebianco; Rome/IT

Purpose: The validate a new bladder cancer (BCa) diagnostic pathway focusing on magnetic resonance imaging (MRI). Moreover, to
identify clinical, pathological, and radiological characteristics, including the Vesical Imaging-Reporting and Data System (VI-RADS)
score, individually demonstrating a significant correlation with muscle-invasive BCa (MIBC).
Methods or Background: Prospective multicentres study (four centres). Patients with BCa suspicion underwent MRI before trans-
urethral resection of bladder tumour (TURBT). Two experienced uro-radiologists in each centre independently analysed MRIs.
According to VI-RADS, a cutoff of ≥ 3 or ≥ 4 was assumed to define MIBC. TURBT and/or cystectomy reports were compared with
preoperative VI-RADS scores to assess MRI accuracy for discriminating between non-muscle-invasive versus MIBC. Performance was
assessed by ROC curve analysis. Univariable and multivariable logistic regression models were implemented.
Results or Findings: A final cohort of 251 patients was enrolled (median age 73 [IQR: 64, 77.5]). MRI showed sensitivity, specificity,
PPV, NPV, and accuracy for MIBC detection ranging from 83-93%, 80-92%, 67-81%, 93-96%, and 84-89% for the more experienced
readers. The area under the curve (AUC) was 0.95 (0.91-0.99). In the multivariable logistic regression model, haematuria (p = 0.02),
tumour size (p=0.03), the VI-RADS score, using both a cutoff of 3 and 4 (p<0.0001), and concomitant hydronephrosis (p=0.02) were
the variables correlating with a bladder cancer staged as ≥ T2. The inter-reader agreement was substantial (k=0.85).
Conclusion: VI-RADS score has been validated as a reliable preoperative tool predicting BCa muscle-invasiveness. Moreover, if
integrated as an MRI biomarker with clinical and imaging data, VI-RADS can be integrated into a customised diagnostic approach to
select patients more precisely, with the goal of reducing the risk of diseases’ understaging and enhancing clinical outcomes.
Limitations: High-volume referral centres, reduced reproducibility limit this study.
Ethics committee - additional information: Institutional Ethics Committee approved this study.
Contrast-enhanced CT radiomic analysis for the preoperative prediction of pathological T3a upstaging in renal cell
carcinoma (7 min)
Enyu Yuan; Chengdu / China
145
Author Block: E. Yuan1, Y. Chen1, B. Song1, J. Yao1, H. Yin2, H. Zhang2; 1Chengdu/CN, 2Beijing/CN
Purpose: The study aims to develop and temporally validate the radiomic models for identifying pathological T3a upstaging in renal
cell carcinoma.
Methods or Background: A total of 1003 patients with renal cell carcinoma in our hospital were retrospectively enrolled. The
patients were assigned to development dataset (n = 729) and temporal validation dataset (n = 274). The pathological T3a status of
each patient was pathologically confirmed. Radiomic features of the lesion and the 5 mm peritumoural area were extracted from the
manually labeled portal venous phase CT images. Tumour model, peritumour model, and combine model were developed using
tumour, peritumour, and both features. The modeling pipelines were internally validated using 1000*100 nested cross validation, and
the final models were temporally validated in an independent temporal validation set. The model in each experiment was evaluated
by using metrics of discrimination, calibration, and clinical utility. The frequencies of features being selected were also analysed.
Results or Findings: The tumour shape based feature "sphericity" was the most frequently selected feature with the largest
coefficient in the final models. In internal validation, the tumour pipeline and combine pipeline showed similar discrimination (AUC =
0.8362±0.0327 vs. 0.8347±0.0331), calibration, and clinical utility, while the peritumour pipeline showed slightly but signification
poor discrimination (AUC = 0.7994±0.0369), calibration, and clinical utility than the other two. In temporal validation, the tumour
model, peritumour model, and combine model showed similar discrimination (AUC = 0.8457 vs. 0.8489 vs. 0.8426), calibration, and
clinical utility.
Conclusion: The radiomic models showed favorable performance in predicting pathological T3a upstaging preoperatively in renal cell
carcinoma patients.
Limitations: This is a single centre study. The models needs further multicentre and prospective validation.
Ethics committee - additional information: The written consent was waived because of the retrospective design.
Neoadjuvant chemotherapy VI-RADS (nacVI-RADS) for the assessment of tumour response to neoadjuvant
immunotherapy in muscle-invasive bladder cancer (7 min)
Giorgio Brembilla; Milan / Italy
Author Block: G. Brembilla, M. Cosenza, T. Russo, G. Basile, D. Raggi, C. Mercinelli, F. Montorsi, A. Necchi, F. De Cobelli; Milan/IT
Purpose: The aim of this study was to investigate the diagnostic accuracy and reproducibility of the neoadjuvant chemotherapy VI-
RADS (nacVI-RADS) in assessing MIBC response to neoadjuvant immunotherapy.
Methods or Background: Two hundred and twenty bladder MRI scans from 110 patients who underwent pre- and post-
immunotherapy MRI prior to radical cystectomy (RC) were retrospectively reviewed by two readers using nacVI-RADS scores. The
diagnostic accuracy of nacVI-RADS was assessed using histopathology of RC specimens as the standard of reference, relative to two
endpoints: complete pathologic response (ypT0) and downstaging (ypT1). The threshold for MRI positivity was considered nacVI-
RADS 3; for downstaging, an additional threshold of nacVI-RADS 4 was also tested. Interobserver agreement was assessed with
agreement coefficient 1 (AC1), Cohen’s k coefficient and percentages of agreement.
Results or Findings: Complete pathologic response was observed in 42% of patients. Complete radiologic response (i.e.: nacVI-
RADS 1-2) was reported in 35.5% and 49.9% by reader 1 and reader 2, respectively. nacVI-RADS 3 had a sensitivity and NPV for
>ypT0 residual disease of 82% (95%CI: 74-88) and 74% (95%CI: 64-83), respectively; for >ypT1 residual disease, sensitivity and NPV
were 97% (95%CI: 90-99) and 97% (95%CI: 91-99), respectively. For the assessment of residual >ypT1 disease, a potentially better
tradeoff between sensitivity, specificity and NPV was obtained for higher positivity thresholds of MRI. nacVI-RADS had an AUC for ypT0
of 79.6%, for ypT1 of 86.3%. Interreader agreement was substantial regardless the MRI positivity threshold (AC1 0.65-0.68; k
0.66-0.67).
Conclusion: nacVI-RADS criteria showed good accuracy and reproducibility in assessing MIBC response to neoadjuvant treatments.
Limitations: Single centre, experienced readers limit this study.
Funding for this study: Funding for this research was provided by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc.,
Rahway, NJ, USA, and Associazione Italiana per la Ricerca sul Cancro (AIRC), Grant number: MFAG: 2017 Id.20617.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Fondazione IRCCS Istituto
Nazionale dei Tumori, Milan, Italy.
Development and validation of a diagnostic model based on contrast-enhanced CT to identify clear cell renal cell
carcinoma in solid small renal masses: a multicentre study (7 min)
Jiayue Han; Zhuhai / China
146
Author Block: J. Han, Y. Tao, Y. Zhang; Zhuhai/CN
Purpose: The aim of this study was to develop and validate a diagnostic model based on contrast-enhanced CT for identifying clear
cell renal cell carcinoma (ccRCC) in small renal masses (SRMs).
Methods or Background: This retrospective multicentre study enrolled patients with pathologically confirmed SRMs. Data from
three centres were used as training set (n=229), with data from one centre serving as an independent external validation set (n=81).
Univariate and multivariate logistic regression analyses were used to screen independent risk factors for ccRCC and build the
classification and regression tree (CART) model. Three radiologists were asked to diagnose the SRMs in each case independently
based on professional experience and re-evaluated using the CART model.
Results or Findings: A total of 71% (220/310) of SRMs were ccRCC. Enhancement pattern, early dark cortical band (EDCB), the ratio
of lesion to normal cortex attenuation (L/C) in the corticomedullary phase, non-enhancing phase L/C and sex were used to develop the
CART model. In the testing cohort, the AUC and accuracy of the CART model were 0.903, 85.1%. The accuracy of radiologists was
67.9%,58.0%, and 56.8%. With the CART model support, the accuracy of radiologists improved to 86.4%,84.0%,79.0%. Interobserver
agreement was significantly improved with the use of model aids (0.323 vs 0.654, P＜0.001).
Conclusion: The CART diagnostic model can identify ccRCC in SRMs and help radiologists make the diagnosis, potentially reducing
the number of unnecessary biopsies.
Limitations: First, since it was a retrospective study, the existing selection bias may affect the results, and further prospective
verification is needed in future work. Second, the data were collected from different institutions, and the scanning protocols were
different. To minimise the impact of internal and external factors on the results, the quantitative data were standardised, and an
independent external validation set was established.
Funding for this study: This work was supported in part by the National Key Research and Development Program of China under
Grant Nos. 2023YFE0204300, in part by the National Natural Science Foundation of China under Grant No: 81801809, 82371917,
81830052, 81971691, 12126610, 62371476; in part by the Guangzhou Technology Program of Agriculture and Social Development of
Key Research and Development Scheme under Grant No: 2023B03J1237, in part by the Basic and Applied Basic Research Foundation
of Guangdong Province under Grant No: 2020A1515010572, and in part by the Zhuhai Basic and Applied Basic Research Foundation
under Grant No: ZH22017003200001PWC.
Ethics committee - additional information: This study was approved by the institutional review boards of the Fifth Hospital of Sun
Yat- sen University.
Comparison of photon-counting and energy-integrating detector CT systems for the characterisation of cystic renal
lesions on virtual non-contrast imaging (7 min)
Ludovica Lofino; Milan / Italy
Author Block: L. Lofino, F. Schwartz, M. Al Tarhuni, A. Abadia, F. Ria, E. Samei, D. Marin; Durham, NC/US
Purpose: The study aims to compare the absolute CT attenuation errors of cystic renal lesions and abdominal organs on virtual non-
contrast images (VNC) between photon-counting (PCCT) and energy-integrating detector CT (EID).
Methods or Background: In this HIPAA compliant, IRB-approved retrospective study, multiphase CT scans from one PCCT and two
EID dual-source dual-energy CTs were retrieved. A total of 56 BMI-matched patients (26 women, 30 men; mean age 58.5 ± 15.3
years; mean BMI 29.0 ± 6.8 kg/m2), were included: 16 for PCCT and 20 each per EID systems. Attenuation measurements of
abdominal organs were recorded on VNC and True Noncontrast (TNC) datasets. Furthermore, attenuation measurements of 16 cystic
renal lesions (eight for PCCT and eight for EID) were compared on VNC and TNC. Absolute CT attenuation errors were calculated and
compared between PCCT and EID for the entire population and a subset of 20 obese patients using paired t-tests. Absolute CT
attenuation errors were also compared for all cystic renal lesions and for renal lesions <1 cm, separately.
Results or Findings: PCCT yielded significantly lower absolute CT attenuation errors than EID on VNC for spleen (2.6 ± 6.2 vs 8.0 ±
10.3) and pancreas (4.4 ± 1.8 vs 7.7 ± 9.7) for all patients and for liver, spleen and pancreas in the obese patient cohort (P<0.05).
Furthermore, PCCT yielded significantly lower absolute CT attenuation errors compared to EID for all cystic renal lesions (2.0 ± 1.3 vs.
12.0 ± 8.9) and for renal lesions <1 cm (1.4 ± 0.9 vs. 19.1 ± 6.8), P<0.01.
Conclusion: PCCT has significantly lower absolute CT attenuation errors for abdominal organs and cystic renal lesions in VNC images,
compared to two dual-energy EID CTs.
Limitations: Single centre, retrospective study with limited number of cases.
Funding for this study: No funding was required for this study.
Ethics committee - additional information: Ethics comittee approved the study.
Diagnostic accuracy and inter-reader agreement of nac-VIRADS scoring system: a prospective validation study (7 min)
Ailin Dehghanpour; Rome / Italy
147
Author Block: A. Dehghanpour, E. Messina, A. Borrelli, S. Lucciola, C. Catalano, V. Panebianco; Rome/IT
Purpose: Recently a novel scoring system named nac-VIRADS (neoadjuvant chemotherapy VI-RADS) was proposed to assess
radiological response to chemotherapy in MIBC (muscle-invasive bladder cancer) patients using mpMRI.
The primary aim of this study was to validate the performance of nac-VIRADS scoring system and assess its inter-reader agreement.
Additionally, to investigate the radiological and clinicopathological features independently correlating with response to systemic
therapy.
Methods or Background: This prospective multicenter study included patients with non-metastatic MIBC who underwent trans-
urethral resection of bladder tumour (TURBT) and/or repeated TUR, followed by neoadjuvant chemotherapy (NAC), radical cystectomy
(RC), and extended pelvic lymph node dissection. Patients underwent pre- and post-treatment multiparametric MRI. Radiological
response was evaluated by two experienced urogenital radiologists using nac-VIRADS scoring system. The reference standard was
histopathologic reports and tumour regression grade (TRG) from RC.
Nac-VIRADS performance was assessed by means of ROC analysis, deriving sensitivity, specificity, PPV, NPV, and accuracy.
Interreader agreement was determined with Cohen’s k statistics. Univariable and multivariable analyses were implemented.
Results or Findings: Fifty-five patients were included. n=13 were classified as nac-VIRADS 1-2, n=14 as nac-VIRADS 3, n=22 as
nac-VIRADS 4, and n=6 as nac-VIRADS 5. Overall, nac-VIRADS score showed a high diagnostic performance in concordance with
histological results expressed by TRG. Diagnostic performance showed a range of 86%-92% sensitivity, 71%-79% specificity,
80%-82% (PPV, and 79%-83% NPV. AUC was 0.92 (95% CI: 0.83–1) for the more experienced reader. Inter-reader agreement was
almost perfect (K = 0.82).
Conclusion: nac-VIRADS scoring system offers a reliable and reproducible approach, employing a clear algorithm, to assess response
to systemic therapy in MIBC patients. Its implementation in a standardised pathway has the potential to significantly influence
therapeutic decision-making and enhance overall patient survival.
Limitations: Limited sample size limits the scope of this study.
Ethics committee - additional information: The study was approved by the Institutional Ethical Committee.
Novel CT-derived biomarkers for prediction of renal split function in preoperative nephrectomy planning (7 min)
Christopher Owen Page; London / United Kingdom
Author Block: C. O. Page, D. Ap Dafydd, C. Kelly-Morland, W. Cazzaniga, D. Nicol, D. Levine, S. J. Withey; London/UK
Purpose: The aim of this study was to investigate the feasibility of utilising preoperative contrast-enhanced CT to estimate renal split
function, compared to standard-of-care nuclear medicine studies (DMSA and MAG3) in preoperative planning for nephrectomy.
Methods or Background: Retrospective analysis was conducted on patients with either DMSA- or MAG3-derived estimates of split
renal function (NM-SRF) and a contemporaneous contrast-enhanced CT scan. Patients with hydronephrosis were excluded. Using
reconstruction software with density thresholding (Syngo.via, Siemens Healthineers), renal volumes and mean attenuation values
were calculated. The product of these from each kidney was used to calculate a CT-derived estimated split renal function (CT-SRF).
The estimated percentage renal function from the right kidney from NM-SRF and CT-SRF were compared using Pearson correlation
coefficient. The estimated remaining renal function following nephrectomy using NM-SRF and CT-SRF was compared to actual GFR
values obtained >3 months post-operative. The difference between estimated and true post-operative GFR values were compared
using Student’s T-test.
Results or Findings: n=49 patients (mean age 65.8 ± 9.3 years; 73.4% male) were included in the initial analysis. The Pearson
correlation coefficient between CT-SRF and NM-SRF was 0.835, indicating a very strong correlation (p=<0.0001)
For n=28 patients, post-operative renal function (>3 months following the date of nephrectomy) was available. CT-SRF and NM-SRF
underestimated post-operative GFR by 12.1(±11.9) and 10.5(±12.2) ml/min/1.73 m2, respectively. There was no significant difference
in performance between these tests (p=0.604). Time to calculate CT-SRF was 157±68 seconds.
Conclusion: CT-SRF was a time-efficient and feasible way of estimating split renal function with comparable performance to
standard-of-care NM-SRF in this small retrospective cohort.
Limitations: Larger prospective studies are required to assess whether it is possible to use the available CT data to avoid the
additional radiation exposure, hospital attendance, and cost associated with DMSA and MAG3 testing.
Ethics committee - additional information: Local institutional board review approved this study.
Utility of intra-voxel incoherent motion MRI in assessing bladder cancer muscle invasiveness and histological grade (7
min)
Anup Selvaraju; New Delhi / India
148
Author Block: A. Selvaraju, C. Das; New Delhi/IN
Purpose: Bladder cancer management relies on histopathology to determine tumor grade and muscle invasion. However, trans-
urethral resection of bladder cancer can sometimes lead to understaging. Our research investigates the use of intra-voxel incoherent
motion MRI to improve pre-operative assessment by assessing muscle invasiveness and histological grade.
Methods or Background: We classified 25 cases into low-grade or high-grade and non-muscle invasive or muscle invasive
categories. We conducted image analysis and measured IVIM-derived parameters (apparent diffusion coefficient standard ADC, true
diffusion coefficient D, pseudo-diffusion coefficient D*, and perfusion fraction f). We used Mann-Whitney U-tests for statistical analysis
and generated ROC curves to compare these parameter areas. We considered p-values below 0.05 as statistically significant.
Results or Findings: Out of 25 patients, 13 had muscle-invasive tumours, and 17 had high-grade tumours. Those with muscle-
invasive bladder cancer had notably lower ADC, D, and f values compared to those with non-muscle-invasive bladder cancer.
Specifically, the D value (AUC = 0.90) had a significantly larger area under the ROC curve than the ADC value (AUC = 0.815) (P <
0.05). Similarly, patients with high-grade bladder cancer exhibited significantly lower ADC and D values than those with low-grade
bladder cancer. Here, the D value (AUC = 0.958) also had a significantly larger area under the ROC curve compared to the ADC value
(AUC = 0.875) (P < 0.05).
Conclusion: D values from IVIM analysis effectively differentiate between non-muscle invasive and muscle invasive bladder cancer,
as well as high-grade and low-grade tumours, making them a promising imaging biomarker for assessing bladder cancer's
invasiveness.
Limitations: The sample size is small and the calculated IVIM parameters in our study are specific to the imaging protocol we have
utilised.
Ethics committee - additional information: Approval taken from the Institute ethics committee
Is the R.E.N.A.L. nephrometry score useful for predicting outcomes in percutaneous ablation of renal masses? (7 min)
Mihail Poida; Madrid / Spain
Author Block: M. Poida, M. Paniagua González, M. Camarena Gea, C. Gerlotti Slusnys, P. López Gómez, J. De Luis Yanes, C. Calles
Blanco, E. De Miguel Campos, A. Garcia Perez; Madrid/ES
Purpose: The R.E.N.A.L. nephrometry score was developed to classify renal masses based on their complexity (low, intermediate,
and high) and guide decision-making in cases that could benefit from partial nephrectomy. Although initially designed for this purpose
it might also be useful for predicting outcomes in percutaneous ablation of renal masses-a treatment that has emerged as an
alternative to surgery and has shown promising results in small renal masses. In this study, we aim to examine the cases of renal
masses treated with ablation in our institution.
Methods or Background: We reviewed cross-sectional studies prior to the intervention and retrospectively applied the R.E.N.A.L.
score to a total of 81 renal masses treated at our institution between January 2020 and June 2023, obtaining two groups based on the
complexity of the lesions: low (N=44) and intermediate-high (N=37). Additionally, we analysed all available follow-up studies to date
to identify complications and signs of tumour recurrence.The Fisher’s exact test was used to test for an association of R.E.N.A.L. score
with complications and recurrences.
Results or Findings: The immediate complications (the majority being hematomas at the site of intervention) occurred in 3 cases of
low complexity (6.8%) and in 9 cases of intermediate-high complexity (24.3%). There was a significant association between the
complexity group and the immediate complication rate (p<0.033). Tumoural recurrences were seen in 8 cases of low complexity
(18.1%) and in 3 cases of intermediate-high complexity (8.1%). No significant association has been found (p=0.331).
Conclusion: The R.E.N.A.L. scoring system could be a useful tool for predicting outcomes in percutaneous ablation for renal masses.
Limitations: Due to a lack of cases in the high complexity group, we had to fuse it with the intermediate complexity group for
statistical analysis.
Ethics committee - additional information: No ethical approval was required for this study.
Sensing Ultrasound Localisation Microscopy for the visualisation of glomeruli in living humans (7 min)
Sylvain Bodard; Paris / France
149
Author Block: S. Bodard; Paris/FR
Purpose: Estimation of glomerular function is necessary to diagnose kidney diseases. However, the study of glomeruli in the clinic is
currently done indirectly through urine and blood tests. A recent imaging technique called Ultrasound Localisation Microscopy (ULM)
has appeared. It is based on the ability to record continuous movements of individual microbubbles in the bloodstream. The aim of
this study was to use ULM for glomeruli visualisation. Although ULM improved the resolution of vascular imaging up to tenfold, the
imaging of the smallest vessels had yet to be reported.
Methods or Background: We acquired ultrasound sequences from living humans and then applied filters to divide the data set into
slow-moving and fast-moving microbubbles. We performed a double tracking to highlight and characterise populations of
microbubbles with singular behaviours. We decided to call this technique “sensing ULM” (sULM).
Results or Findings: In this study, we report the observation of microbubbles flowing in the glomeruli in living humans. We present
a set of analysis tools to extract quantitative information from individual microbubbles, such as remanence time or normalised
distance.
Conclusion: As glomeruli play a key role in kidney function, it would be possible that their observation yields a deeper understanding
of the kidney. It could also be a tool to diagnose kidney diseases in patients. More generally, it will bring imaging capabilities closer to
the functional units of organs, which is a key to understand most diseases, such as cancer, diabetes, or kidney failures.
Limitations: The spatial resolution of CEUS, i.e. between 70 and 150 μm, the localisation error of 43±4 μm - that increases with
depth due to divergent ultrasound beam - and the presence of efferent and afferent arterioles in the same track as glomeruli leaded
us to a rough PDW estimation with sULM
Funding for this study: This study was funded by the European Research Council under the European Union Horizon H2020
program (ERC Consolidator grant agreement No 772786-ResolveStroke).
Ethics committee - additional information: The Ethics Committee of the French Society of Radiology approved this study (CERF,
reference number CRM-2203-240).
Multiparametric MRI to evaluate tumour response of locally advanced bladder cancer after radioimmunotherapy (7 min)
Stefan Reischl; Munich / Germany
Author Block: S. Reischl, M. Makowski, A. Sauter; Munich/DE

Purpose: Neoadjuvanttherapy, including radioimmunotherapy (RIT), is considered for locally advanced bladder cancer (BC) patients
that are not candidates for primary surgery. Evaluating tumor response to neoadjuvant therapy is crucial for determining the
feasibility of bladder-sparing strategies. This study aims to evaluate the utility of multiparametric magnetic resonance imaging
(mpMRI) using the Vesical Imaging-Reporting And Data System (VI-RADS) criteria in evaluation of tumour response to RIT.
Methods or Background: This study was designed as secondary analysis of a tertiary-center, prospective study including patients
with locally advance BC treated by neoadjuvant RIT and followed by cystectomy. MpMRI was performed pre- and post-RIT and the
imaging protocol followed the VI-RADS guidelines, including T2-weighted, diffusion-weighted and contrast-enhanced sequences. Semi-
quantiative scoring image analysis was performed by two radiologists. Inter- and intrareader agreement were assessed using
Kendall's τ. Histopathological examination served as ground truth for the evaluation of the diagnostic accuracy of mpMRI in post-RIT
mpMRI.
Results or Findings: Fifteen patients underwent post-RIT mpMRI and subsequent cystectomy. Intrareader agreement (τ=0.74;
p<0.001 and τ=0.53; p<0.015) was superior to interreader agreement (τ=0.38; p<0.081) post-RIT. Of nine patients with residual
tumour only one case of carcinoma in situ (pTis) was scored false-negative, while of five cases with muscle-invasive residual tumour
(≥pT2) only one patient was scored false-negative (VI-RADS≤2) for muscle invasion, in each case by one reader.
Conclusion: This study suggests VI-RADS criteria in mpMRI can be valuable tools for evaluating tumour response in BC patients after
RIT. These findings open the door to expanding bladder-sparing strategies for a broader patient population, pending validation in
larger studies.
Limitations: This study was restricted by its limited cohort size.
Funding for this study: This study was funded by inhouse ressources the Department of Radiology of the TUM.
Ethics committee - additional information: This study was approved by the ethical review board of our institution.
Utilisation of lesion heterogeneity on plain CT imaging using standard deviation and mean HU values in differentiating
between solid/complex cystic and benign cystic renal lesions (7 min)
Balqees Khamees Alabri; Muscat / Oman
150
Author Block: B. K. Alabri, I. Alsalmi; Muscat/OM
Purpose: The main goal of this study is to evaluate the utility of standard deviation and mean Hounsfield Unit (HU) values in plain CT
scans to distinguish between solid/complex cystic lesions and benign cystic lesions in the kidneys. The hypothesis is that there is a
statistically significant difference in HU standard deviation between cystic and solid renal masses.
Methods or Background: A retrospective analysis of 239 cases over the past 13 years was conducted, focusing on histologically
proven renal masses for patients who underwent a plain CT scan of the abdomen prior to biopsy or resection of a renal lesion, with
exclusion criteria applied for cases of angiomyolipomas and lesions smaller than one cm. Attenuation values were assessed through
region of interest (ROI) measurements, ensuring comprehensive coverage while avoiding areas of calcifications and necrosis. Renal
MRI data from the same period were used as a control group for simple cysts.
Results or Findings: The study included 239 cases, dividing renal masses into simple cystic (n=115, 48.1%) and solid (n=124,
51.9%) groups based on CT scans. Average Hounsfield Units (HU) in simple cysts were much lower (mean=7.10) than in solid lesions
(Mean=37.35) with a p-value <0.001. Standard deviation (SD) values followed a similar pattern, being lower in simple cysts
(Mean=12.85) compared to solid lesions (Mean=16.61) with a p-value <0.001. HU was more effective than SD in distinguishing them.
A scoring system combining HU and SD effectively differentiated between cystic and solid renal lesions, showing potential clinical use.
Conclusion: Combining Hounsfield unit and standard deviation values improves renal lesion characterisation on CT scans, offering
potential for more precise clinical decisions and patient management.
Limitations: Single-centre retrospective nature, raising possible issues with regard to generalisation of results and selection bias and
sample size limit this study.
Ethics committee - additional information: The study is retrospective and hence, no ethical approval was sought.
151
E³ 420 - The future of oncologic and vascular interventional radiology
Categories: Interventional Oncologic Radiology, Interventional Radiology

CME Credits: 1.5
The session gives an overview of the main novelties expected to be introduced and soon used in vascular and extra-vascular
interventional radiology. The session intends to familiarise the audience with new technologies and clinical advancements that could
be implemented in interventional radiology departments.
Moderator:
Christoph Binkert; Winterthur / Switzerland

Evolution and perspective in oncological IR (18 min)

Laura Crocetti; Pisa / Italy
Multimodality and combined treatment in IR (18 min)

Roberto Luigi Cazzato; Stasbourg / France
Perspective of vascular IR (18 min)

Heather Moriarty; Dublin / Ireland
Stereotactic and robotic interventions (18 min)

Brad J. Wood; Bethesda / United States
Panel discussion: How IR departments should integrate all these novelties (13 min)
152
CUBE 4 - Utility of augmented reality with computed tomography fusion in chronic occlusive femoral
interventions

The "EFRS @ the Cube" sessions focus on current radiography topics in interventional radiology.
Utility of augmented reality with computed tomography fusion in chronic occlusive femoral interventions (30 min)
Irene Nieri; Paris / France
1. To define the role of augmented reality (AR) with CT fusion in chronic occlusive femoral interventions.
2. To explain the training and governance arrangements for radiographers when using AR.
3. To discuss the future role of AR and CT fusion in IR procedures.
153
E³ 421a - Current issues in chest radiology
Categories: Chest, General Radiology, Imaging Methods

CME Credits: 1.5
Differences and overlap between pulmonary infections and non-infectious types of inflammations in HRCT (45 min)
1. To describe the typical HRCT features of pulmonary infections and non-infectious inflammations.
2. To understand possible overlaps and important differences between the two groups of diseases.
Imaging of thoracic vascular malformations (45 min)

Gianluca Milanese; Parma / Italy
1. To learn the most common radiological findings in thoracic vascular malformations.

2. To emphasise the importance of HRCT for a differential diagnosis.
154
Meets 4 - Radiology and Nuclear Medicine: opportunities for joint research
Categories: Imaging Methods, Interventional Radiology, Multidisciplinary, Research

CME Credits: 1
Moderators:
Rudi Dierckx; Groningen / Netherlands
Rudi Dierckx; Groningen / Netherlands
The beautiful world of radiopharmaceuticals (9 min)

Verena Pichler; Vienna / Austria
1. To name and identify key radiopharmaceuticals in nuclear medicine.

2. To describe the revolutionary concept of theranostic radiopharmaceuticals.
3. To explore nuclear imaging techniques for complementary diagnostics.
Theranostics using MRI (9 min)
Theranostics: beyond imaging (9 min)

Ken Herrmann; Essen / Germany
1. To name and identify two focus areas of radioligand therapy in oncology.

2. To list three innovative theranostic targets.
3. To describe the challenges and opportunities of theranostics.
Theranostics in interventional radiology (9 min)

Gilles Soulez; Montreal / Canada
1. To understand the current role of nuclear medicine to improve targeting during cTACE and TARE procedures.
2. To discuss the potential role of nanoparticles and liposomes in future theranostic strategies.
3. To discuss the potential role of magnetic nanoparticles and magnetotactic bacteria to improve MR-guided
targeting.
Nuclear medicine: interdisciplinarity and multidisciplinarity (9 min)

Wim J.G. Oyen; Arnhem / Netherlands
1. To understand the dynamics of a nuclear medicine department in the evolving prospects of theranostics.
2. To understand the role of nuclear medicine in the multidisciplinary teams in the hospital.
3. To understand the specific training aspects to successfully run a nuclear medicine service.
155
E³ 421b - Neuroradiology: an update
Categories: Imaging Methods, Neuro, Oncologic Imaging, Paediatric, Vascular

CME Credits: 1.5
Paediatric neuroimaging (45 min)

Daniela Prayer; Vienna / Austria
1. To understand the advantages and disadvantages of pre-, and postnatal neuroimaging.

2. To become familiar with MR imaging in uncooperative patients.
3. To get information about age-dependent MR protocols.
4. To understand prognostic information derived from imaging studies.
Oncologic neuroimaging (45 min)

Julia Johanna Furtner; Vienna / Austria
1. To understand the best techniques in the evaluation of CNS primary tumours.
156
RPS 401 - Diagnosis and response assessment in colorectal cancer
Categories: GI Tract, Imaging Methods, Nuclear Medicine, Oncologic Imaging

CME Credits: 1.5
Moderator:
Sofia Gourtsoyianni; Athens / Greece
Value of Iodine uptake derived from DECT in differentiating malignant from benign colorectal wall thickening (7 min)
Lucien Widmer; Villars-Sur-Glâne / Switzerland
Author Block: L. Widmer1, F. Cherbanyk1, H. Thoeny1, S. Malekzadehlashkariani2; 1Fribourg/CH, 2Sion/CH

Purpose: The aim of this study was to assess the performance of iodine uptake derived from Dual-energy computed tomography
(DECT) in differentiating malignant from benign colorectal wall thickening.
Methods or Background: DECT is widely used to characterise and differentiate tumours. However, data regarding its diagnostic
performance for the characterisation of colorectal tumours are limited. In this single-centre IRB-approved study, 50 consecutive
patients with both histopathology-proven colorectal adenocarcinoma and benign colon wall pseudo-thickening were prospectively
enrolled. Two radiologists quantitatively analysed the iodine uptake of the tumors and iodine ratios as well as density measurements
in Hounsfield units (HU) from portal venous phase DECT. Paired t-test and predictive power score were used to compare variables of
malignant and benign wall thickening. A priori statistical power analysis using G*Power estimated that 45 patients were necessary to
reach a 95% statistical power for a one-sided paired t-test with a medium effect size and an alpha of 0.05.
Results or Findings: Significantly greater iodine uptakes were found in tumours than in benign wall pseudo-thickenings (medium to
strong predictive power, delta = 0.13, t(49) = 7.3, p < 0.001, pps = 0.64). There were no equivalence between the tumours and the
benign wall pseudo-thickenings (delta = 0.17, t(49) = 6.4, p = 1), but significantly greater HU values in tumours compared to benign
wall pseudo-thickenings (strong predictive power, t(49) = 15.8, p < 0.001, pps = 0.79).
Conclusion: DECT demonstrated a significant difference in HU and iodine uptake in malignant colorectal tumours in comparison to
benign colorectal wall pseudo-thickening. However, this study surprisingly shows that HU is a variable of better value than iodine
uptake in discriminating a tumour from a non-tumour pseudo-thickening.
Limitations: Single-centre study and small sample size; adenocarcinoma as only histopathological subtype; manual delineation of
ROIs limit the scope of this study.
Ethics committee - additional information: CER-VD, swissethics approved this study.
Performance of CT in the locoregional staging of non-rectal colon cancer: detailed radiology-pathology correlation with
a special emphasis on tumour deposits and extramural venous invasion (7 min)
Duygu Karahacıoğlu; Istanbul / Turkey
157
Author Block: D. Karahacıoğlu, O. C. Taskin, A. Armutlu, B. Saka, R. Esmer, D. Bugra, E. Balik, V. Adsay, B. Gurses; Istanbul/TR
Purpose: The purpose of this study was to investigate the performance of CT in detecting the diagnostic and prognostic
characteristics (established in rectal adenocarcinomas) in non-rectal colon malignancies via close radiologic and pathologic
correlation in order to improve patient care.
Methods or Background: CT and pathology data from 137 resected colonic adenocarcinoma cases were reviewed for prognostic
parameters by dedicated radiologists and pathologists whose primary practice is GI cancers. Discrepant cases were re-evaluated
together with correlation of radiologic-gross photographs and microscopic findings.
Results or Findings: For T-stage, overall CT tended to understage (in 12.4% of cases), performed better in left versus right colon,
with the lowest performance observed in transverse colon, and the highest rate of misstaging at sigmoid. The sensitivity, specificity,
and accuracy of CT in the detection of EMVI were 58,5 %, 82,1 %, 73 % (which was slightly better for high-load EMVI compared to low-
load). These figures for TDs were 57,9 %, 92,4 %, 87,6 and for LNs, 44.7%, 72.7%, and 63.5%. In 13 patients with discrepancies, on-
site correlation resulted in modification of the initial radiology or pathology findings.
Conclusion: The criteria increasingly employed for rectal cancers and were elucidated by careful pathologic radiologic correlation in
our studies demonstrate that CT can be more effectively used in prognostication of colon cancers. CT appears to be more applicable
in left colon, although mis-stagings are most common in the sigmoid, for which more non-committal reporting may be necessary in
uncertain cases. CT is highly effective in detection of TDs and EMVI (especially high-load), but less so for LNs. TD and EMVI should be
considered for incorporation into routine reporting CT proformas for colon cancer.
Limitations: The major limitation is the absence of specimen radiology, which prevents the correlation of LN/TD on a case-by-case
basis.
Funding for this study: No funding was recieved for this study.
Ethics committee - additional information: This study was approved by the IRB number: 2022.092.IRB1.038
Preoperative evaluating peritoneal cancer index in peritoneal metastatic colon cancer patients using CT-based deep-
learning model (7 min)
Shaoting Zhang; Jinan / China
Author Block: S. Zhang1, F. Shen2, C. Shao2; 1Jinan/CN, 2Shanghai/CN

Purpose: This study aimed to build a convolution neural network to automatically segment the colon cancer lesion and evaluate
peritoneal cancer index (PCI) in peritoneal metastasis (PM) cases based on preoperative CT scans.
Methods or Background: A total of 177 consecutive colon cancer patients with preoperative abdominopelvic CT scans and PM
administered exploratory laparotomy were retrospectively analysed. Surgical PCIs (sPCIs) were confirmed by the surgery team, and
CT-PCI scores were assessed by radiologists. Totally 90 cases were assigned to training set. Then, 87 cases were enrolled as test set.
Initially, we utilised nn-UNet to build an automatic segmentation model based on CT scans at portal venous phase. Secondly, we
constructed deep learning model (DL-Model) to assess sPCIs. The performances in distinguishing between low-sPCI (≤ 20) and high-
sPCI (> 20) cases were also assessed by receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA).
Results or Findings: The automatic segmentation model showed excellent performance. The DL-Model had higher classification
performance than the subjective evaluation in all datasets (AUCs were 0.867 and 0.830 in training and test sets, vs 0.767 and 0.694,
respectively). The usefulness of the proposed DL-Model was confirmed by DCA.
Conclusion: This study showed that the DL-Model based on abdominopelvic CT scans were useful for preoperative detecting PCI in
colon cancer patients with PM, especially showed increased diagnostic performance than subjective CT-PCI.
Limitations: It is unclear whether our findings would apply to the lesion with neoadjuvant treatment, which needs further research.
Ethics committee - additional information: Committee on Medical Research Ethics of Changhai Hospital approved this study.
Early regression index (ERI) on MR images predicts complete responders after neo-adjuvant chemo-radiotherapy (7
min)
Martina Mori; Milan / Italy
158
Author Block: C. Fiorino, D. Palumbo, M. Mori, G. Palazzo, A. Pellegrini, A. Damascelli, S. Steidler, A. Del Vecchio, F. De Cobelli;
Milan/IT
Purpose: The early regression index (ERI) resulted to predict treatment response in rectal cancer patients. Aim of the current study
was to prospectively assess tumour response to neo-adjuvant chemo-radiotherapy (nCRT) of locally advanced oesophageal cancer
using ERI, based on MRI.
Methods or Background: From January 2020 to May 2023, 30 patients with oesophageal cancer were enrolled in a single arm
prospective study (ESCAPE). Patients were scanned with PET-MRI at three time points: (i) before nCRT (tpre); (ii) mid-radiotherapy,
tmid; (iii) after nCRT, 8-12 weeks before surgery (tpost); nCRT delivered 41.4Gy/23fr with concurrent carboplatin and paclitaxel. For
patients that skipped surgery, complete clinical response (cCR) was assessed if patients showed no local relapse after one year;
patients with pathological complete response (pCR) or with cCR were considered as complete responders (pCR+cCR). Segmentations
of GTV volumes were performed by two observers (Vpre, Vmid, Vpost) on T2w MRI: ERI and other volume regression parameters at
tmid and tpost were tested as predictors of pCR+cCR.
Results or Findings: At the time of the analysis complete data of 25 patients were available: 3/25 with complete response at
imaging refused surgery and 2/3 were cCR; in total, 10/25 patients showed pCR+cCR (pCR=8/22). Both ERImid and ERIpost identified
pCR+cCR patients, with ERImid showing better performance (AUC:0.78, p=0.014). A logistic model combining ERImid and Vpre
improved performances (AUC:0.93, p<0.0001). Inter-observer variability in contouring GTV did not affect the results.
Conclusion: Despite the limited numbers, interim analysis of ESCAPE study suggests ERI as a robust predictor of complete response
after nCRT for esophageal cancer. Further validation on larger populations is warranted.
Limitations: Interruptions due to Covid-19 pandemia caused the downsizing of the sample size.
Funding for this study: The study was supported by AIRC (Italian Association for Cancer Research) under Investigator Grant – IG
2019 – ID 23015 project.
Predicting response to chemotherapy in patients with advanced colon cancer based on body composition (7 min)
Damiano Grasso; Rome / Italy
Author Block: D. Grasso, F. Marruzzo, L. Cosma, F. Iafrate; Rome/IT

Purpose: The purpose of this study was to investigate the association between body composition parameters and response in of
adjuvant therapy in patients with metastatic colon cancer.
Methods or Background: Patients with metastatic colon cancer Stage IV treated with neoadjuvant chemotherapy were enrolled;
incomplete scan set was adopted as an exclusion criterion. Body composition was assessed via software AI Quantib. Visceral Adipose
Tissue (VAT), Subcutaneous Adipose Tissue (SAT) and Skeletal Muscle Index (SMI) were measured before starting adjuvant therapy.
Measurements were performed on a computed tomography-derived abdominal image at third lumbar vertebra (L3) level by an
automatic dedicated software. Probit model and Linear regression analyses were performed.
Results or Findings: One hundred and four patients were included in the evaluation, (mean age 55 ± 26 years), 12 patients were
sarcopaenic, 47 were obese, while 45 patients were neither sarcopaenic nor obese. Statistical analyses showed that good response to
therapy was correlated to higher SMI values (p < 0.001).
Conclusion: Our results showed an interesting correlation between sarcopaenia and progression of disease and demonstrated that
SMI can positively influence the response to adjuvant chemotherapy.
Limitations: The few limitations of our study are linked to the fact that it is a retrospective, monocentric analysis with a limited
number of patients.
Ethics committee - additional information: No ethical approval was sought for this study.
Role of IVIM MRI in response assessment in colorectal carcinoma (7 min)

Stuti Chandola; New Delhi / India
159
Author Block: C. J. Das, S. Chandola, A. Soni; New Delhi/IN
Purpose: The aim of this study was to. assess the usefulness and performance of Diffusion-weighted imaging (DWI) with IVIM- DKI for
assessing post chemotherapy (CT)/Chemo and Radiotherapy (CTRT) response in colorectal carcinoma and compare with PET/CT
parameters
Methods or Background: Forty patients with rectosigmoid cancer underwent baseline staging multiparametric MRI and 18-FDG
PET/CT and follow-up with both scans post chemo radiotherapy. Quantitative diffusion, IVIM and DKI parameters, viz. apparent (ADC)
and molecular (D) diffusion coefficient, perfusion coefficient (f), and kurtosis (K) were measured from non-necrotic areas and semi-
quantitative PET parameters including SUV max, SUV ratio, Metabolic tumour volume (MTV), and total lesion glycolysis (TLG) were
derived from the PET/CT images; and correlated with the patient’s response keeping RECIST 1.1 criteria as the gold standard.
Results or Findings: A statistically significant increase in D and ADC with a decline in K was noted after therapy in all patients. No
significant difference was seen among the percentage change in the parameters observed post-therapy among the responders and
non-responders. Both the responders as well as non-responders depicted a statistically significant increase in D and ADC, and a
significant decline in K values post-therapy. Among 17 patients with follow-up PET/CT imaging, a significant decline in all parameters
of the primary lesion was seen post-therapy. Responders (n=12) showed a significant decline in all parameters from baseline after
therapy, whereas non-responders did not show any such decline. Post-therapy MTV, followed by TLG were found to have strongest
correlation with a positive response, with AUCs of 0.933 and 0.900 on receiver operator curves.
Conclusion: 18-FDG PET/CT is the more accurate single modality for assessing both response as well as tumour burden post therapy,
while ADC and D derived from DWI and IVIM respectively are useful adjuncts for assessment of response.
Limitations: No information provided by the submitted.
Ethics committee - additional information: The ethics committee notification can be found under the number ID - IEC
PG-621/25.11.20, RT-06/23.12.2020.
Predictive value of modified MRI‑based split scar sign (mrSSS) score for pathological complete response after
neoadjuvant chemoradiotherapy for patients with rectal cancer (7 min)
Fangying Chen; Shanghai / China
Author Block: F. Chen, F. Shen, C. Shao; Shanghai/CN

Purpose: The aim of this study was to measure the diagnostic performance of modified MRI-based split scar sign (mrSSS) score for
the prediction of pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) for patients with rectal cancer.
Methods or Background: The modified MRI-based split scar sign (mrSSS) score, which consists of T2-weighted images (T2WI)-based
score and diffusion-weighted images (DWI)-based score. The sensitivity, specificity, and accuracy of modified mrSSS score,
endoscopic gross type, and MRI-based tumour regression grading (mrTRG) score, in the prediction of pCR, were compared. The
prognostic value of the modified mrSSS score was also studied.
Results or Findings: A total of 189 patients were included in the study. The Kendall’s coefficient of interobserver concordance of
modified mrSSS score, T2WI -based score, and DWI-based score were 0.899, 0.890, and 0.789 respectively. And the maximum and
minimum k value of the modified mrSSS score was 0.797 (0.742–0.853) and 0.562 (0.490–0.634). The sensitivity, specificity, and
accuracy of prediction of pCR were 0.66, 0.97, and 0.90 for modified mrSSS score; 0.37, 0.89, and 0.78 for endoscopic gross type
(scar); and 0.24, 0.92, and 0.77 for mrTRG score (mrTRG=1). The modified mrSSS score had significantly higher sensitivity than the
endoscopic gross type and the mrTRG score in predicting pCR. Patients with lower modified mrSSS scores had significantly longer
disease-free survival (P<0.05).
Conclusion: The modified mrSSS score showed satisfactory interobserver agreement and higher sensitivity in predicting pCR
after nCRT in patients with rectal cancer. The modified mrSSS score is also a predictor of disease-free survival.
Limitations: The major limitations of this study were: (1) This was a retrospective and single-centre study with relatively short follow-
up. (2) We did not perform an analysis of lymph node response. (3) This modified mrSSS score system in this study lacks enough
sensitivity.
Ethics committee - additional information: Committee on Medical Research Ethics of Changhai Hospital approved this study.
Can MRI accurately assess the TN staging of rectal cancer patients with local regrowth during watchful waiting? (7 min)
Xinde Ou; Amsterdam / Netherlands
160
Author Block: X. Ou, B. M. Geubels, D. M. J. Lambregts, B. Grotenhuis, G. L. Beets, R. G. H. Beets-Tan, M. Maas; Amsterdam/NL
Purpose: Local regrowth during watchful waiting occurs in +/-25% of patients. Salvage surgery is possible in the vast majority. MRI is
used for regrowth staging, but its accuracy is unknown. The aim is to evaluate the accuracy of MRI for preoperative TN-staging of
rectal cancer patients with local regrowth during organ preservation.
Methods or Background: Rectal cancer patients identified with local regrowth during watchful waiting all underwent salvage
surgery (TME or local excision). Patients who underwent local excision as salvage surgery were excluded from N-status analyses if
their follow-up after local excision was < 24 months. An expert radiologist scored the ymrTN-stage on pre-surgery MRIs. The ymrT and
ymrN-stages were compared with final ypT and ypN-stages at histopathology. For analysis, T0, T1 and T2 were combined as T0-2 and
T3 and T4 combined as T3-4; N1 and N2 were combined as N+. Diagnostic performance (accuracy, sensitivity, specificity) were
calculated.
Results or Findings: Among these patients, 26 had ypT-stage available (17 ypT0-2, 9 ypT3-4) and 22 had ypN-stage available (15
ypN0, 7ypN+). The MRI-to-surgery interval ranged from 2 to 19 weeks (median: 7 weeks). 20 had luminal regrowth, one had nodal
regrowth, and six had both. Accuracy, sensitivity and specificity for ypT-staging were 81%, 76% and 89%. Overstaging occurred in
15%. Accuracy, sensitivity and specificity for ypN-staging were 82%, 80% and 82%. Understaging occurred in 14%.
Conclusion: MRI can accurately stage local regrowth during watchful waiting before surgery. In ypT-staging, overstaging is common,
while ypN-staging is more prone to understaging.
Limitations: Multiple readers need to be included to assess the interobserver agreement and evaluate the impact of radiologists'
experience on ymrTN-staging accuracy. Sample size is rather small. Selection bias needs to be considered for the highly selected
group.
The involvement of mesorectal fascia by tumour deposits and extramural venous invasion predicts poor overall
survival in locally advanced rectal cancer (7 min)
Yaru Feng; Shanghai / China
Author Block: Y. Feng, T. Tong; Shanghai/CN

Purpose: Mesorectal fascia (MRF) involvement serves as an adverse prognostic indicator for locally advanced rectal cancer (LARC).
However, prognoses among MRF+ patients vary. This study aims to further classify MRF involvement in MRF+ cases and investigate
its prognostic implications.
Methods or Background: We conducted a retrospective analysis of data from LARC patients who were evaluated MRF+ and
underwent surgery after neoadjuvant chemoradiotherapy. Patients were categorised into four groups based on MRF involvement:
MRF+ (Tumour) vs MRF+ (Non-tumour), MRF+ (Extramural venous invasion; EMVI) vs MRF+ (Non-EMVI), MRF+ (lymph node; LN) vs
MRF+ (Non-LN), and MRF+ (Tumour deposits; TDs) vs MRF+ (Non-TDs). Survival curves were generated using the Kaplan-Meier (K-M)
method, and differences in overall survival (OS) were assessed using the log-rank test. Prognostic factors were evaluated through
multivariate Cox proportional hazards models.
Results or Findings: A total of 400 MRF+ LARC patients were included in the study. The occurrence of MRF+ (Tumour), MRF+ (TDs),
and MRF+ (EMVI) was higher in patients aged ≥60 years. In K-M survival analysis, both MRF+ (TDs) and MRF+ (EMVI) cases were
associated with significantly worse OS compared to MRF+(non-TDs) and MRF+ (non-EMVI). Conversely, there were no significant
differences in OS between MRF+ (Tumour) and MRF+ (LN) cases and MRF+ (Non-tumour) and MRF+ (non-LN). In the multivariate COX
regression analysis, MRF+ (TDs) demonstrated a significant association with OS (HR 1.744; 95% CI, 1.064-2.860). Similarly, MRF+
(EMVI) displayed a significant correlation OS (HR 1.886; 95% CI, 1.076-3.306). Conversely, there were no significant associations
observed between MRF+ (Tumour) or MRF+ (LN) and OS.
Conclusion: MRF+ (TDs) and MRF+ (EMVI) independently indicated higher risks for adverse prognosis in MRF+ LARC patients,
providing valuable insights for treatment and follow-up strategies.
Limitations: This was a retrospective and single centre study.
Ethics committee - additional information: This research was approved ethically.
Performance of dual-layer spectrum CT virtual monoenergetic images to assess early rectal adenocarcinoma T-stage:
comparison with MR (7 min)
Ziqi Jia; Guangzhou / China
161
Author Block: Z. Jia, W. j. Yuan, X. M. Liu, H. Zhang, J. Dai, L. Guo, W. C. Chen, X. Liu; Guang Zhou/CN
Purpose: Dual-layer spectrum CT (DLSCT) can provide information about material decomposition and improve lesion visualization
that may be useful to assess preoperative T-stage in early rectal carcinoma (ERC). This study aimed to investigate the image quality
of virtual monoenergetic images (VMI) and conventional polyenergetic image (PEI) from DLSCT, and compare the performance with
MR in assessing preoperative T-stage for early rectal adenocarcinoma (ERA).
Methods or Background: This retrospective study included 67 ERA patients (mean age 62±11.1 years) who underwent DLSCT and
MR examination. VMI 40-200 keV and poly energetic image (PEI) were reconstructed. The image noise, signal-to-noise ratio (SNR),
contrast-to-noise ratio (CNR), and tumour contrast of different energy levels were calculated and compared, respectively. Two
radiologists independently assess the image quality of the VMIs and PEI using five-point scales. The diagnostic accuracies of DLSCT
and HR-MRI for ERA T-staging were evaluated and compared.
Results or Findings: The maximum noise was observed at VMI 40 keV, and noise at VMI 40-200keV in the arterial and venous
phases showed no statistically significant difference (all p>0.05). The highest SNR and CNR were obtained at VMI 40 keV, significantly
greater than other energy levels and PEI (all p<0.05). Tumour contrast was more evident than PEI at 40-100keV in the arterial phase
and at 40keV in the venous phase (all p <0.05). When compared with PEI, VMI 40keV yielded the highest scores for overall image
quality, tumour visibility, and tumor margin delineation, especially in the venous phase (p<0.05). The overall diagnostic accuracy of
DLSCT and HR-MRI for T-stage was 65.67% and 71.74% and showed no statistically significant difference (p>0.05).
Conclusion: VMI 40keV provides the best image quality and improves the diagnostic confidence for ERA staging.
Funding for this study: This work was supported by Guangzhou Basic and Applied Basic Research Foundation 2023A03J024,
National Nature Science of Foundation of China [Grant No.82202259] and the Youth Talent Project of The Second Affiliated Hospital of
Guangzhou University of Chinese Medicine(ZY2022YL05)
was obtained from all participants
162
OF 4T - Radiology Trainees Forum (RTF): supporting young radiologists
Categories: Education, General Radiology, Management/Leadership, Professional Issues, Students

CME Credits: 1
Moderators:
Saif Afat; Tübingen / Germany
Judith Herrmann; Tübingen / Germany

Radiology Trainees Forum: who we are and our aims (10 min)
1. To describe the radiology trainees' forum and aims.

2. To discuss how the advantages of being involved in the forum for young radiologists.
Radiology Trainees Forum: opportunities to collaborate (20 min)

Viktoriia Pozdniakova; London / United Kingdom
Isabel Molwitz; Hamburg / Germany
1. To describe the collaborative activities of the radiology trainees' forum.

2. To discuss how forum activities can shape the future of national societies.
How trainees can influence next-generation radiology (10 min)

Michail Klontzas; Heraklion / Greece
1. To describe research within the radiology trainee forum.

2. To discuss how young radiologists can shape future practice.
Open forum discussion: How to make the most of the radiology trainees' forum? (15 min)
163
EU 4 - SAMIRA: a European Commission initiative to strengthen quality and safety in imaging
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Professional Issues

CME Credits: 1.5
Moderator:
Georgi Simeonov; Luxembourg / Luxembourg
Chairperson's introduction and SAMIRA overview (8 min)

Update on EU-REST study on workforce and staffing issues (8 min)

1. To learn about the aims of the EU-REST project.

2. To appreciate problems in education and staffing guidelines of the radiology, radiation oncology and nuclear medicine workforce in
Europe.
3. To understand variability in EU-27 countries and how the project may influence the future of the workforce in the EU.
Update on MARLIN: radiation incident reporting (8 min)

CARLOS Prieto PRIETO MARTIN; Madrid / Spain
1. To learn that errors can be used to improve processes, systems, and equipment and then reduce the probability of repeating them.
2. To appreciate that there can and should be different incident learning systems (ILS) with different objectives and organisational
structures; ILS at the local (departmental or hospital) level, ILS of professional societies and ILS of competent authorities.
3. To understand that although there are various tools for safety management, the use of ILS is a fundamental element that must be
effectively implemented to achieve increasingly safer procedures in the use of radiation in medicine.
Update on the BSS equipment study, Nucadvisor (15 min)

Jonas Seth Andersson; Umeå / Sweden
1. To learn about the background and aims of the BSS equipment study.
2. To appreciate the challenges and possibilities in patient exposure monitoring in radiology, interventional specialities, nuclear
medicine, and radiotherapy.
3. To understand the potential difficulties with patient exposure monitoring due to the heterogeneity in Europe and what may be
achieved with legislation, standardisation, and dose monitoring systems.
Guidelines and recommendations for therapeutic radiopharmaceuticals from the SIMPLERAD project (8 min)
Michael Lassmann; Würzburg / Germany
1. To learn about the implementation of Euratom and the EU legal bases concerning the therapeutic uses of radiopharmaceuticals.
2. To appreciate the links and interdependencies between the European pharmaceutical legislations and Euratom radiation protection
requirements.
3. To understand potential barriers to implementation, ways to advance a coherent implementation of these requirements for the
therapeutic use of radiopharmaceuticals, and that there are quality and safety issues related to the current use and introduction of
novel therapeutic radiopharmaceuticals into clinical practice, including requirements for dosimetry, the role of MPEs, the release of
patients from the hospital, and management of radioactive waste.
164
Results of the EURAMED rocc'n'roll project and next steps in implementing the proposed strategic research agenda (15
min)
Christoph Hoeschen; Magdeburg / Germany
1. To learn about the development of a European strategic research agenda and corresponding roadmap on medical applications of
ionising radiation.
2. To appreciate the way that the documents have been developed and that these shall be seen as living documents, which depend
on the community’s input.
3. To understand how these documents can be used for fostering individual research efforts and to lobby for research on medical
applications of ionising radiation.
Forthcoming opportunities under SAMIRA (8 min)

1. To learn about the SAMIRA action plan and its three pillars on quality and safety of medical radiation applications, security of supply
of medical radioisotopes and innovation and development in these areas.
2. To appreciate the progress made in implementing the SAMIRA action plan, focusing on the quality and safety of radiology.
3. To understand the opportunities offered by different EU programmes to support radiology's quality and safety.
Panel discussion: How to improve the implementation of SAMIRA studies and guidance through professional
initiatives? (20 min)
165
PC 4 - Management of localised prostate cancer
CME Credits: 1.5
Moderator:
Harriet Thoeny; Bern / Switzerland

Harriet Thoeny; Bern / Switzerland
Radiologist (20 min)

1. To explain what kind of information the radiologist needs to provide to the MDT.
2. To illustrate the impact of this information on the correct diagnosis, staging and treatment stratification.
3. To stress the importance of MDT meetings in radiological quality assurance.
Urologist (20 min)

Massimo Valerio; Lausanne / Switzerland
1. To understand how clinical, biological, radiological and pathological information influence clinical decision-making and informed
consent.
2. To explore current treatment options and modern algorithms in patients with low to high-risk prostate cancer.
Pathologist (20 min)

Eva Compérat; Vienna / Austria
1. To be able to correctly understand the information given by the pathologist, in order to decide the best treatment according to the
pathology report.
2. To understand the complexity of the molecular underpinnings in the setting of prostate cancer, and to understand the correlation
between molecular changes and prostate cancer histology.
Patient perspective on prostate cancer and screening for prostate cancer (15 min)
Erik Briers; Brussels / Belgium
1. To understand that early detection through risk-based population screening will save lives and prevent unnecessary treatments.
2. To explain that in low-risk prostate cancer, "active surveillance and watchful waiting" is the only option according to the EAU
guidelines, but that this treatment needs a good explanation to the patient.
3. To learn that from a patient's perspective, learning of side effects is equally important as learning what the clinician can do when
they happen to help the patient.
Panel discussion: How to offer the best treatment to the individualised patient? (10 min)
166
SF 4 - The green hospital: enhancing sustainability
Categories: Education, Management/Leadership, Multidisciplinary, Radiographers, Students

CME Credits: 1.5
Moderators:
Jose Guilherme Couto; Msida / Malta
Philipp Brantner; Rheinfelden / Switzerland

Philipp Brantner; Rheinfelden / Switzerland
Tips for making more effective use of departmental resources (16 min)
1. To be familiar with the evidence relating to the relative contribution of departmental resources to the overall environmental impact,
focussing on energy consumption and single-use items.
2. To understand the costs and savings of green practice.
3. To understand the environmental impact of streamlining clinical care.
Applicability of current green practices in healthcare facilities (16 min)

Ana Luísa Ferreira Soares; Porto / Portugal
1. To define the concepts of "circular economy" and "carbon footprint".

2. To identify the different dimensions across which circular economy can be applied to healthcare.
3. To provide examples of sustainable practices that can be applied by healthcare professionals.
Safe and sustainable use of contrast media (16 min)

Moreno Zanardo; Limbiate / Italy
1. To understand the concept of green radiology and the importance of enhancing sustainability in radiology facilities.
2. To learn about the safe and sustainable use of contrast media.
3. To gain knowledge on strategies for promoting contrast media sustainability.
Reducing waste in imaging (16 min)

Elin Kjelle; Gjøvik / Norway
1. To name typical imaging examinations of low-value to patients.

2. To discuss resources wasted on unnecessary imaging examinations.
3. To discuss measures to effectively reduce the use of unnecessary imaging examinations.
Panel discussion: How do we instil green skills in future healthcare professionals (21 min)
167
OF 4R - A holistic approach to osteoporosis
Categories: Imaging Methods, Management/Leadership, Professional Issues, Radiographers

CME Credits: 1
This session delves into the multifaceted landscape of osteoporosis care, highlighting the pivotal role of radiographers in the
assessment and management of this prevalent bone disorder. This session comprises three insightful talks that collectively contribute
to a comprehensive understanding of osteoporosis, emphasising the significance of early detection and prevention. Furthermore, the
session will highlight the pivotal role radiographers can play in early detection, accurate reporting, and proactive fracture prevention.
Attendees of this session will gain a profound understanding of how radiographers may further develop and/or extend their role and
make a difference in the fight against osteoporosis, ultimately contributing to better bone health and enhanced patient outcomes.
Moderator:
Maria Cauchi; Birkirkara / Malta

Maria Cauchi; Birkirkara / Malta
How radiographers can make a difference in DEXA patient pathways (16 min)
Eilish Mcdermott; Dublin / Ireland
Optimised reporting for DEXA scanning: the vertebral fracture identification toolkit (16 min)
Jill Griffin; Bath / United Kingdom
Radiographers as fracture prevention practitioners:a hybrid role model for the future? (16 min)
Rosemary James; Somerset / United Kingdom
168
RC 406 - Imaging-guided theranostics: current clinical practice and recent developments
Categories: Hybrid Imaging, Interventional Oncologic Radiology, Molecular Imaging, Nuclear Medicine, Oncologic Imaging
ETC Level: LEVEL II
CME Credits: 1
Moderator:
Jose Luis Vercher Conejero; Barcelona / Spain

Prostate-specific membrane antigen (PSMA)-based theranostics: prostate cancer and beyond (15 min)
Tobias Maurer; Hamburg / Germany
1. To become familiar with PSMA-based radiopharmaceuticals.

2. To become familiar with clinical indications for PSMA-targeted radioligand therapy.
3. To understand the role of hybrid imaging in therapy planning and response assessment.
Somatostatin receptor (SSR)-based theranostics: neuroendocrine tumours and beyond (15 min)
Irene Virgolini; Innsbruck / Austria
1. To become familiar with SSR-based radiopharmaceuticals.

2. To become familiar with clinical indications for SSR-based peptide receptor radionuclide therapy.
3. To understand the role of hybrid imaging in therapy planning and response assessment.
Recent developments in the field of imaging-guided theranostics (15 min)

Désirée Deandreis; Villejuif / France
1. To understand basic concepts of imaging-guided theranostics.

2. To get insight into current clinical research directions.
3. To discuss future applications of imaging-guided theranostics.
Panel discussion: The role of radiologists in oncologic therapy (10 min)
169
E³ 418 - Coronary heart disease: radiology is key
Categories: Cardiac, Emergency Imaging, Evidence-Based Imaging, Imaging Methods

CME Credits: 1.5
At the end of this session, participants will be able to determine which imaging modality is preferred for the specific patient. They
should know how to assess the important imaging features in coronary heart disease, whether stable or acute, and they should learn
to describe the imaging findings of (M)INOCA and coronary dissection.
Moderator:
Konstantin Nikolaou; Tuebingen / Germany

The patient with stable chest pain (24 min)

Michelle Claire Williams; Edinburgh / United Kingdom
The patient with acute chest pain (24 min)

Christopher L. Schlett; Freiburg im Breisgau / Germany
(M)INOCA and coronary dissection (24 min)

Maja Hrabak Paar; Zagreb / Croatia
Panel discussion: How do we ensure sufficient coverage? 24/7? (16 min)
170
E³ 426 - Imaging of the brachial plexus and nerves around the shoulder
Categories: Musculoskeletal
CME Credits: 1
Moderator:
Paolo Simoni; Bruxelles / Belgium

Paolo Simoni; Bruxelles / Belgium
Imaging of brachial plexus injury (20 min)

Hannes Platzgummer; Vienna / Austria
1. To learn the anatomy of the brachial plexus.

2. To recognise the different traumatic lesions.
3. To identify the consequences of the lesions.
4. To understand what is important to describe in the report.
Imaging of other small nerves around the neck and the shoulder (20 min)
Carlo Martinoli; Rapallo / Italy
1. To become familiar with the complex anatomy of some small but clinically relevant nerves around the neck and the shoulder, at the
most common sites of pathology.
2. To learn the imaging appearance and scanning techniques used to image these nerves.
3. To be able to understand the range of pathologic conditions for which diagnostic imaging based on ultrasound and MR imaging is
appropriate for nerve assessment.
Panel discussion: MR neurography: a game changer? (15 min)
171
CUBE 5 - Endovascular bypass: natural evolution or science fiction?

Advanced Session - Topic Coordinator: Prof. Miltiadis Krokidis
The “Advanced Session: The Extra Mile” introduces the audience to techniques and treatments offered for challenging cases where an
out-of-the-box approach was required or where there has been an impactful learning point for clinical practice.
Moderator:

Endovascular bypass: natural evolution or science fiction? (28 min)

Fabrizio Fanelli; Roma / Italy
1. To define the indications of performing complex endovascular recanalizations.

2. To describe the most common techniques and tools of complex recanalizations.
3. To illustrate the outcomes.
172
ST 2 - The Perception Lab - Medical Image Perception at ECR
Categories: Research
The Perception Lab at ECR is a "pop-up" version of an academic research lab focused on medical image perception.
Led by Dr. Jeremy Wolfe from Harvard Medical School, several groups of researchers have come to Vienna to test radiologists and
other experts during the meeting. This session will provide insights about medical image perception in general and Dr. Wolfe's studies
on the "look but fail to see" errors in particular.
The Perception Lab is located on the first floor of the ACV, within the "Innovation in Focus" area. Be sure to stop by and learn more!
Moderator:
Interview (30 min)

Jeremy Michael Wolfe; Boston / United States
173
RC 502 - Alternative ways to MRI-guided breast biopsy
Categories: Breast, Interventional Radiology

CME Credits: 1
Moderator:
Gordana Ivanac; Zagreb / Croatia

Targeted conventional breast imaging (15 min)

Panagiotis Kapetas; Vienna / Austria
1. To understand the indications of targeted ultrasound (US).

2. To learn how to look for MRI-detected lesions on second-look ultrasound.
3. To understand lesion position change from supine imaging to prone imaging.
4. To understand the role of elastography and contrast-enhanced ultrasound in second-look ultrasound.
Volume navigation guided biopsy (15 min)

Mustafa Erkin Aribal; Istanbul / Turkey
1. To present the indications and basic principles of volume navigation guided biopsy.
2. To learn how to perform this technique and review cases.
3. To learn tips and tricks for a successful application of this technique.
Contrast-enhanced mammography guidance: a new approach to image-guided procedures (15 min)

Rodrigo Alcantara Souza; Barcelona / Spain
1. To understand the procedure and indications of CEM guidance.

2. To evaluate the strengths, limitations, and future relevance of CEM guidance.
3. To examine practical cases and procedural tips.
Panel discussion: Challenges in the adoption of alternative ways to MRI-guided biopsy (10 min)
174
OF 5T - Career development: being a better next-generation radiologist
Categories: Education, General Radiology, Management/Leadership, Professional Issues, Research, Students

CME Credits: 1
Moderator:

Being a better clinician (15 min)

1. To discuss opportunities to develop a rounded clinical practice.

2. To discuss how this can benefit the next generation of radiologists and patients.
Being fit for practice (15 min)

Gennaro D'Anna; Legnano / Italy
1. To discuss opportunities to optimise work-life balance.

2. To discuss how this can benefit young radiologists.
Radiology for the long haul: avoiding burnout (15 min)

1. To discuss opportunities to develop a rounded clinical practice.

2. To discuss how this can benefit the next generation of radiologists and patients.
Open forum discussion: How can a next-generation radiologist excel? (10 min)
175
MR 5 - MR quality: what, why, and how
Categories: Education, Evidence-Based Imaging, Management/Leadership

ETC Level: LEVEL II
CME Credits: 1
Moderators:
Vítor Manuel F. Silva; Porto / Portugal
Johanna Maria Lieb; Basel / Switzerland

Johanna Maria Lieb; Basel / Switzerland
What is MRI quality and how do you assess it? (19 min)
Simone Busoni; Firenze / Italy
1. To learn the fundamental aspects of MRI quality, including image acquisition, technical parameters, image artefacts, and overall
image interpretation.
2. To appreciate the importance of comprehensive quality assessment in MRI, considering both technical aspects (e.g., image
resolution, signal-to-noise ratio) and clinical relevance (e.g., diagnostic accuracy, impact on patient management).
3. To understand the methodologies and tools available for assessing MRI quality, such as objective metrics, visual evaluation, and
quality assurance programmes, and their role in ensuring reliable and high-quality imaging outcomes.
How to improve the quality and the value of the MR imaging process (19 min)
Ruth Tuura O’Gorman; Zurich / Switzerland
1. To learn the significance of multidisciplinary collaboration between radiographers, radiologists, and physicists in optimising the MR
imaging process and enhancing patient outcomes.
2. To appreciate the value of a coordinated MR quality assessment programme that involves teamwork, regular communication, and
shared responsibilities among the key stakeholders.
3. To understand the practical strategies and best practices for implementing a collaborative approach, including standardised
protocols, continuous education, and quality improvement initiatives, to improve the overall quality and value of MR imaging for
patients.
Identifying imaging artefacts and optimising the acquisition (19 min)

1. To learn practical techniques for identifying and minimising image artefacts in MRI, including common sources of artefacts and
troubleshooting strategies.
2. To appreciate the impact of optimising image resolution, signal-to-noise ratio (SNR), and contrast on diagnostic image quality and
the clinical interpretation of MR images.
3. To understand the practical considerations and technical parameters involved in optimising resolution, SNR, and contrast, along
with the trade-offs and potential limitations associated.
176
OF 5R - Playful imaging: where play meets care
Categories: Education, Imaging Methods, Multidisciplinary, Paediatric, Radiographers

CME Credits: 1
In the ever-evolving landscape of healthcare, the incorporation of innovative and compassionate approaches is paramount. This
session embodies this ethos, exploring novel methodologies that infuse elements of play and empathy into healthcare practices
within our radiology and radiotherapy departments. This session should be of interest to radiographers, radiologists, healthcare
professionals, researchers, and educators as it will explore the confluence of advanced medical imaging, empathetic care, and the
human spirit, showcasing how playful imaging can revolutionise healthcare, creating a brighter, more compassionate future and
enhanced experiences for patients and health practitioners alike.
Moderator:
Berit Møller Møller Christensen; Jönköping / Sweden

Berit Møller Møller Christensen; Jönköping / Sweden
Using play and child participation to support children in the radiology department (16 min)
Jannie Bøge Steinmeier Larsen; Aarhus V / Denmark
The use of VR and gaming to reduce patient anxiety (16 min)

Stefan Liszio; Düsseldorf / Germany
Embracing canine companions in orthopaedic and cancer care (16 min)

Martin Peyton; Cork / Ireland
177
ESR/EFSUMB - Multiparametric ultrasound (MPUS) in small parts examinations: where are we

standing?
Categories: Education, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods

CME Credits: 1
Moderators:
Caroline Ewertsen; Copenhagen Oe / Denmark

Caroline Ewertsen; Copenhagen Oe / Denmark
MPUS of the thyroid (13 min)

Vito Cantisani; Roma / Italy
1. To show baseline, CEUS and elastography features of a thyroid nodule.

2. To discuss the accuracy of single techniques.
3. To provide an algorithm for thyroid nodules assessment.
MPUS of the breast (13 min)

1. To show baseline, CEUS and elastography features of breast lesions, benign and malignant.
2. To discuss the value of MPUS.
3. To provide an algorithm for the assessment of breast lesions.
MPUS of the testes (13 min)

Michele Bertolotto; Trieste / Italy
1. To show baseline, CEUS and elastography features of lesions in the testes.

2. To discuss the accuracy of each technique.
3. To discuss when to use more than B-mode ultrasound.
Future perspectives of MPUS, including AI (13 min)

1. To present the available evidence of AI in US.

2. To discuss how AI may be helpful in US.
3. To discuss the challenges of AI in US.
178
HW 5R - Dose management software in interventional radiology and cardiology
Categories: Cardiac, EuroSafe Imaging/Radiation Protection, Interventional Radiology, Radiographers

CME Credits: 1
In this session the participants will be able:

1. To describe best practices, tips and tricks for using dose management software in interventional radiology and cardiology.
2. To demonstrate and apply best practices, tips and tricks for using dose management software in interventional radiology and
cardiology.
Minimising peak skin dose: presentation, live software demonstration, and Q&A (30 min)
Dania Kawood
Real-time dosimetry - managing occupational radiation exposure in IR and cath-labs: presentation, live software
demonstration, and Q&A (30 min)
Maxime Denis
179

CME Credits: 1


180
RC 501 - What shall I quantify in abdominal organs?
Categories: Abdominal Viscera, Imaging Methods

CME Credits: 1
Moderator:

Fat: liver, pancreas, visceral fat (15 min)

Manuela M. França; Maia / Portugal
1. To understand the different clinical scenarios and the clinical need for liver, pancreas and visceral fat quantification.
2. To know the role of different imaging techniques for assessing fat deposition in liver, pancreas and visceral fat.
3. To apply the techniques in clinical practice and know their strengths and limitations.
Iron: liver, pancreas, spleen (15 min)

Benjamin Henninger; Innsbruck / Austria
1. To reflect the basic MR techniques for iron quantification in the abdomen.

2. To list the relevant diseases associated with increased iron storage in the liver, spleen and pancreas.
3. To describe the relationship between iron storage in the liver, spleen and pancreas.
Inflammation and fibrosis: liver, pancreas (15 min)

1. To understand the diagnostic value of US and MR elastography in assessing liver fibrosis.

2. To be aware of the role of US and MRI in evaluating liver inflammation.
3. To discuss quantitative MRI in the detection and quantification of fibrosis and inflammation in the pancreas.
Panel discussion: Integrating quantification of imaging biomarkers in clinical routine (10 min)
181
OF 5E - A glimpse inside the life of a journal

CME Credits: 1
In this session, organised by the ESR Journal Family, prominent editors from the ESR and other journals will give you an insight into
editorial work. You will learn about the qualities of good reviewers, considerations of editors' decisions, and current challenges of
scientific publishing.
Moderator:
Yves Menu; Paris / France

Do's and Don'ts when reviewing and commenting (15 min)

Jose Maria Garcia Santos; Murcia / Spain
A day in an editor's life and how not to get desk-rejected (15 min)
Challenges for journals in a fast-changing world (15 min)

Vicky Goh; London / United Kingdom
182
AI-SC 2 - Interoperability for AI in radiology: a collaborative approach between different stakeholders

Moderator:
Nadya Pyatigorskaya; PARIS / France

Panel Discussion (57 min)

David Goyard; Neuchâtel / Switzerland
Benoît Rizk; Villars-sur-Glane / Switzerland
Yannick Re; Berre L'Etang / France
Annalisa Trianni; Trento / Italy
1. To discuss industry experience.

2. To present radiologists’ expectations.
3. To identify current gaps and possible solutions.
183
RC 515 - Aortic aneurysm and dissection: classification, imaging, and management
Categories: Imaging Methods, Interventional Radiology, Multidisciplinary, Vascular

CME Credits: 1
Moderator:
Luca Saba; Cagliari / Italy

Anatomy, classification, pathophysiology and clinical assessment (12 min)

Robert Anthony Morgan; Dorking / United Kingdom
1. To have detailed knowledge of the anatomy and physiology of the abdominal aorta.
2. To name the major classifications of the aortic dissections (DeBakey, Stanford) and possible complications.
3. To describe the role of each imaging modality in dissection and aneurysm, including US, CTA and MRA.
The role of aortic CTA (12 min)

Ruediger Schernthaner; Vienna / Austria
1. To discuss the technical details of aortic CTA.

2. To identify common findings of aneurysm and dissection at CTA.
3. To interpret post-interventional follow-up CTA.
The role of aortic MRA (12 min)

Ferenc Imre Suhai; Budapest / Hungary
1. To summarise the technical details of aortic MRA.

2. To describe common findings of aneurysm and dissection at MRA.
3. To learn how to choose the appropriate imaging modality.
Role of interventional radiology in the management of aortic aneurysm and dissection (12 min)
Fabrizio Fanelli; Roma / Italy
1. To name and identify the difficulties and risk factors for endovascular treatment of patients with aneurysms and vasculitis.
2. To select endovascular treatment options for patients with aneurysms and vasculitis.
3. To list the characteristics of post-intervention management of these patient groups.
Panel discussion: How best to perform and report my CTA and MRA study? (7 min)
184
BS 5 - Embracing new techniques and technologies
Categories: Education, Imaging Methods, Nuclear Medicine, Radiographers, Research

CME Credits: 1
Moderator:
Luís Pedro Vieira Ribeiro; Lagoa - Parchal / Portugal

Luís Pedro Vieira Ribeiro; Lagoa - Parchal / Portugal
Dynamic PET using parametric imaging (18 min)

Christina Baun; Odense / Denmark
1. To describe the principles of dynamic PET imaging and its significance in medical and preclinical research, including the impact of
acquisition techniques on parametric imaging analysis.
2. To explore the applications of dynamic PET in preclinical and clinical fields such as oncology, neurology, and cardiology.
3. To discuss the challenges and limitations associated with dynamic PET and parametric imaging.
A radiographer's guide to functional MRI (18 min)

Claude Portanier Mifsud; Msida / Malta
1. To define functional MRI of the brain and list potential uses of functional MRI.
2. To describe the physiologic response of neuronal activity and how this effect is captured with MRI.
3. To examine the role of the radiographer during functional MRI.
Exploring the potential benefits and ethical considerations of ChatGPT for radiography and scientific research (18 min)
Moreno Zanardo; Limbiate / Italy
1. To describe the potential benefits and challenges of ChatGPT in radiography.

2. To explore the applications of ChatGPT in scientific research.
3. To discuss the ethical concerns of using ChatGPT in radiography and scientific research.
185
RPS 503 - Trials and meta-analyses: cardiac CT and MRI
Categories: Cardiac, Evidence-Based Imaging, Imaging Methods, Research

CME Credits: 1
Moderator:
François Pontana; Lille / France
Cardiovascular CT and MR imaging in Europe: insights from the ESCR registry (7 min)
Federica Catapano; Milan / Italy
Author Block: F. Catapano1, L. J. Moser2, M. Francone1, R. Vliegenthart3, C. Catalano4, M. Gutberlet5, H. Alkadhi2; 1Milan/IT, 2Zürich/CH,
3
Groningen/NL, 4Rome/IT, 5Leipzig/DE
Purpose: The aim of this study was to provide an overview of advanced cardiovascular imaging practices in Europe using structured
data from the European society of cardiovascular radiology (ESCR) registry.
Methods or Background: Numbers on cardiovascular CT and MRI examinations were extracted from the ESCR-registry between
2009 and October 2023. Data collection included the total/annual numbers of examinations, indications, complications, and reporting
habits.
Results or Findings: The ESCR registry demonstrates a 6.8-fold increase of annually submitted CT examinations from 2,244 to
15,267, and a 4.7-fold increase of MRI examinations from 2,803 to 13,183 between 2010 and 2022. Reporting of CT (76%) and MRI
(71%) was mostly performed by radiologists, and, to a lesser degree, in consensus with non-radiologists (19% and 27%, respectively).
Main indications for cardiac CT were suspected coronary artery disease (CAD) (59%), TAVI-planning (21%), valve disease (7%) and
preablation (6%). Main MRI indications were myocarditis (26%), suspected CAD (including stress-imaging) (21%), and cardiomyopathy
(19%). Adverse event rates were very low for CT (0.3%) and MRI (0.7%).
Conclusion: The largest available registry on Cardiovascular CT and MRI in Europe demonstrates a considerable increase in exam
numbers, in particular for CAD and CT. These findings collectively contribute to our understanding of the current state of
cardiovascular imaging in Europe.
Limitations: Our data, extracted from an ESCR registry, may inherently favour a radiological perspective in the representation of
clinical practice in cardiac imaging. Nevertheless, they do represent the largest cases archive in Europe.
Ethics committee - additional information: The study is retrospective or educational.
Cost-effectiveness of computed tomography in patients with atypical chest pain clinically referred for invasive
coronary angiography: randomised controlled trial (7 min)
Mahmoud Mohamed; Berlin / Germany
186
Author Block: M. Bosserdt1, M. Mohamed1, K. Neumann1, N. Rieckmann1, H. Dreger1, V. Brodszky2, T. Reinhold1, A-M. Mielke1, M.
Dewey1; 1Berlin/DE, 2Budapest/HU
Purpose: Is coronary computed tomography (CT) cost-effective compared with invasive coronary angiography (ICA) in patients with
atypical chest pain who are clinically referred for ICA?
Methods or Background: A prespecified cost-effectiveness analysis of 329 patients with atypical angina or chest pain from a
randomised pragmatic trial comparing CT and ICA conducted at a university hospital in Germany was performed. Cost-effectiveness
was analysed for up to 3 years of follow-up from the health sector perspective using quality-adjusted life years (QALYs) derived from
the EQ-5D-3L questionnaire. Costs were obtained from each individual's outpatient and inpatient billing data and included
cardiovascular medications, hospitalisations, emergency department visits, cardiologist visits, and cardiac examinations. Data
analysis included 500 multiple imputations followed by 1,000 bootstrapping iterations for each imputed data set, and the net
monetary benefit was calculated.
Results or Findings: There was no statistically significant difference in mean QALYs at either one-year or three-year follow-up, while
the mean cost per patient was significantly lower in the CT group compared with the ICA group, both at one year (difference in €:
-1,647.8, -2,198.3 to -1,937.0) and at three years (difference in €: -1,543.3, -2,228.0 to -830.0). At a willingness-to-pay threshold of
€20,000/QALY, the average incremental net monetary benefit of CT over ICA was €1,256.5 (164.8 to 2331.8) at one year and €1202.0
(-1,378.7 to 3,961) at three years. The incremental net monetary benefit of CT over ICA at three years was the highest in patients
with a pretest probability of CAD above 30% (€ 1445.6, -1803.1 to 4637.0).
Conclusion: A CT-first strategy for the management of patients with atypical angina or chest pain was more cost-effective than a
direct-to-ICA strategy.
Limitations: This analysis considered only cost from the health sector perspective.
Funding for this study: This study was funded by a grant of the Heisenberg programme of the German Research Foundation to
Marc Dewey.
Ethics committee - additional information: This trial was approved by the ethics committee of the Charité –Universitätsmedizin
Berlin (EA1-1-080-08) and by the German Federal Office for Radiation Protection (Z5–22462/2–2008-048). Before the randomisation all
participants gave written informed consent.
Detection of calcified plaques on coronary CT angiography compared to thin-slice non-contrast CT; multicentre trial
evaluation (7 min)
Kenrick Schulze; Berlin / Germany
1 1 1 1 1 1 2 3 1
Author Block: K. Schulze , B. Föllmer , F. Biavati , R. Bockelmann , M. Bosserdt , F. Michallek , J. D. Dodd , K. F. Kofoed , M. Dewey ;
1 2 3
Berlin/DE, Dublin/IE, Copenhagen/DK
Purpose: The aim of this study was to assess the feasibility of assessing calcified coronary artery plaques on CT angiography (CTA)
compared to non-contrast CT (NCCT) in a multicentre study.
Methods or Background: This study included 47 patients from the DISCHARGE trial subgroup (mean age 62.0 ± 11.0 years, 57.4%
male) with available thin-slice (< 0.7 mm) NCCT and CTA. The diagnostic accuracy and detection of manually segmented coronary
calcified plaques was automatically assessed for CTA and NCCT using a definition of a volume of at least 1 mm³. Plaques on CTA were
defined as missed if there was no spatial overlap with a calcified plaque on NCCT after registration. Sensitivity and specificity were
calculated using NCCT as reference standard. Lesion level statistics were analysed for plaque density and volume parameters, with
plaques categorised into the groups 'all' and 'missed'.
Results or Findings: NCCT identified 314 coronary calcified plaques of which 213 (32% sensitivity) were missed in CTA alone. CTA
was not associated with false positive calcified plaques. Missed coronary calcified plaques on CTA had higher density (537.3 HU ±
224.9 HU) compared to NCCT (467.9 HU ± 215.3 HU). These plaques were detectable on NCCT based on density, while CTA density
was closer to aortic density (393.4 HU ± 101.0 HU). Additionally, missed plaques were smaller in volume (7.15 mm³ ± 9.0 mm³)
compared to all plaques (21.0 mm³ ± 57.3 mm³). Overlooking calcified coronary plaques would have led to the omission of 11 out of
47 (23%) patients with CAD-RADS 1, incorrectly categorising them as CAD-RADS 0.
Conclusion: CTA can miss up to two thirds of coronary calcified plaques visible on NCCT, highlighting the important role of NCCT.
Limitations: This study involved a small subset of 47 patients from three of the 26 DISCHARGE trial centres.
Funding for this study: This study was funded by grants from the EU-FP7 Framework Program (FP 2007-2013, EC-GA 603266) and
funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) in the graduate program on quantitative
biomedical imaging (BIOQIC, DFG project number: 289347353, GRK 2260/1).
Ethics committee - additional information: The DISCHARGE trial was approved by the ethics committee at
Charité–Universitätsmedizin Berlin as the coordinating centre, by the German
Federal Office for Radiation Protection, and by local or national ethics
committees.
Dedicated cardiovascular screening in lung cancer screening: preliminary results from the European 4-IN-THE-LUNG-
RUN trial (7 min)
Daiwei Han; Groningen / Netherlands
187
1 1 2 3 4 5 2 1
Author Block: D. Han , M. Vonder , C. Van Der Aalst , A. Schmitz , J. W. C. Gratama , M. Silva , H. J. De Koning , M. Oudkerk ;
1 2 3 4 5
Groningen/NL, Rotterdam/NL, Amsterdam/NL, Apeldoorn/NL, Parma/IT
Purpose: The 4-IN-THE-LUNG-RUN (4ITLR) trial, which was recently started and aims to enroll 26,000 participants, offers the
opportunity for prospective cardiovascular screening within a lung cancer screening program. Although current guidelines advise the
assessment of coronary calcifications on chest CT scans of any kind, this requires specific imaging acquisition and reconstruction for
accurate Agatston score evaluation. This study explores the potential benefits of cardiovascular screening within the initial
participants of 4ITLR trial.
Methods or Background: The inclusion criteria were: age 60-79 years, smoking history of ≥35 pack-years, and current smoking or
quitting within the last 10 years. High-temporal-resolution low-dose chest CT scans using a third-generation dual-source CT scanner
were performed on 443 participants between January 15th and March 29th, 2023, at a single centre. An automatic assessment of the
Agatston score was conducted on dedicated cardiac reconstructions that utilised a slice thickness/increment of 3.0/1.5 mm, a
medium-sharp kernel, and high-pitch acquisition, with an FBP algorithm and 120 kVp. This allowed for the reliable categorisation of
the participants' risk based on their Agatston score, with categories being low risk (0 score), moderate risk (1-99), high risk (100-399),
and very high risk (≥400).
Results or Findings: The mean age was 68.6 years (SD 4.9), with 56.9% male. Median Agatston scores were 242.1 (IQR 34.8-939.9)
for men and 56.3 (IQR 2.5-365.0) for women. 16.0%, 30.7%, 19.6%, and 33.6% of participants were at low, moderate, high, and very
high CHD risk, respectively. About 47% fell into low/moderate CHD risk categories.
Conclusion: The initial lung cancer screening identified one third of participants at significantly high CHD risk, while half were
suitable for preventive CHD treatment. Notably, 16% had a low CHD risk, exempting them from CHD preventive medication as per
existing guidelines.
Funding for this study: This project has received funding from the European Union's Horizon 2020 programme under grant
agreement no 848294.
Ethics committee - additional information: The minister of public health, well-being and sport has approved the execution of the
4-IN-THE-LUNG-RUN study.
Prognostic value of CT-derived myocardium-related parameters in patients with aortic stenosis: a systematic review
and meta-analysis (7 min)
ZiXian Chen; Lanzhou / China
Author Block: X. He1, Y. Li1, Y. Wang1, X. Lu1, J. Nan1, L. Cao1, Y. Wang2, G. Wang1, Z. Chen1; 1Lan Zhou city/CN, 2Shanghai/CN
Purpose: The aim of this study was to investigate the prognostic value of CT-derived myocardium-related parameters in patients
with severe aortic stenosis (AS) who have undergone aortic valve replacement (AVR).
Methods or Background: Four databases (PubMed, Web of Science, Embase, and Cochrane) were searched to identify studies
investigating the prognostic performance of CT-derived myocardium-related parameters in patients with AS. A random effects model
for meta-analysis was conducted to calculate pooled hazard ratios (HR) and 95% confidence intervals (CI) in order to assess the
prognostic value. The I2 statistic was used to assess heterogeneity.
Results or Findings: In this analysis, ten studies were identified, six of which involved 662 patients reporting CT-derived
extracellular volume fraction (ECV), and four studies, including 1244 patients reporting CT-derived left ventricular global longitudinal
strain (LVGLS). The meta-analysis revealed that ECV, whether considered as a dichotomous variable (pooled HR: 4.12, 95% CI:
2.76-6.15, I2 =0%, P< 0.001), or as a continuous variable (pooled HR: 1.15, 95% CI: 1.05-1.25, I2 =74%, P=0.002), and LVGLS,
whether considered as a dichotomous variable (pooled HR: 1.70, 95% CI: 1.31-2.19, I2 =0%, P< 0.001) or a continuous variable
(pooled HR: 1.07, 95% CI: 1.05-1.10, I2=0%, P< 0.001) were all significant predictors for all-cause mortality in patients with AS after
AVR.
Conclusion: This study has demonstrated the significant prognostic value of pre-AVR CT-derived ECV and LVGLS, both as
dichotomous and continuous variables, in predicting all-cause mortality in patients with AS. These findings enhance our
understanding of the pathophysiology of AS and assist in optimizing the timing of AVR.
Limitations: Limited number of studies in meta-analysis.
Funding for this study: This study is supported by the Lanzhou Science and technology project Foundation (2020-2D-80) and First
Hospital of Lanzhou University Hospital Foundation (ldyyyn2019-78).
Reporting trends and clinical impact of high-risk coronary plaque features on coronary CT angiography: a 4-year
analysis at tertiary referral hospital in the CAD-RADS era (7 min)
Won-Seok Yoo; Seoul / Korea, Republic of
188
Author Block: W-S. Yoo, Y. J. Suh; Seoul/KR
Purpose: The study aimed to evaluate the reporting trends of coronary artery disease-reporting and data system (CAD-RADS)
modifier HRP (high-risk plaque) on coronary CT angiography and its clinical impact on the management of patients with coronary
artery disease in a large tertiary referral hospital over a 4-year period.
Methods or Background: A total of 24578 cardiac CT performed between 2017 to 2021 were retrospectively reviewed. Reporting
prevalence of modifier HRP were analysed according to the patient characteristics, CT protocol or indication, CAD-RADS category, and
reader. To evaluate the effect of modifier HRP on patient management, medical records mentioning the presence of HRP and
additional studies or treatments that patients received after CT were reviewed in HRP-reported cases.
Results or Findings: Modifier HRP were reported in 0.8% of cases, ranging 0.1% to 1.6% depending on the readers. Modifier HRP
were more frequently reported in male than female (1.1% vs 0.6%) cases with clinical suspicion of acute coronary syndrome or stable
angina than others (1.4% versus 1.1% versus 0.2%), CT protocols for evaluation of CAD or triple-rule-out than others (1.2% versus
0.8% versus 0.2%), and higher CAD-RADS category (OR per category, 1.607)(P< 0.001 for all). The presence of modifier HRP was
recorded by clinicians in 53.1% (110/207) of HRP-reported cases, which was not different among the CAD-RADS categories (P=0.716).
Invasive coronary angiography was performed in 8.3% (2/24) of HRP-reported minimal stenosis (CAD-RADS category 1), 20.0% (6/30)
of mild stenosis (category 2), 32.7% (16/49) of moderate stenosis (category 3), and 84.6% (88/104) of severe stenosis or total
occlusion (category 4-5).
Conclusion: CAD-RADS modifier HRP had considerable impact on the management of patients with CAD. However, reporting trends
of modifier HRP varied according to the readers and CT protocols or indications.
Ethics committee - additional information: Currently under review
Evaluating the diagnostic performance of magnetic resonance imaging in diagnosis of cardiac sarcoidosis: a diagnostic
accuracy meta-analysis (7 min)
Ahmed Tarek Saeed Benghatnsh; Tripoli / Libya
Author Block: A. T. S. Benghatnsh1, A. Msherghi1, A. Khaled1, A. Elfaituri1, H. Faraj2, M. K. Elfaituri1; 1Tripoli/LY, 2Doha/QA
Purpose: This study aimed to analyse the diagnostic accuracy of cardiac MRI in diagnosing Cardiac Sarcoidosis.
Methods or Background: A systematic search was conducted across PubMed, Embase, and the Cochrane Library for studies
published up to November 2022 that assessed the diagnostic accuracy of cardiac MRI for CS, using biopsy as the reference standard.
For the statistical analysis, R software (version 4.0.3) and the mada package were employed, enabling the aggregation of sensitivity,
specificity, false-positive rate estimates, diagnostic odds ratio, as well as positive and negative likelihood ratios (LR). All these
outcomes are expressed with a 95% confidence interval (CI).
Results or Findings: The analysis incorporated a total of 8 studies, including 622 participants, of whom 146 were confirmed
diseased patients and 476 were non-diseased. Our findings revealed that cardiac MRI demonstrated a high sensitivity of 95.9% (95%
CI: 91.2-98.1%, I^2=0%) and a specificity of 87.8% (95% CI: 72.5-95.2%, I^2=0%). The false-positive rate was 12.2% (95% CI:
4.8-27.5%). The positive and negative likelihood ratios were 7.88 (95% CI: 3.24-19.13) and 0.05 (95% CI: 0.02-0.10), respectively, and
the diagnostic odds ratio was 168.45 (95% CI: 45.97-617.23). A very low heterogeneity was observed among the studies (I^2 = 0%),
indicating a high level of consistency in the diagnostic performance of cardiac MRI across the different studies.
Conclusion: This study demonstrates cardiac MRI's high sensitivity and specificity in diagnosing Cardiac Sarcoidosis. It highlights
cardiac MRI's potential to enable early detection and accurate diagnosis, leading to timely treatment and improved prognosis.
However, additional research is needed to standardise MRI protocols and confirm these findings across diverse patient populations.
Limitations: The study included a low number of studies in the analysis, which necessitated the need for further large future studies.
Ethics committee - additional information: This is a meta-analysis study based on published data available without the need for
ethical committee approval.
Mitral annular disjunction: the boundary between normal and pathological. Results from a large multicentre National
Register of the Section of Cardioradiology of the SIRM (7 min)
Elisa Bruno; Milan / Italy
189
Author Block: E. Bruno1, A. Palmisano1, S. Dell'Aversana2, R. Faletti3, N. Galea4, M. Gatti3, C. Liguori2, S. Pradella5, A. Esposito1;
1
Milan/IT, 2Naples/IT, 3Turin/IT, 4Rome/IT, 5Florence/IT
Purpose: Mitral annular disjunction (MAD) is an anatomic variant characterised by the atrialisation of the mitral valve junction; its
prevalence and pathological role are still debated.
The aim of the study is to evaluate the incidence of MAD in a vast cohort of patients who underwent cardiac magnetic resonance
(CMR), defining the correlation with morpho-functional or myocardial tissue alterations and arrhythmias.
Methods or Background: Multi-center observational study involving 13 Italian hospitals. CMR from January to June 2019 were
evaluated, assessing the presence of MAD, structural (prolapse, curling, regurgitation) and tissue alterations (LGE, T1, T2, ECV),
volumetric and functional features, clinical suspicion, diagnosis, presence and type of arrhythmias.
Results or Findings: From a total of 2611 patients (67% XX, 53 [IQR 39-65] years old), 5.4% (142 patients, 65% XX, 52 [IQR39-63]
year-old) had MAD. Of them, 8% underwent CMR for the suspicion of valvopathy, 5% for arrhythmias and 87% for other causes.
47/142 (33%) patients had arrhythmias, associated with valve prolapse (p=0.004) and bigger MAD length (p< 0.001).
83/142 (58%) had MAD without other cardiomyopathies, with prolapse in 43% of cases, associated with increased incidence of curling
(64% vs 17%; p-value<0.001) and higher ECV values (29% vs 25%; p=0.003).
Bi-leaflet prolapse was associated to more severe MAD compared to patients with single-leaflet prolapse or without (6 vs 3.5 and 4
mm; p=0.083), bigger left atrial volume (40 mL/ m2 vs 2 mL/ m2 and 27 mL/ m2; p=0.011), left ventricle volume (170 mL vs 134 and
111 mL; p< 0.001), higher rate of moderate-severe regurgitation (45% vs 6% and 0%; p< 0.001) and arrhythmias (63% vs 36%;
p=0.037), without significant differences in the presence of LGE, whereas more frequent (35% vs 25% and 27%; p=0.794).
Conclusion: MAD is frequent in the population even without valvopathies. Its severity, the association with structural alterations lead
to a higher risk of myocardial remodelling and arrhythmias.
Limitations: Absence of follow-up data.
Ethics committee - additional information: Prot. MIAMI
190
RPS 510 - Upper extremities and tumours
Categories: Imaging Methods, Musculoskeletal, Oncologic Imaging

CME Credits: 1
Moderator:
Mitja Rupreht; Maribor / Slovenia
Deep-learning accelerated 3T MRI of the shoulder after rotator cuff repair: a comparison to standard 3T MRI (7 min)
Author Block: R. P. Marcus, G. C. Feuerriegel, A. A. Marth, D. Nanz, R. Sutter; Zurich/CH

Purpose: This study aimed to assess the diagnostic performance of accelerated MRI acquisition with deep-learning reconstruction
and edge sharpening in patients after rotator cuff repair.
Methods or Background: Unenhanced MRI scans of the operated shoulder were performed on 116 patients using a 3 Tesla (T)
scanner (Magnteom Prisma, Siemens Healthineers). Standard and accelerated protocols were used during the same scanning session
acquiring three sequences in coronal plane (STIR, PD TSE and PD FS), two sequences in sagittal plane (STIR and T1 TSE), and one
sequence in axial plane (PD FS) with a slice thickness of 3mm each. The acquisition times of the standard and the accelerated
protocols were recorded. Two experienced radiologists, blinded to the acquisition protocol, evaluated postsurgical changes of the
tendons as per Sugaya classification (1: sufficient thickness, 2: sufficient thickness with partial high intensity, 3: insufficient thickness,
4: minor discontinuity, 5: major discontinuity).
Results or Findings: Interrater and intrarater reliabilities of the evaluated 207 operated tendons were strong (Kappa=0.824-0.863).
The accelerated protocol reduced the acquisition time by 80% (3.7mins (2.5-6.3) vs. 18.4mins (17-33.7) for accelerated vs. standard
protocol, p<0.001). A total of four intrarater disagreements were observed (2%), mostly due to motion artifacts in the standard
acquisition. No motion artifact was recorded during the accelerated protocol.
Conclusion: Accelerated MRI acquisition with deep-learning reconstruction and edge sharpening significantly shortened the
acquisition time while an excellent depiction of evaluating the condition of the shoulder after rotator cuff repair.
Limitations: This study has a small sample size.
Ethics committee - additional information: This study was approved by a local Ethics Committee in the Canton of Zurich.
Bone mineral density assessment from spectral localiser radiographs: proof-of-concept with clinical photon counting
detector CT (7 min)
Lukas Jakob Moser; Zürich / Switzerland
191
Author Block: L. J. Moser1, J. Pitteloud1, V. Mergen1, D. Frey1, T. Nowak2, O. Distler1, M. Eberhard3, H. Alkadhi1; 1Zurich/CH,
2
Forchheim/DE, 3Unterseen/CH
Purpose: This study aimed to evaluate the feasibility and accuracy of bone mineral density (BMD) quantification from spectrally
acquired localiser radiographs (LRs) with photon-counting detector (PCD)-CT, in comparison with DXA.
Methods or Background: For this experimental study, spectrally-resolved LRs on a clinical dual-source PCD-CT scanner (NEAOTOM
Alpha, Siemens) acquired with 140kVp and with four different tube currents (10mA, 20mA, 30mA and 40mA) of the European Spine
Phantom were obtained. Scans were repeated three times. We used two setups: (A) the phantom placed directly on the CT table and
(B) the phantom submerged in water with an anterior-posterior diameter of 30cm simulating a slightly overweight patient.
Corresponding DXA scans were acquired. Assessment of aBMD in LRs was enabled by leveraging the spectral information with
material decomposition maps for hydroxyapatite and water. The spine phantom-supplied values were used as the reference standard.
Results or Findings: The mean absolute error (MAE) in setup A for LR-derived aBMD across all tube currents and vertebrae was
0.003g/cm2 with a relative difference of ΔaBMD ranging from -1.5% to 1.1%. Coefficients of variance (CV) were below 1% stratified for
tube current in setup A. In setup B, the MAE was 0.018g/cm2, and the ΔaBMD ranged from +1.0% to +5.3%. CV was below 3% for all
tube currents and below 2% excluding 10mA. The absolute error of DXA scans in both setups (-0.041g/cm2 for setup A, -0.06g/cm2
for setup B) was higher than corresponding MAEs from PCD-CT.
Conclusion: Our phantom study indicates that LRs obtained with PCD-CT enable the calculation of BMD with high accuracy.
Limitations: This was a phantom study and represents pre-clinical software to reconstruct raw data from LRs.
Ethics committee - additional information: This study was performed on a phantom.
Automatic motion analysis of the wrist using dynamic CT imaging for diagnosis of scapholunate ligament injuries (7
min)
Maranda Haenen; Nijmegen / Netherlands
Author Block: M. Haenen, E. Teule, S. Hummelink, I. Sechopoulos, B. van der Heijden; Nijmegen/NL
Purpose: This study aimed to analyse wrist kinematics in healthy and scapholunate interosseous ligament (SLIL) injured wrists using
four-dimensional computed tomography (4DCT) and a fully automated motion analysis algorithm.
Methods or Background: 4DCT scans of 41 healthy wrists and eight wrists with arthroscopically-confirmed Geissler 4 SLIL injuries
were acquired. Each scan consisted of one conventional static CT image and two dynamic imaging sequences; wrist radial-ulnar
deviation (RUD) and flexion-extension (FE). Scanner rotation time was 0.275s, and acquisition parameters include 80 kV and 40 mAs.
A sampling rate of 10Hz resulted in 144 dynamic 4DCT frames per wrist. Bones were automatically segmented in each frame using an
artificial intelligence-based algorithm, and the scapholunate distance (SLD) and sagittal scapholunate angle (SLA) were automatically
estimated per dynamic frame and resampled to a uniform distribution of wrist positions for intersubject comparison. The median and
maximum SLD and SLA values were calculated. A Mann-Whitney U test was employed to compare both groups.
Results or Findings: The median and maximum SLDs and SLAs were significantly larger in the injured group compared to the
healthy group. For example, the median and interquartile range were as follows: RUD (SLD) 2.49mm [1.36-3.40mm] vs 0.86mm
[0.64-1.06mm], p<.001; RUD (SLA) 93.0⁰ [83.8-103.0⁰] vs 69.2⁰ [62.3-74.1⁰], p<.001 and FE (SLD) 1.99mm [1.31-3.23mm] vs 0.92mm
[0.71-1.17mm], p<.001; FE (SLA) 92.3⁰ [66.6-102.3⁰] vs 71.9⁰ [57.0-85.2⁰], p<.001).
Conclusion: 4DCT is an emerging modality that enables non-invasive analysis of wrist motion. SLDs and SLAs automatically
estimated from 4DCT scans during wrist motion could be used to diagnose SLIL injuries non-invasively.
Limitations: The small sample size of injured wrists and the lack of independent validation of quantitative estimates limit this study.
Funding for this study: This study was internally funded.
Ethics committee - additional information: This study was approved with the ABR number: NL72518.091.19.
Initial findings of IVIM-DWI validation using DCE-MRI in soft tissue sarcomas (7 min)
Gizem Timoçin Yığman; Istanbul / Turkey
192
Author Block: G. Timoçin Yığman, A. Peker, Y. E. Senturk, H. Özen Atalay; Istanbul/TR
Purpose: The aim of this study was to investigate the relationship between quantitative DCE-MRI and intravoxel incoherent motion
diffusion-weighted imaging (IVIM-DWI) parameters in patients with soft tissue sarcomas (STS).
Methods or Background: From March 2022 to August 2023, patients were included who were histopathologically diagnosed as STS
in our centre and who had a DCE-MRI and IVIM-DWI. Patients with a history of chemotherapy or radiotherapy before MRI and patients
with a diagnosis of liposarcoma were excluded. Two radiologists with specific musculoskeletal imaging experience evaluated all the
exams (3 and 4 years). Patients' age, gender, and the longest diameter of the mass (LD) were noted. DCE-MRI measurements were
made from the Ktrans maps, avoiding bleeding, necrosis, and cystic areas. ROIs were placed the area with the highest perfusion in
both. Permeability parameters of the masses (Ktrans, AUC, Kep, Ve) were noted. IVIM-DWI was obtained using axial EPI with 11-b
values. IVIM-DWI was processed semi-automatically using Syngo.via V8.5 Frontiers (Siemens, Germany) and FD, D, and DP (D*) were
calculated.
Results or Findings: Ten patients, two females and eight males, were included. The mean age of the patients was 58±15.05
(23-70), and the mean LD was 124±66.44 mm (56-280). Interobserver agreement was excellent for DCE-MRI and good for IVIM-DWI
parameters. There was a strong correlation between K trans and FP (r=0.76; p=0.011), Kep and DP (r=0.77; p=0.009), and AUC and
FP (r=0.72; p=0.020). No significant correlation was observed in other parameters.
Conclusion: No study in the literature compares IVIM-DWI and DCE-MRI parameters in tumours. Our study demonstrated a strong
correlation between parameters. Studies with larger patient groups are needed to correctly interpret IVIM-DWI parameters and
replace DCE-MRI.
Limitations: The study is limited by the small number of patients from a single centre.
Ethics committee - additional information: This study was approved by the Koc University Biomedical Research Ethics
Committee, Istanbul.
Ultrasound-guided five per cent dextrose and corticosteroid injections in patients with carpal tunnel syndrome: one-
year follow-up (7 min)
Vesna Potočnik Tumpaj; Ljubljana / Slovenia
Author Block: V. Potočnik Tumpaj, G. Omejec, Ž. Snoj; Ljubljana/SI

Purpose: This study aimed to compare the long-term treatment effects of ultrasound (US)-guided five per cent dextrose and
corticosteroid injections in patients with carpal tunnel syndrome (CTS).
Methods or Background: We prospectively recruited 57 patients with mild to moderate bilateral CTS. Symptom severity and
functional status were assessed with the Boston Carpal Tunnel Questionnaire (BCTQ) and visual analogue scale (VAS).
Electrodiagnostic studies (EDx) and median nerve US were performed. For the left wrist, injectate was randomly selected (6ml of five
per cent dextrose or 2ml of corticosteroid mixed with 4ml saline). Afterwards, the right wrist was injected with the remaining solution.
Patients and examiners were blinded to the selection of the solution. Follow-up was performed at one, three, six, and 12 months.
Results or Findings: Both solutions demonstrated significant improvement for clinical outcome in VAS (at one, three and six months
follow-up; p<0.05) and BCTQ score (at one, three, six, and 12 months follow-up; p<0.05) compared to baseline. No changes in
median nerve cross-sectional area (CSA) were present, however, the flattening ratio significantly improved at three, six, and 12
months follow-up (p<0.05). No significant difference was present between solutions at each follow-up (p>0.05).
Conclusion: Ultrasound-guided carpal tunnel injections of corticosteroids and five per cent dextrose significantly improve clinical
outcomes in patients with CTS up to one-year follow-up. No differences were present between solutions.
Ethics committee - additional information: This study was approved by the Slovenian National Ethics Committee.
Deep learning-based reconstruction enhances image quality and improves diagnosis in MR imaging of the shoulder
joint (7 min)
Zijun Liu; Zhengzhou / China
193
Author Block: Z. Liu1, B. Wen1, Z. Wang1, K. Wang2, L. Xie2, Y. Zhang1, J. Cheng1, Y. Zhang1; 1Zhengzhou/CN, 2Beijing/CN
Purpose: Accelerated MRI sequences reconstructed using a deep learning (DL) algorithm were found to be more likely than
Conventional-MRI sequences to detect subacromial bursal (SbA) thickening, however, the extent of this thickening was not
extensively explored. This study aimed to compare the image quality of DL-MRI with Conventional-MRI for shoulder examinations and
evaluate the effectiveness of DL-MRI in accurately demonstrating the degree of SbA and subcoracoid (SC) bursa thickening.
Methods or Background: A total of forty-one patients who underwent 3-T MRI of the shoulder were included in this study. Both
Conventional-MRI and accelerated MRI sequences were acquired. The image quality and degree of artifacts were assessed using a 5-
point Likert scale for both Conventional-MRI and DL-MRI. Additionally, the degree of SbA and SC thickening, as well as the relative
signal-to-noise ratio (rSNR) and relative contrast-to-noise ratio (rCNR), were analysed using the Paired Wilcoxon test. Statistical
significance was defined as p<0.05.
Results or Findings: The utilisation of accelerated sequences resulted in a remarkable 58% reduction in total scan time. Overall, DL-
MRI exhibited superior image quality scores and fewer artifacts compared to Conventional-MRI. Specifically, DL-MRI demonstrated
significantly higher measurements of SC thickenings in the oblique sagittal fat suppression (FS) density-weighted imaging (PDWI)
sequence (p=0.028) compared to Conventional-MRI. Moreover, DL-MRI exhibited higher detection of SbA thickenings in the oblique
coronal FS PDWI sequence, with a notable trend towards significant differences. Furthermore, DL-MRI exhibited better noise reduction
in muscle tissue compared to bone when compared to Conventional-MRI. All DL-MRI images exhibited significantly greater rSNRs and
rCNRs compared to Non-DL-MRI (p<0.001).
Conclusion: The utilisation of DL-MRI enhances image quality and improves diagnostic capabilities, making it a promising alternative
to Conventional-MRI for shoulder imaging.
Limitations: Bursal measurement accuracy could not be ensured as this study did not utilise an external reference standard.
Ethics committee - additional information: This study was retrospectively approved by the Review Committee of the First
Affiliated Hospital of Zhengzhou University (No: 2019-KY-0015-002).
Impact of different tube voltage in accuracy assessment for bone mineral density measurement on European spine
phantom: phantom and clinical research (7 min)
Yali Li; Zhengzhou / China
Author Block: Y. Li, H. Yuan; Beijing/CN

Purpose: The aim of this study was to evaluate the accuracy of bone mineral density (BMD) measurement in different tube voltages
on European Spine Phantom (ESP) and clinical patients.
Methods or Background: Image acquisitions were performed on Revolution CT (GE Healthcare, Waukesha, WI). Scan parameters
were: tube voltages (80, 100, 120, and 140kV), and tube currents (100, 125, 150, and 200mAs) on ESP, whilst tube voltage was 100
and 120kV with the automatic tube current modulation in patients. The relative errors (RE) and relative standard deviations (RSD)
were compared. The association and consistency of the two measurements were analysed by a Pearson correlation and Bland-Altman
plot. The intraclass correlation (ICC) was used to quantify the intertester agreement between the two measurements.
Results or Findings: The RE and RSD of the L1, L2, L3, and L1-3 were (17.82%, 7.66%, 5.25%, and 7.76% versus 0.20%, 0.57%,
0.13%, and 0.17%) in 80kV, (13.35%, 6.12%, 1.00%, and 4.26% versus 0.18%, 0.18%, 0.25%, and 0.18%) in 100kV, (3.87%, 1.28%,
2.68%, and 2.46% versus 2.22%, 1.11%, and 0.53%, and 0.94%) in 120kV, and (7.35%, 4.34%, 4.38%, and 4.8% versus 0.40%,
0.64%, 0.15%, and 0.16%) in 140kV, respectively. There were significant differences in different tube voltages (P<0.05), whilst there
were no statistically significant differences in different tube current groups (P>0.05). The average BMD 100kV was slightly higher than
BMD 120kV (127.85±43.55 versus 124.29±43.69 mg/cm3). The ICC was excellent (ICC=0.964, P=0.000) and there was a strong
positive correlation between the two methods (r=0.93, P<0.0001). A Bland-Altman plot revealed a high consistency of BMD 100kV
and BMD 120kV.
Conclusion: With the increase of the tube voltage, the BMD values of the lumbar spine decreased gradually. The BMD 100kV
revealed a highly correlation with BMD 120kV and can be utilised in determining osteoporosis and reducing radiation exposure on
patients.
Bone lymphoma presenting with lower back pain: a case series (7 min)
Vito Di Martino; Siena / Italy
194
Author Block: V. Di Martino, N. Di Meglio, A. Perrella, G. Bagnacci, C. Sica, M. A. Mazzei; Siena, SI/IT
Purpose: We report six cases of bone lymphoma presenting with lower back pain which occurred over a two year period, from 2020
to 2022, in our centre. This study highlights relevant imaging features with a special focus on MR.
Methods or Background: Bone lymphoma presenting with musculoskeletal pain is an often challenging diagnosis because its
imaging features are nonspecific. It is classified either as primary bone lymphoma (PBL, no other extraosseous site involved), or
secondary bone involvement in systemic lymphoma (SBL, secondary bone lymphoma). SBLs are much more frequent (30-50% of all
non-Hodgkin Lymphomas, NHLs, and 5- 10% of all Hodgkin Lymphomas, HLs), while primary bone lymphomas are rare (<5% of
primary bone tumours and 1% of all NHLs, more often a form of diffuse large B cell lymphoma, DLBCL).
Results or Findings: Four females and two males (aged 21 to 80 years) with persistent lower back pain underwent an MRI and
subsequently a body contrast-enhanced CT and/or a PET/CT. The diagnosis was confirmed with a biopsy: four PMLs (DLBCL), two SBLs
(one HL and one NHL). In all cases bone lesions were characterised by marked signal hypointensity on T1-weighted MRI corresponding
to heterogeneous hyperintensity on T2-weighted images, asymmetric, not confined to the subchondral bone and, in two cases, also
associated with soft tissue involvement but without significant cortical bone destruction. In four cases bone lesions were visible at CT
with a permeative or lytic pattern. Both cases of SBL presented systemic symptoms as well (fever and night sweats).
Conclusion: Bone lymphoma should always be considered in the differential diagnosis of patients presenting with lower back pain,
especially if the imaging features of bone lesions are atypical or associated with fever, night sweats, and weight loss.
Limitations: This study is limited by its case series methodology.
195
RPS 516 - Whole-body MRI in oncology
Categories: Hybrid Imaging, Imaging Methods, Molecular Imaging, Oncologic Imaging

CME Credits: 1
Moderator:
Marta Zerunian; Rome / Italy
Patients' selection for pelvic exenteration for gynaecologic neoplasm using whole-body MRI: preliminary results (7 min)
Salvatore Persiani; Policoro / Italy
Author Block: S. Persiani, G. Avesani, A. Perazzolo, L. D'Erme, C. Panico, V. Rufini, B. Gui, E. Sala; Rome/IT
Purpose: Pelvic exenteration is used to treat recurrent pelvic localisation in gynaecological patients. Critical patient selection is
crucial due to high postoperative morbidity. Typically, patients undergo preoperative MRI to evaluate local disease extent and FDG-
PET-CT to detect lymph nodes and distant metastasis. This study aims to evaluate whole-body-MRI (WB-MRI) in detecting extrapelvic
disease in gynaecologic cancer patients awaiting pelvic exenteration and to test WB-MRI's reproducibility among radiologists.
Methods or Background: In this IRB-approved study from June 2021 to September 2023, patients with recurring gynaecological
cancer prepped for pelvic exenteration underwent standard FDG-PET-CT, pelvic MRI, and additional sequences to assess extrapelvic
disease, adding 20-25 minutes to the MRI exam time. Two radiologists reviewed WB-MRIs for extrapelvic involvement, comparing
findings to FDG-PET-CT. Diagnostic accuracy and interradiologist agreement were evaluated.
Results or Findings: We included 36 patients. Seven patients had extra-pelvic localisations at PET-CT and were excluded from
surgery. WB-MRI correctly identified six patients with extrapelvic involvement; the patient not identified had a PET-positive para-aortic
lymph node with a short axis of 8mm. WB-MRI identified one patient with a suspicious peritoneal nodule, which was PET-negative and
histologically defined as fibrotic area after surgery. The radiological PI showed a sensitivity of 0.85 (CI 0.42-0.99), specificity of 0.96
(CI 0.82-0.99), positive predictive value of 0.91 (CI 0.60-0.99) and negative predictive value of 0.94 (CI 0.72-0.97), with an overall
accuracy of 0.93 (CI 0.80-0.99). The agreement between radiologists was 0.72.
Conclusion: WB-MRI has a good performance in detecting extrapelvic involvement in patients with gynaecologic malignancies and
has good reproducibility; it may be considered a stand-alone technique for patient selection for pelvic exenteration for gynaecologic
malignancies. However, these findings need to be confirmed in a larger sample.
Limitations: This was a monocentric study with a small sample size.
Ethics committee - additional information: This study was approved by the Fondazione Policlinico Gemelli Ethics Committee with
reference number: 3813.
Performance comparison between whole body MRI and 18F-PSMA-1007 PET CT in the detection of tumour recurrence in
prostate cancers (7 min)
Margarita Garcia Fontes; Montevideo / Uruguay
196
Author Block: M. Garcia Fontes, E. Otero, L. Valuntas, V. Gigirey, J. Mattos, G. Dos Santos, J. P. Gambini, P. Duarte, O. Alonso;
Montevideo/UY
Purpose: The aim of this study was to determine whether Whole Body Magnetic Resonance (WBMRI) with 3 Tesla offers similar
performance to 18F-PSMA-1007 PET CT for recurrent tumour detection in prostate cancers.
Methods or Background: A WBMRI scan with a 3T GE equipment and a 18F-PSMA-1007 PET CT with a 64 GE PET -CT were
performed in patients with prostate cancer and the suspicion of tumour recurrence. Thirty patients between 55 and 81 years old, with
a PSA value between 0.4 and 41 ng/ml, were included. The number of bone lesions, suspicious nodes, local recurrences, and
incidental findings with both methods were compared. The WBMRI study protocol was based on the METastasis Reporting and Data
System for Prostate Cancer (MET-RADS-P). An AXIAL T2 high Resolution and AXIAL DWI FOCUS sequences of the prostate were added
to better the assessment of local tumour recurrence. Statistics analysis was performed.
Results or Findings: Both WBMRI and 18F-PSMA 1007 PET CT detected the same number of bone and lymph node metastases and
local tumour recurrence. WBMRI found more bone lesions in two cases. Incidental findings revealed that both detected a
neuroendocrine tumour in the right iliac fossa and the WBMRI detected a kidney tumour.
Conclusion: WBMRI seems to be an adequate method for the evaluation of tumour recurrences of prostate cancer with the
advantages of greater accessibility, lower cost, and repeatability compared to PET CT18F-PSMA-1007. Further advantages include its
lack of radiation and lack of contrast media usage.
Limitations: WBMRI requires patient collaboration due to its longer duration compared to PET CT PSMA, and some of these patients
may be uncomfortable during the study. Biosafety limitations must also be considered. In order to validate the use of WBMRI in
control of patients with prostate cancer recurrence, a larger series is necessary.
Applying ONCO-RADS to whole-body MRI cancer screening in a retrospective cohort of asymptomatic subjects (7 min)
Chung-Jung Lin; Taipei / Taiwan, Chinese Taipei
Author Block: C-J. Lin1, Y-S. Hu2, C-A. Wu1, H-J. Chiou1; 1Taipei/TW, 2New Taipei/TW
Purpose: The aim of this study was to evaluate the frequencies of cancer in asymptomatic subjects undergoing whole-body magnetic
resonance imaging (WB-MRI) within the Oncologically Relevant Findings Reporting and Data System (ONCO-RADS) categories.
Methods or Background: We retrospectively included 2064 asymptomatic subjects participating in the WB-MRI cancer screening
program with a 3-T scanner between 2017 and 2022. Results of further investigations, including additional imaging and
histopathology exams performed at our institute, were used to confirm cancer. Two radiologists blinded to the clinical outcome
categorised the WB-MRI findings by using the ONCO–RADS categories as follows: 1 (normal), 2 (benign finding highly likely), 3 (benign
finding likely), 4 (malignant finding likely), and 5 (malignant finding highly likely). Firth logistic regression analysis was conducted to
determine the associations between the subject characteristics and findings of ONCO-RADS category ≥4.
Results or Findings: Of the 2064 subjects, 43 (2.1%) individuals had findings of ONCO-RADS category ≥4 and 24 (1.2%) had cancer
confirmed. The cancer frequencies per subject were 0.1%, 5.4%, 42.9%, and 75% for ONCO-RADS category 2, 3, 4, and 5,
respectively. In the multivariable model, older age (odds ratio (OR): 1.035, p=0.029), history of hypertension (OR: 2.051, p=0.026),
hepatitis B carrier (OR: 2.584, p=0.013), and prior surgery (OR: 3.787, p<0.001) were independently associated with ONCO-RADS-
category-≥4 findings.
Conclusion: The ONCO-RADS designed for cancer risk stratification was validated that higher numbers indicate greater likelihood of
cancer. The application of ONCO-RADS facilitates risk-based management after WB-MRI for cancer screening.
Limitations: This was a retrospective study with contrast usage in the general population.
Funding for this study: Funding for this study was received with the code: MOST 112-2314-B-A49-064.
Ethics committee - additional information: This study was approved by the Taipei Veterans General Hospital Internal Review
Board.
Comparison of fast 'all in one' whole-body MRI and 68Ga-prostate specific membrane antigen (PSMA) (PET/CT) in the
staging of high-risk prostate cancer (PCa) (7 min)
Sandy Van Nieuwenhove; Brussels / Belgium
197
Author Block: S. Van Nieuwenhove, R. Lhommel, V. Pasoglou, J. Van Damme, N. Michoux, B. Tombal, F. Lecouvet; Brussels/BE
Purpose: Next-generation imaging, including PET/CT and whole-body MRI, is disrupting the management of cancer patients. Few
studies have compared differences in staging of PCa patients. The objective of the study was to compare a Fast 'All in one' WB-MRI
study and a 68GaPSMA-PET/CT for local (T), nodal (N), and distant staging (M1a, M1b, M1c).
Methods or Background: This single-centre prospective study included 52 patients with high-risk PCa who underwent one fast 'All
in one' WB-MR examination (bi-parametric prostate assessment with optimised rapid T2w and DWI sequences and whole-body
assesment with 3DT1 and DWI sequences) and one 68GaPSMA PET/CT within one month.
Lecture of randomized anonymized exams using PSMA-RADS 1.0.
Gwet’s AC1 coefficient was used to quantify the agreement between techniques and a 2-sided Exact test was used to assess potential
proportional differences in positive findings.
Results or Findings: The Fast 'All in one' WB-MRI detected significantly more extracapsular extension (47.5 vs 30% p=0.002) with a
confirmation by radical prostatectomy in eight patients, while PSMA-PET/CT detected significantly more nodes (40.38% vs 28.85%;
p=0.0313) with a median small axis of 5mm±1.52. There was no significant difference between each modality in detecting extra
nodal metastases (p=0.625). Agreement between Fast 'All in One' WB-MRI and PSMA PET/CT was moderate in T (0.72 (0.54; 0.90)),
M1a (0.68 (0.48; 0.87)), good in N (0.79 (0.62; 0.95)) and M1b (0.86 (0.72; 0.99)), and very good in M1c (0.96 (0.89; 1.00)).
Conclusion: These preliminary observations show that a fast 'All in one' WB-MRI study better detects extracapsular extension. In
contrast, PSMA-PET/CT detects more nodal metastasis without differences regarding extra nodal metastasis.
Limitations: The limitations of the study were its monocentric design and small patient cohort.
Funding for this study: Funding for this study was received from GE Healthcare.
Ethics committee - additional information: This study was approved by an Institutional Review Board with code:
(2019/06FEV/060) and written informed consent was obtained from all participants.
Predictive role of sustained imaging MRD negativity assessed by diffusion-weighted whole-body MRI in multiple
myeloma (7 min)
Barbara Frittoli; Brescia / Italy
Author Block: B. Frittoli, A. Belotti, T. Falcone, C. Saeli, S. Gambarini, R. Ambrosini, A. Tucci, L. Grazioli; Brescia/IT
Purpose: Diffusion-weighted whole-body magnetic resonance imaging (DW-MRI) is used in the management of multiple myeloma
(MM) as part of the criteria for Response Assessment Category (RAC), as stated by the Myeloma Response Assessment and Diagnosis
System
(MY-RADS). Its prognostic usefulness in the detection of minimal residual disease (MRD) after
autologous stem cell transplantation (ASCT) has been previously established. The goal of our study is to examine the predictive role
of sustained imaging MRD-neg assessed by DW-MRI.
Methods or Background: DW-MRI and RAC criteria were performed in 70 MM patients, both after ASCT and at 1-year during
maintenance therapy in order to evaluate imaging-residual-disease. We combined DW-MRI results with those of bone marrow
samples, collected at day +100 after ASCT and at 1-year in MRD-neg patients.
Results or Findings: Multivariable analysis showed that progression free survival (PFS) and overall survival
(OS) were predicted by DW-MRI persistent disease (RAC≥2): p=0.001, HR 0.12 (95% CI:
0.05–0.30) for PFS, p=0.032, HR 0.20 (95% CI: 0.05–0.87) for OS. Median PFS was significantly
longer for patients with imaging MRD-neg at 1-year (RAC 1) compared to patients with residual disease on DW-MRI (RAC≥2) (median
PFS: 55.4 vs. 28.4 months; 3-years PFS: 91% vs. 30%, respectively (HR 0.12; 95% CI: 0.04–0.35; p=0.0001)). OS of imaging MRD-neg
patients at 1-year was significantly longer for patients with RAC1 vs. RAC≥2 (median OS: not reached (NR) vs. 63 months; 3-years OS:
100% vs. 82%, respectively (HR 0.13; 95% CI: 0.03–0.66; p=0.0007)).
Conclusion: Sustained imaging MRD negativity assessed by DW-MRI with RAC criteria has strong predictive relevance for survival in
MM patients on maintenance therapy after ASCT. This work has also been published in a journal (well done) (doi: 10.1002/ajh.26995).
Limitations: The retrospective nature of observations and the relatively low number of patients represent major limitations of our
study.
Ethics committee - additional information: This study has been done in accordance with the Declaration of Helsinki. The
submitter provided no additional information.
Potential added value of whole-body DWI-MRI in the diagnostic workup of patients with adenocarcinoma of unknown
primary (ACUP): a prospective pilot study (7 min)
Jeroen Willemse; Amsterdam / Netherlands
198
Author Block: J. Willemse, M. Lahaye, P. Snaebjornsson, S. Marchetti, M. Vollebergh, L. van Golen, W. Vogel, R. G. H. Beets-Tan, D. M.
J. Lambregts; Amsterdam/NL
Purpose: The purpose of this prospective diagnostic pilot study was to investigate the potential added benefit of whole-body (chest-
abdomen) MRI including DWI (further referred to as WB-MRI) in this diagnostic work-up of patients with adenocarcinoma of unknown
primary (ACUP).
Methods or Background: From January 2022 to August 2023, the option of WB-MRI was added as an adjunct diagnostic tool to the
routine clinical workup, including CT, FDG-PET/CT, lab tests, and biopsies for ACUP patients in our institution. The choice of whether to
perform an MRI was discussed by a multidisciplinary team and guided by all available clinical information (e.g. if the primary tumour
could likely be suspected to be located within the abdomen/pelvis). We analysed the impact of WB-MRI in terms of primary tumour
identification and detection of additional metastatic sites.
Results or Findings: WB-MRI was performed in 27 ACUP patients. In 6/27 (22%) of patients, WB-MRI suggested a possible primary
tumour location undiagnosed on previous CT and/or FDG-PET/CT, including 2 bile duct cancers, 1 ovarian, 1 appendiceal, 1 duodenal
and 1 pancreatic cancer. In 5 of these 6 cases, the WB-MRI diagnosis aligned with and supported the final diagnosis, established by
integration of clinicopathological data with whole genome sequencing. In addition, WB-MRI discovered extra metastatic sites in 6/27
(22%) of patients, including peritoneal metastases (n=3), bone (n=1), kidney (n=1) and testicular metastases (n=1).
Conclusion: This study demonstrates the potential added value of WB-MRI in the search for the underlying primary tumour in the
complex diagnostic work-up of patients with disseminated adenocarcinoma of unknown primary.
Limitations: This was a small pilot study with a correspondingly small cohort.
Ethics committee - additional information: This was a retrospective analysis of data acquired prospectively as part of routine
clinical care, approved by the local Institutional Research Board.
The impact of whole-body MRI for staging unfavourable intermediate- and high-risk prostate cancer prior to curative-
intent surgery, radiotherapy, hormonal therapy: all-in-one MRI preliminary data (7 min)
Cristiano Michele Girlando; Milan / Italy
Author Block: C. M. Girlando1, S. Alessi1, Y. Zambanini1, S. Luzzago1, B. Frittoli2, S. Gambarini2, L. Grazioli2, G. Petralia1; 1Milan/IT,
2
Brescia/IT
Purpose: This is a multicentric, prospective interventional study, with the aim of comparing the accuracy of Whole-Body Magnetic
Resonance Imaging (WB-MRI) with that of conventional imaging (Computed Tomography and Bone Scintigraphy) for the systemic
staging of patients with unfavourable intermediate- or high-risk prostate cancer (PCa) and how WB-MRI staging results affect the
treatment decision.
Methods or Background: In two IRCCS centres, patients with unfavourable intermediate- or high-risk PCa were enrolled and
underwent systemic staging with conventional imaging and WB-MRI. Staging examinations were reported blinded to each other. A
first systemic staging and treatment decision was made by the multi-disciplinary team based on conventional imaging staging only
and, after the outcome of the WB-MRI, was formulated the definitive therapeutic proposal based on all available test results. We
compared the diagnostic accuracy of WB-MRI staging with that of conventional imaging staging and reported how WB-MRI staging
resulted in changes in treatment decision.
Results or Findings: This is a preliminary evaluation based on the first 80 subjects enrolled. Of these, 2 were excluded due to drop-
out. In the remaining 78 patients WB-MRI, compared to conventional imaging, led to an upstage in 19 patients (24.4%) and to a
downstage in 9 patients (11.5%), leading to a change in the therapeutic proposal in 17 patients (21.8%).
Conclusion: WB-MRI has shown superior diagnostic accuracy compared to conventional imaging in the systemic staging of patients
with unfavourable intermediate- or high-risk PCa with better risk stratification and more appropriate treatment proposals.
Funding for this study: Funding was received from the Italian Ministry of Health with Ricerca Corrente and 5x1000 funds.
Ethics committee - additional information: This study was approved by the Institutional Ethics Committee (UID 2080), and all
patients provided informed consent for their participation in the study in addition to that for the individual radiological procedures.
Whole-body MRI in staging of diffuse large B-cell lymphoma (7 min)

Monika Wagnerová; Praha / Czechia
199
Author Block: M. Wagnerová, P. Vodicka, K. Benesova, A. Bocan, D. Zogala, A. Burgetova, M. Trneny, L. Lambert; Prague/CZ
Purpose: In this study, we evaluated the diagnostic performance of whole body magnetic resonance imaging (wbMRI) in the staging
of diffuse large B-cell lymphoma and patients' preference for the staging modality.
Methods or Background: In this single-centre prospective study, adult patients diagnosed with DLBCL underwent wbMRI and 18-F
FDG PET/CT as the reference standard. Two radiologists evaluated wbMRI for nodal (12 nodal regions) and extranodal involvement.
Patients received a questionnaire about wbMRI.
Results or Findings: 14 of 78 patients consented to participate in and complete the study. The sensitivity and specificity of wbMRI
in the assessment of nodal involvement was 0.84 and 1.00. Extranodal involvement was apparent in all 14 instances. Staging was
concordant with PET/CT in all patients. The most frequent concern was the fear of an enclosed environment and the duration of the
examination.
Conclusion: Despite excellent sensitivity and specificity of wbMRI in staging of DLBCL, wbMRI is preferred less by the patients due to
the fear of an enclosed environment and the duration of the examination.
Limitations: There was a low number of patients willing to participate, and no evaluation of treatment response.
Funding for this study: Funding for this study was received from the Czech Ministry of Health (MH CZ-DRO, General University
Hospital in Prague - VFN, 00064165; VES NU21-03-00411) and from institutional funding at the Charles University in Prague
(Cooperatio, Medical Diagnostics and Basic Medical Sciences and Haematology-Oncology).
Ethics committee - additional information: This study was approved by the Ethics Committee of the General University Hospital
in Prague (12/20 Grant VFN IGP).
200
RPS 505 - Optimising radiological excellence: the synergy of AI and quality in radiology
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Physics in Medical Imaging
CME Credits: 1
Moderator:
Nico Buls; Brussels / Belgium
Assessing visual quality variability in deep learning reconstructed MRI: a focus on lesions (7 min)
Quintin Yves van Lohuizen; Groningen / Netherlands
Author Block: Q. Y. van Lohuizen1, S. Fransen1, C. Roest1, T. Kwee1, F. Simonis2, D. Yakar1, H. Huisman3; 1Groningen/NL,
2
Enschede/NL, 3Nijmegen/NL
Purpose: This study assessed the structural similarity index measure (SSIM), the leading metric for image quality, for its reliability in
evaluating deep learning reconstructed (DLRecon) MRI scans. The study particularly focused on images reconstructed using a state-
of-the-art DLRecon algorithm, targeting full transversal, prostate regions and lesion-specific fields of view (FOV) for SSIM assessment.
Methods or Background: A retrospective analysis was conducted using two datasets. The recurrent inference machine (RIM), a k-
space-based DLRecon algorithm, was trained on the public prostate NYU fastMRI k-space dataset (N=312) and externally validated
using prostate MRI scans of eight patients from the University Medical Center Groningen. Clinically significant prostate cancer lesions
(N=17) with PI-RADS scores from three to five were delineated by expert radiologists. Image quality was assessed using SSIM on
three FOVs: the full transversal FOV, the prostate, and the lesion level. Assessments were conducted at varying acceleration factors
from 2x to 8x.
Results or Findings: Significant differences in SSIM values were observed across FOVs in MRI scans reconstructed by the DLRecon
algorithm. Specifically, lesion FOVs had lower SSIM values (0.482± 0.057 at 8-fold acceleration) compared to the full transversal FOV
(0.905± 0.018) and the prostate FOV (0.870± 0.023). All differences were statistically significant (p< 0.001, Wilcoxon tests).
Furthermore, linear mixed-effects models revealed a significantly steeper rate of SSIM degradation in lesion-specific FOVs, suggesting
greater variability in image quality in these critical areas.
Conclusion: DLRecon algorithms like the RIM showed significantly lower SSIM values in lesion-specific FOVs compared to the full
transversal and prostate FOVs. This discrepancy calls into question the adequacy of SSIM as a standalone quality metric, emphasising
the need for more targeted quality assessments in future DLRecon algorithm development.
Funding for this study: Funding was provided by Health~Holland and Siemens Healthineers.
Image quality and sharpness improvement in coronary CT angiography using a deep-learning super-resolution
reconstruction algorithm: a phantom study (7 min)
Amir Pourmorteza; Atlanta, GA / United States
201
Author Block: T. W. Holmes1, S. Sharma2, S. Ross2, K. Schultz2, P. Gleason1, J. Schuzer2, Z. Yu2, R. Thompson2, A. Pourmorteza1;
1
Atlanta, GA/US, 2Vernon Hills, IL/US
Purpose: The purpose of this study was to investigate the performance of a super-resolution deep-learning-based reconstruction
(DLR) algorithm named precision image quality engine (PIQE) developed for cardiac CT against two clinical reconstruction algorithms:
adaptive iterative dose reduction (AIDR) and high-resolution DLR (AiCE).
Methods or Background: We 3D-printed inserts with microfluidic channels (d= 0.25 - 3.5 mm) with stents and calcified plaques
embedded and filled with dilutions of iodinated contrast agent. The inserts were placed inside a 12-cm diameter water tank and
scanned on a clinical CT scanner with prospective ECG-gating: 120 kVp, exposure: 25, 50, 250, and 400 mAs. Images were
reconstructed with matched parameters using AIDR, AiCE, and PIQE: 512 x 512 matrix, 0.312 x 0.312 x 0.5 mm3 voxel size. PIQE
images were also reconstructed with a 1024 x 1024 matrix and 0.156 x 0.156 x 0.5 mm3. We evaluated CT number stability, contrast-
to-noise ratio (CNR), and image sharpness as a function of radiation dose.
Results or Findings: CT number deviations from the 400 mAs baseline were measured in iodine, water, and fat inserts and were in
the [-1.1 3.1], [-1.1, 3.4], and [-2.2 0.26] for AIDR, AiCE, and PIQE, respectively. CNRs between iodine, water, fat (soft plaque), and
calcium (hard plaque), were between 36% to 97% higher for PIQE compared to AIDR, with maximum CNR improvement observed in
the lowest dose (25 mAs) scans. AiCE images showed a 0% - 37% increase in CNR in low-dose scans (25,50 mAs), however, their CNR
was between 11% to 27% lower for the higher-dose scans (400,250 mAs), compared to AIDR. MTF cutoff at 10% was 8.98, 10.68,
10.44, and 13.61 lp/cm for AIDR, AiCE, PIQE, and PIQE1024 respectively.
Conclusion: Overall, DLR algorithms improved CNR and image sharpness between 16%-18% at normal resolution voxel size.
Furthermore, PIQE improved image sharpness by 51% when reconstructed at high-resolution voxel size.
Limitations: More experiments mimicking different patient sizes are warranted.
Funding for this study: This study was sponsored research agreement with Canon Medical Research USA.
Elevating TIPSS procedures: AI denoising's impact on cone-beam CT image quality and diagnostic confidence (7 min)
Reza Dehdab; Tuebingen / Germany
Author Block: R. Dehdab, A. S. Brendlin, S. Afat, G. Grözinger, K. Nikolaou, J. Mück; Tuebingen/DE

Purpose: This study aimed to assess the utility of artificial intelligence (AI)-based denoising techniques in enhancing cone-beam
computed tomography (CBCT) images during transjugular intrahepatic portosystemic shunt (TIPSS) procedures, taking into account
variations in body mass index (BMI).
Methods or Background: A retrospective review of 60 patients undergoing TIPSS between 2016 and 2022 was approved. Patients
were categorised and paired based on BMI and divided according to image acquisition durations of three seconds (3s) and six seconds
(6s). CBCT images were processed with AI denoising (ClariCT.AI, ClariPI®). Image quality was quantitatively and qualitatively
evaluated, considering image noise, radiation dose, and Contrast-to-noise ratio (CNR) and diagnostic confidence. Relationships
between BMI, noise, CNR and radiation doses were also analysed.
Results or Findings: AI-based deep learning denoising (DLR) reduced image noise in 3s and 6s acquisition groups. Despite higher
initial noise in the 3s group, Post-DLR, noise equalised (3s-Original vs. DLR: T-Statistic: 57.06, p<0.001; 6s-Original vs. DLR: T-
Statistic: 41.11, p<0.001). Positive BMI-noise correlation occurred in 3s (r=0.564, P-value=0.006), not in 6s (r=0.3738, P=0.09). Both
groups showed a strong BMI-radiation dose correlation (3s- Correlation Coefficient (r): 0.681, p<0.001; 6s- Correlation Coefficient (r):
0.681, p<0.001). Diagnostic confidence was higher in 6s DLR (Kruskal-Wallis H Test: Statistic: 152.63, p<0.001), with 3s DLR
achieving comparable confidence to 6s originals (Mann-Whitney U Test: Statistic: 3943.5, p=0.83).
Conclusion: In conclusion, our study demonstrates that AI-enhanced denoising techniques in CBCT images during TIPSS procedures
not only significantly improve image quality but also reduce radiation exposure, highlighting the potential of AI in medical imaging for
optimising diagnostic accuracy while prioritising patient safety.
Limitations: The study's relatively small sample size and single-centre setting may limit the generalisability of the findings.
Funding for this study: No funding was received for this research.
Ethics committee - additional information: This study was retrospectively approved by the Institutional Review Board with a
waiver for the need for informed consent (#167/2022BO2).
3D human reconstruction algorithm: a novel technique for prospective image quality control in chest radiography (7
min)
Yuqi Tan; Chengdu / China
202
Author Block: Y. Tan, P. Wu, Z. Ye, Y. Hou, C. Xia, Z. Li; Chengdu/CN
Purpose: The purpose of this study was to propose a novel real-time 3D human reconstruction algorithm and investigate its
performance of identifying incorrect body postures and radiographer’s operations in chest radiography.
Methods or Background: A total of 83 chest post-anterior (PA) images and 71 chest lateral (LA) images shot by different
radiographers were included for this study. The 3D human reconstruction algorithm took a photo as input and output a 3D mesh
containing body morphology information, which was mainly based on a series of deep neural networks, including SMPL-X and HybrIK-
X. A camera fixed in front of the beam limiter was used to capture photos at the time of exposure. Automatic measurement tools were
developed for 3D human evaluation. Indexes including shrug (PA), scapula position (PA), arms up (LA), postures, exposure field, and
projection point were assessed in both subjective and 3D human evaluation. Subjective results were regarded as reference standard.
Sensitivity, specificity, and Kappa consistency of each index were calculated.
Results or Findings: In the chest PA of 3D human evaluation, the accuracy of identifying exposure field was 100%. The sensitivity,
specificity, and Kappa value of shrug, scapula position, postures, and projection point were 0.82, 0.92, 0.73; 0.86, 0.83, 0.69; 1, 0.93,
0.73; 0.54, 0.86, 0.42. In the chest LA of 3D human evaluation, the accuracy of identifying exposure field was also 100%. The
sensitivity, specificity and Kappa value of arms up, postures and projection point were 0.83, 0.94, 0.63; 0.95, 0.80, 0.65; 0.57, 1, 0.68.
Conclusion: 3D human reconstruction algorithm showed good ability in identifying incorrect body postures and radiographer’s
operations in chest radiography. Further improvement of this algorithm is needed to enhance its accuracy.
Limitations: The sample size was small and the measurements of projection point in 3D human evaluation needed modification.
Ethics committee - additional information: This study was retrospective, so did not require an ethics review.
MRI-radiomics for MGMT promoter methylation prediction in glioma: methodological quality, systematic review, and
meta-analysis (7 min)
Fabio Martino Doniselli; Milan / Italy
Author Block: F. M. Doniselli, R. Pascuzzo, M. Moscatelli, M. Grisoli, L. M. Sconfienza; Milan/IT

Purpose: This study aimed to evaluate the methodological quality and diagnostic accuracy of MRI-based radiomic studies predicting
O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status in gliomas.
Methods or Background: PubMed Medline, EMBASE, and Web of Science were searched to identify MRI-based radiomic studies on
MGMT methylation in gliomas published until December 31, 2022. Three raters evaluated the study methodological quality with
Radiomics Quality Score (RQS, 16 components) and Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis
Or Diagnosis (TRIPOD, 22 items) scales. Risk of bias and applicability concerns were assessed with QUADAS-2 tool. A meta-analysis
was performed to estimate the pooled area under the curve (AUC) and to assess inter-study heterogeneity.
Results or Findings: We included 26 studies. The median RQS total score was 8 out of 36 (22%, range: 8%-44%). Thirteen studies
performed external validation. All studies reported AUC or accuracy, but only 4 (15%) performed calibration and decision curve
analysis. No studies performed phantom analysis, cost-effectiveness analysis, and prospective validation. The overall TRIPOD
adherence score was between 50% and 70% in 16 studies and below 50% in 10 studies. The pooled AUC was 0.78 (95% CI: 0.73-0.83,
I2=93.4%) with a high inter-study heterogeneity. Studies with external validation and including only WHO-grade IV gliomas had
significantly lower AUC values (0.65, 95% CI: 0.57-0.73, p<0.01).
Conclusion: Study RQS and adherence to TRIPOD guidelines was generally low. Radiomic prediction of MGMT methylation status
showed great heterogeneity of results and lower performances in grade IV gliomas, which hinders its current implementation in
clinical practice.
Limitations: Some included studies did not report AUC values; grey literature was not included.
Ethics committee - additional information: This study was a meta-analysis.
Automatic uncertainty-based quality controlled segmentation of multimodal cardiac MR images (7 min)

Stéphanie Bricq; Dijon / France
203
Author Block: J. Michaud, T. W. Arega, S. Bricq; Dijon/FR
Purpose: Although deep learning-based segmentation methods have shown promise in automating the segmentation of cardiac MRI
images, they are not widely used in clinical practice due to the lack of robustness and reliability of the models. We propose an
uncertainty-based quality control (QC) framework to identify failed segmentations and to enhance the reliability of multimodal cardiac
MRI segmentation models.
Methods or Background: Automatic and accurate analysis of myocardial tissue characterisation is highly dependent on the quality
of the segmentation result. We proposed here an automatic quality controlled T1 mapping and LGE segmentation. The cardiac
structures were segmented from LGE, native and post-contrast T1 mapping images using a Bayesian Swin Transformer-based U-Net.
The quality of the segmentation output is assessed using uncertainty-based QC metrics. These uncertainty features are used as
inputs to a random forest-based classifier to evaluate the segmentation quality and reject bad segmentations. The proposed
framework was tested on a private cardiac MR dataset with various diseases.
Results or Findings: The proposed uncertainty-based quality control framework is robust in detecting inaccurate segmentations.
Proposed QC method achieves an area under the ROC curve (AUC) of 0.922 for native T1 images, 0.886 for post-contrast T1 images,
and 0.918 for LGE images on binary classification (bad or good segmentation).
Conclusion: The proposed framework automatically segments cardiac structures on multimodal MR images and rejects inaccurate
segmentation results. It can be applied to other segmentation methods to detect segmentation failures and to enhance the reliability
of the segmentation models.
Limitations: In addition to identify segmentation failures, it could be interesting to understand the underlying causes or sources
behind them. By gaining insights into why certain segmentation results are rejected, the model’s reliability could be enhanced.
Funding for this study: Funding was received from the French National Research Agency (ANR) with reference number: ANR-19-
CE45-0001-01-ACCECIT.
The performance of a commercial artificial intelligence algorithm in an external quality assurance scheme regularly
used by humans in the NHS breast screening programme (7 min)
Yan Chen; Nottingham / United Kingdom
Author Block: Y. Chen, A. Taib, I. Darker, J. James; Nottingham/UK

Purpose: The purpose of this study was to compare the performance of human readers in the National Health Service Breast
Screening Programme and a commercially available artificial intelligence (AI) algorithm when interpreting Personal Performance in
Mammographic Screening (PERFORMS) external quality assurance test sets.
Methods or Background: Two consecutive PERFORMS sets, each consisting of 60 challenging normal, benign, and biopsy proven
malignant mammography cases, were assessed by humans between 2018 and 2021 and AI in 2022. Suspicion of malignancy scores
were assigned to features detected. Each breast was considered separately, and the highest score was used to assess performance
using a pre-defined recall threshold. Sensitivity, specificity and ROC analysis was used to compare the performance of AI and human
readers retrospectively.
Results or Findings: 552 human readers interpreted both PERFORMS sets. There were 161 normal, 70 malignant, and 9 benign
breasts. There was no difference in the area under the receiver operating curve for AI and human readers (0.93 and 0.88 respectively,
p=0.15). At the developer suggested recall threshold, there was no difference in AI and human reader sensitivity (84% and 90%
respectively, p=0.34), but the specificity of the AI was higher than the human readers (89% and 76% respectively, p=0.003). Using
recall thresholds to match mean human reader performance (90% sensitivity and 76% specificity), the AI showed no difference in
performance with a sensitivity of 91% (p=0.73) and specificity of 77% (p=0.85).
Conclusion: Diagnostic performance of AI was comparable to that of the average human reader when evaluating cases from two
enriched test sets of the PERFORMS scheme.
Limitations: The PERFORMS sets consist of a small number of mammograms. These are not representative of typical screening
populations as they are enriched with challenging malignant cases.
Funding for this study: Funding for this study was received from Lunit Inc.
Ethics committee - additional information: Following discussion with the local research and development team, the requirement
for ethical approval was waived due to the retrospective nature of the study.
204
CUBE 6 - How to improve outcomes?

Advanced Session - Topic Coordinator: Prof. Gianpaolo Carrafiello
The “Advanced Session: Percutaneous Interventions” is aimed at a more advanced audience and covers percutaneous interventions
in various areas of interventional radiology.
1. To learn about new tools of the trade.

2. To understand how to use them.
3. To discuss the advantages and improvements in clinical practice.
Moderator:
Gianpaolo Carrafiello; Milan / Italy

Imaging intra-operatorio: IVUS and perfusion angiography (14 min)

Matteo Stefanini; Roma / Italy
New tools and innovations in peripheral endovascular treatment (14 min)

Chiara Floridi; Perugia / Italy
205
E³ 626 - AI benefits and limits
Categories: Abdominal Viscera, Artificial Intelligence & Machine Learning, Musculoskeletal, Oncologic Imaging, Paediatric
CME Credits: 1
Moderator:
Susan Cheng Shelmerdine; London / United Kingdom

Susan Cheng Shelmerdine; London / United Kingdom
Latest updates on AI for paediatric musculoskeletal diseases (10 min)

Amaka Cynthia Offiah; Sheffield / United Kingdom
1. To understand the current spectrum of available AI applications for paediatric musculoskeletal radiology.
2. To learn about the emerging AI applications within paediatric musculoskeletal radiology.
3. To discuss the future directions of AI in paediatric musculoskeletal radiology.
Latest updates on AI for paediatric abdominal imaging (10 min)

Latest updates on AI for paediatric oncological imaging (10 min)

1. To describe the main roles of AI software in image preparation and labelling.

2. To discuss the AI approaches to organ and tumour segmentation.
3. To evaluate the AI improvements in the decision support tools regarding children's management.
Developing an AI tool for hip dysplasia US (10 min)

Thomas L. A. van den Heuvel; Nijmegen / Netherlands
1. To understand the use of AI in combination with ultrasound for developmental hip dysplasia.
Panel discussion: Which kind of radiology-based AI tool/use case would bring the most promise to paediatric
healthcare? (15 min)
206
RC 614 - Redeveloping the role of the radiographer in clinical audit
Categories: EuroSafe Imaging/Radiation Protection, Management/Leadership, Professional Issues, Radiographers

CME Credits: 1
Moderators:
Eija Metsälä; Helsinki / Finland
Patricia Cunningham; Dunshaughlin / Ireland

Eija Metsälä; Helsinki / Finland
Patricia Cunningham; Dunshaughlin / Ireland
Requirements and best practices for radiographer-led clinical audit (15 min)
Matteo Migliorini; FERRARA / Italy
1. To name and identify the types of audits and the clinical audit cycle.
2. To list the essential elements and requirements to design an audit question and define an audit standard.
3. To describe and reflect on audit reports and dissemination.
How to develop an effective clinical audit programme (15 min)

Karen Borg Grima; Naxxar / Malta
1. To demonstrate how to compile a clinical audit programme.

2. To examine the reviewing process and practical steps involved in setting up the clinical audit programme.
3. To discuss how to monitor the clinical audit programme once it is established.
Clinical governance in radiology departments: is audit enough? (15 min)

Theophilus N Akudjedu; Bournemouth / United Kingdom
1. To identify the key clinical governance practices for quality improvement across radiology departments.
2. To critique a case study in relation to clinical audit and research in radiography practice.
3. To reflect on components of their practice that are informed by quality improvement activities, including research and audit.
Panel discussion: What will clinical audit need to look like in the future to be fit for purpose? (10 min)
207
OF 6T - The impact of emerging technologies on radiology education and practice
Categories: Artificial Intelligence & Machine Learning, Education, Imaging Methods, Interventional Radiology, Students
CME Credits: 1
Moderator:
Edith Vassallo; Imsida / Malta

How will artificial intelligence affect radiology? (15 min)

1. To explain the basic concepts of AI related to imaging.

2. To appreciate how AI will impact radiology practice through case examples.
How will virtual reality and simulation affect radiology? (15 min)
Jim Zhong; Leeds / United Kingdom
1. To explain the basic concepts of virtual reality and simulation.

2. To explore how these will impact the radiology practice through case examples.
What will interventional radiology of the future look like? (15 min)
1. To highlight the new technologies in interventional radiology.

2. To appreciate how these technologies will be applied through case examples.
Panel discussion: What does the future of radiology look like? (10 min)
208
RC 604 - A road less travelled: imaging the tracheobronchial tree
Categories: Chest
CME Credits: 1
Moderator:
Benoit Ghaye; Brussels / Belgium

Benoit Ghaye; Brussels / Belgium
Diseases involving the trachea (15 min)

Guillaume Chassagnon; Paris / France
1. To identify main tracheal lesions.

2. To name the main causes of tracheal stenosis.
3. To differentiate between the main causes of tracheal thickening.
Bronchial and bronchiolar diseases (15 min)

Arjun Nair; London / United Kingdom
Assessment of emphysema (15 min)

Mariaelena Occhipinti; Pisa / Italy
1. To recognise emphysema subtypes as defined on CT.

2. To compare and contrast the appearance of emphysema according to different acquisition and reconstruction algorithms, including
at ultra-high spatial resolution on PCCT.
3. To reflect on the importance of quantitative imaging and CXR in the assessment of COPD patients.
Panel discussion: Potential use of quantification in diseases of the tracheobronchial tree (10 min)
209
OF 6R - Embracing generations: empowering elderly care in radiology
Categories: Education, Imaging Methods, Management/Leadership, Professional Issues, Radiographers

CME Credits: 1
This session is dedicated to the critical intersection of radiology and elderly patient care. In an era characterised by an ageing
population, it is essential to address the unique challenges and opportunities that radiological practice offers in the realm of elder
care. This session will feature three compelling talks that collectively contribute to a holistic understanding of the specialised care
needed when imaging elderly patients. Attendees of this session will benefit from a better understanding of the critical importance of
empowering radiological practice to meet the evolving needs of our ageing population. The talks will delve into practical solutions and
best practices that bridge the generations, ensuring that elderly patients receive the compassionate and effective care they deserve
in the realm of radiology and radiotherapy.
Moderator:
Jose Miguel Saude; Porto / Portugal

Jose Miguel Saude; Porto / Portugal
Tips to enhance the experience of radiological examinations in patients with cognitive impairment (16 min)
Overcoming image quality challenges when performing radiography in frail elderly patients (16 min)
Elona Dybeli; Elbasan / Albania
Strategies to improve communication when imaging elderly patients (16 min)

Uffe Lindberg Lindberg Wewer Jakobsen; Odense NV / Denmark
210
IF 6 - Research in radiology: how much do we contribute to population health?
Categories: Education, Imaging Methods, Management/Leadership, Multidisciplinary, Research

CME Credits: 1
Radiology has always been an interdisciplinary speciality and was often crucial to progressing in treating and managing various
conditions. Radiological screening - e.g., breast and lung cancer - has been established to improve overall population health. This
session will discuss how research in radiology has impacted many patients' lives and how it contributed to population health. Past and
current high-impact studies will be highlighted, and new frontiers will be explored on a European level.
Moderator:
Martina Pecoraro; Rome / Italy

Clinical trials in radiology: the impact on population health (15 min)

EU-funded radiology research: what can we expect for the future? (15 min)
Monika Hierath; Vienna / Austria
The RTF perspective on radiology research and population health (15 min)
Panel discussion: Research in radiology: where can we have the biggest impact? (10 min)
211
ETAP - Why choose an ETAP-certified centre for radiology training?
Categories: Education, Management/Leadership

CME Credits: 1
Moderator:
Oğuz Dicle; Izmir / Turkey

Oğuz Dicle; Izmir / Turkey
Why certify your training programme? (15 min)

Jussi Hirvonen; Turku / Finland
1. To learn the primary motivation of a centre to apply for ETAP.

2. To appreciate the importance of the ETAP certificate for the centre's reputation.
3. To understand the benefits of the assessment.
Can the assessment certificate enhance the selection of a radiology residency programme? (15 min)
Hanna Maria Nikkila; Tampere / Finland
1. To learn the importance of an assessment certificate when selecting a radiology training programme.
2. To appreciate the relevance of the certificate when enrolling into the programme.
3. To understand what the certificate says about the centre.
Practical tips and key concepts in the assessment process (15 min)
Dean Huang; London / United Kingdom
1. To learn how to prepare for a training program assessment.

2. To appreciate the importance of preparation efforts.
3. To understand the format and structure of the assessment.
Panel discussion: Why is the ETAP certificate beneficial for a radiology training programme? (13 min)
212
RC 610 - Imaging hip joint replacement

CME Credits: 1
Moderator:
Alberto Bazzocchi; Bologna / Italy

Radiography and ultrasound: how far can you go? (15 min)
Davide Orlandi; GENOVA / Italy
1. To differentiate the strengths and weaknesses of conventional radiography in assessing the structural integrity of a hip implant
(THA).
2. To show the correct ultrasound examination technique of soft tissues around THA, including assessment of periprosthetic
pathologic conditions such as inflammatory pseudotumor, infections, and soft tissue impingement.
3. To realise the real-time capabilities of ultrasound, providing a valuable dynamic assessment of hip muscles and tendons functional
status and furnishing an excellent tool for the guidance of diagnostic and therapeutic interventional procedures, such as
periprosthetic fluid collection aspiration
and postoperative hematoma drainage.
CT: when should you do it, and how? (15 min)

Tobias Dietrich; Romanshorn / Switzerland
1. To present CT techniques of metal artefact reduction in hip joint replacement.

2. To show indications of CT imaging in hip joint replacement.
3. To describe hip joint replacement disorders on CT images.
MRI: can it replace the above? (15 min)

Amanda Isaac; London / United Kingdom
Panel discussion: Can we define an algorithm for the assessment of painful hip replacement? (10 min)
213
BS 6 - Imaging of the upper abdominal viscera variants and congenital anomalies: from current to rare
aspects
Categories: Abdominal Viscera, Education, Imaging Methods

ETC Level: LEVEL II
CME Credits: 1
Moderator:
Ioana Gabriela Lupescu; Bucharest / Romania

Liver vascularisation variants and their clinical implications (15 min)

Raffaella Basilico; Chieti / Italy
1. To learn the most frequent anatomical variants regarding arterial and venous vascularisations.
2. To appreciate the clinical implications in hepatobiliary surgery, transplantation and interventional radiology of the liver variants.
3. To present clinical cases of rare anatomical liver variants.
Anatomical variants and malformations of the biliary tree (15 min)

Razvan Capsa; Bucuresti / Romania
1. To present the most common anatomical variants and malformations of the biliary tree.
2. To assess the role of non-invasive imaging in the diagnosis of these findings.
3. To emphasize the role of different MRI sequences and protocols in bile duct assessment and potential challenges.
Pancreas and spleen variants and malformations (15 min)

Marc Zins; Paris / France
1. To identify the most frequent anatomic variants and congenital abnormalities of the pancreas and spleen,
and discuss the appropriate indications for imaging.
2. To summarize the imaging features of the most common anatomic variants and congenital abnormalities of the pancreas and
spleen.
3. To understand the critical role of imaging in the diagnosis, treatment, and management of these patients.
Panel discussion: Structured report and importance of the multidisciplinary team (10 min)
214

CME Credits: 1


215
RC 605 - Quality control and post-market surveillance of AI medical software
Categories: Artificial Intelligence & Machine Learning, Evidence-Based Imaging, Imaging Informatics, Multidisciplinary
CME Credits: 1
Moderator:
Charlotte Brouwer; Groningen / Netherlands

Post-market surveillance of a software as a medical device (15 min)

Kicky Gerhilde van Leeuwen; De Bilt / Netherlands
1. To learn about the need for post-market surveillance.

2. To understand how to perform post-market surveillance on AI applications.
3. To learn how to establish a fruitful industry-hospital collaboration for post-market surveillance of AI tools.
How to confirm the effectiveness and safety of the use of an AI medical software? (15 min)
Emilia Niemiec; Copenhagen / Denmark
1. To learn how the safety and performance of an AI software are defined following the MDR.
2. To understand how to identify key performance indicators to assess safety and performance in clinical settings.
3. To review examples of safety and performance metrics of clinically in-use AI tools.
An example of quality control of AI applications in imaging (15 min)

Federica Zanca; Leuven / Belgium
1. To learn about a possible approach for real-time tracking of AI applications.

2. To learn about operational and clinical key performance indicators for tracking selected AI applications.
3. To learn about typical errors related to AI applications implemented in clinical settings.
Panel discussion: How to implement quality control and post-market surveillance of AI tools in daily practice? (10 min)
216
HW 6R - Practical dose management in computed tomography
Categories: Contrast Media, EuroSafe Imaging/Radiation Protection, Radiographers

CME Credits: 1

1. To describe best practices, tips and tricks for reducing contrast media dose and radiation exposure in CT using computer software.
2. To demonstrate and apply best practices, tips and tricks for reducing contrast media dose and radiation exposure in CT using
computer software.
Reducing CT radiation dose using dose management software: presentation, live software demonstration and Q&A (30
min)
Michiel Heynderickx; Sint-Niklaas / Belgium
Software opportunities for optimising CT contrast media dose: presentation, live software demonstration, and Q&A (30
min)
Cristian Colmo; Padova / Italy
217
RPS 605 - Impact of AI on diagnosis of prostate cancer
Categories: Artificial Intelligence & Machine Learning, Genitourinary, Oncologic Imaging

CME Credits: 1
Moderator:
Marie-France Bellin; Le Kremlin Bicêtre / France
Artificial intelligence and radiologists at prostate cancer detection in MRI: outcomes of the PI-CAI challenge (7 min)
Anindo Saha; Nijmegen / Netherlands
Author Block: A. Saha, J. Bosma, J. J. Twilt, D. Yakar, M. Elschot, J. Veltman, J. Fütterer, M. De Rooij, H. Huisman; Nijmegen/NL
Purpose: Diagnostic performance of AI systems at detecting clinically significant prostate cancer in MRI, in comparison to radiologists
using PI-RADS v2.1, has not been studied at scale. Autonomous AI systems can alleviate the increasing demand in medical imaging
and reduce overdiagnosis in prostate cancer management.
Methods or Background: We trained, tuned, and tested an independently developed AI system at detecting Gleason grade group
≥2 prostate cancer, using a retrospective cohort of 10,207 MRI examinations (9129 patients) from four European tertiary care
centres. In parallel, we facilitated an observer study with 62 radiologists (45 centres, 20 countries) and 400 testing cases. Reference
standard was histopathology and ≥3 years of follow-up. Our study design was established and preregistered with 16 international
multidisciplinary experts.
Results or Findings: In the subset of 400 testing cases that was used to facilitate the observer study, on average, the AI system
demonstrated superior diagnostic performance with an AUROC of 0.91 (95% CI, 0.87-0.94), than the pool of 62 radiologists at PI-RADS
v2.1 with an AUROC of 0.86 (95% CI, 0.83-0.89). In all 1000 testing cases, the AI system showed marginally lower specificity of 68.9%
(95% CI, 65.3-72.4%) than the standard of care during routine practice with a specificity of 69.0% (95% CI, 65.5-72.5%), when
thresholded to match the same sensitivity of 96.1% (95% CI, 94.0-98.2%) as the PI-RADS ≥3 operating point.
Conclusion: An AI system, trained on thousands of cases, is superior in differentiating significant prostate cancer at MRI in
comparison to radiologists at PI-RADS v2.1, but marginally less specific in comparison to the standard of care in routine practice.
Limitations: The study utilised a retrospective design and was based on histologic verification guided by routine practice. There was
an absence of intercontinental, multi-ethnic patient data and MRI examinations from all major commercial vendors.
Funding for this study: This study received funding from the EU Horizon 2020: ProCAncer-I (grant number 952159),
Health~Holland (grant number LSHM20103).
Ethics committee - additional information: This study was approved by institutional or regional review boards at each
contributing center (identifiers: REK 2017/576; CMO 2016-3045; IRB 2018-597; ZGT23-37), and was conducted in accordance with the
principles of the Declaration of Helsinki. Informed consent was exempted, given the retrospective scientific use of deidentified MRI
scans and clinical data.
Data integration using AI, PI-RADS, and clinical data to reduce false positives in prostate MRI (7 min)
Antony William Rix; Cambridge / United Kingdom
218
Author Block: A. W. Rix1, P. Burn2, N. Vasdev3, A. Bradley4, A. Andreou5, J. Aning6, T. Barrett1, A. R. Padhani7, A. Shah8; 1Cambridge/UK,
2
Taunton/UK, 3Stevenage/UK, 4Truro/UK, 5Bath/UK, 6Bristol/UK, 7Northwood/UK, 8Winchester/UK
Purpose: This study aimed to determine how multi-modal decision support models, integrating clinical data, PI-RADS, and AI, could
help optimise patient selection for biopsy following MRI for suspected prostate cancer.
Methods or Background: Clinical history, MRI, PI-RADS, and histopathology data were obtained retrospectively from a five-site,
multi-vendor study of a diagnostic patient population. 352 patients were assigned for model training/ tuning, and 235 patients (Grade
Group≥2 prevalence 34%) for held-out testing. GG≥2 cancer was verified by standard-of-care MRI-directed biopsy. Patients scored PI-
RADS 1/2 without biopsy were considered negative. Automated AI-based software that identifies and scores patients/ lesions for risk
of GG≥2 was separately trained using the same training data. Multi-modal machine learning models were trained for combinations of
AI scores, clinical variables including PSA-density (PSAD), and the original reporting radiologists’ PI-RADS scores. Sensitivity,
specificity, and AUC were compared per-patient on the held-out test data with the PI-RADS assessments and AI scores alone.
Results or Findings: The original PI-RADS scores identified GG≥2 patients with sensitivity 1.00 (95% CI 1.00-1.00), specificity 0.67
(0.61-0.75) and AUC 0.94 (0.91-0.97). AI detected GG≥2 patients with sensitivity 0.97 (0.93-1.00), specificity 0.55 (0.47-0.62) and
AUC 0.88 (0.84-0.92) using bpMRI data. Combining AI scores and PSAD based on TZ volume (TZ-PSAD) gave sensitivity 0.95
(0.90-0.99, p<0.001), specificity 0.70 (0.63-0.77, p<0.001) and AUC 0.90 (0.85-0.93, p=0.25). Combining PI-RADS, AI, and TZ-PSAD
gave sensitivity 0.99 (0.96-1.00, p<0.001), specificity 0.83 (0.77-0.89, p<0.001), and AUC 0.96 (0.93-0.98, p=0.003). TZ-PSAD gave
slightly better AUC than whole-prostate PSAD. Other clinical variables had no statistically significant benefit. Findings with bpMRI and
mpMRI AI were similar.
Conclusion: Decision support models combining PI-RADS, AI scores, and PSAD could significantly reduce false positive biopsies while
maintaining sensitivity, compared to AI or PI-RADS assessments alone.
Limitations: This study used standard-of-care limited biopsy for the ground truth.
Funding for this study: Funding was received from Lucida Medical.
Ethics committee - additional information: This study was approved with the UK HRA IRAS number: 278640.
DL algorithm for MRI prostate volume can automatically tailor the threshold of PSA density in combination with other
risk factors for the prediction of clinically significant PCa (7 min)
Alessandro Venturi; Florence / Italy
1 2 1 1 1 1 1 1 2
Author Block: A. Venturi , A. Colarieti , D. Fazzini , M. Interlenghi , E. Schiavon , M. Alì , I. Castiglioni , S. Papa , F. Sardanelli ;
1 2
Milan/IT, San Donato Milanese, Milan/IT
Purpose: The aim of our study was to determine the optimal threshold for Prostate-Specific Antigen density (PSAd) when the prostate
volume is automatically computed by a deep learning (DL) algorithm on T2-weighted MRI images as a contouring method. This
standardised, repeatable, and reliable predictor was then combined with other predictors of clinically significant prostate cancer
(csPCa).
Methods or Background: We conducted a multicentric retrospective study, including patients assessed by mpMRI prior to prostate
biopsy. csPCa was defined as a PCa with any ISUP grade group ≥2 (Gleason ≥3+4). We trained a U-Net based DL algorithm on T2-
weighted images and tested by Dice Similarity Coefficient (DSC) in comparison with three board certified radiologists that segmented
the prostate slice-by-slice blinded to each other. Twenty repetitions were performed.
Results or Findings: We included 279 patients, aged 65.5±8.0 years. The developed DL algorithm achieved a reliability (DSC) of
0.86. Repeatability was 100%. The computed PSAd ranged from 0.02−2.36ng/ml/cm3. A PSAd threshold of 0.10ng/ml/cm3 showed the
best balanced sensitivity/ specificity of 0.66/ 0.64, respectively, on an external dataset of 86 patients. However, when combined with
patient age, a PSAd threshold of 0.11ng/ml/cm3 and an age threshold of 67 improved sensitivity up to 0.84, without affecting the
specificity.
Conclusion: Our results showed how PSAd threshold can be obtained by an automatic DL algorithm applied on T2-weighted images,
considering the slice-by-slice prostate volume (i.e. not based on geometric approximations, such as ellipsoid diameters), and
specifically optimised in combination with patient age. The inclusion of radiomics features from T2-weighted and DWI could allow a
further specific optimisation of PSAd threshold.
Limitations: These patient cohorts were collected exclusively by Italian centres.
Ethics committee - additional information: This study was approved by the Ethics Committees of relevant centres.
Impact of AI on the diagnosis of prostate cancer with mprMRI for novice radiologists: results of a single-centre study (7
min)
Tommaso Russo; Milan / Italy
219
Author Block: T. Russo, G. Brembilla, E. Camisassa, M. Cosenza, R. Pennella, L. Quarta, G. Gandaglia, A. Briganti, F. De Cobelli;
Milan/IT
Purpose: The purpose of this study was to assess the impact of a commercial AI software (Quantib Prostate) on the diagnostic
accuracy and interreader agreement of novice radiologists in the interpretation of multiparametric MRI of the prostate (mpMRI).
Methods or Background: Accurate interpretation of mpMRI of the prostate requires training on large case series and is affected by
interreader variability. AI software have been developed to overcome these limitations and assist radiologists in evaluating mpMRI.
This is a single-centre retrospective study on 110 patients who underwent mpMRI for clinical suspicion of PCa (+/- targeted biopsy) at
a single center. All mpMRIs were reviewed by three novice readers (radiology residents; Reader 1, 2, 3 – R1, R2, R3) with four years
(R1 and R2) and one year (R3) experience in body imaging. All MRI exams were interpreted and reported in a sequential fashion: first,
radiologists interpreted the exam without AI assistance; then, they were unblinded to AI results and re-reported the MRI exam.
Histopathological results from MRI-targeted and concomitant systematic biopsies were considered the standard of reference; clinically
significant PCa (csPCa) was considered ISUP>1. The primary objective was to compare the diagnostic accuracy of the readers without
and with AI assistance.
Results or Findings: 61% (67/110) of patients had any PCa (ISUP≥1), and 43% (48/110) had csPCa (ISUP≥2). The diagnostic
performance of R1 and R2 remained similar with and without Quantib Prostate. R3’s sensitivity and overall accuracy for csPCa
improved from 81% and 55% to 91% and 60%, respectively. Percentage of interreader agreement was 74% (IC 0.684 to 0.807)
without Quantib and 73% (IC 0.678 - 0.801) with Quantib.
Conclusion: AI-based software (Quantib Prostate) may improve the diagnostic accuracy of novice radiologists for identifying csPCa.
Limitations: This was a retrospective single-centre study.
Ethics committee - additional information: This study was approved by an ethics committee.
Multireader evaluation of a deep learning computer-aided system for prostate MRI in men with prostate cancer being
considered for active surveillance (7 min)
Laura Isabel Loebelenz; Bern / Switzerland
Author Block: L. I. Loebelenz1, A. Samani2, A. Azam2, D. Manea3, D. Prezzi2, A. Sharkey2, O. Williams2, S. J. Withey2, V. Goh2; 1Bern/CH,
2
London/UK, 3Iasi/RO
Purpose: The Prostate Imaging-Reporting and Data System (PI-RADS) has standardised practice but variation in radiologist reporting
performance remains an issue. Artificial intelligence (AI) may improve this. We aimed to evaluate the impact of a commercial deep
learning (DL) software across readers of different experience in men with low-intermediate risk prostate cancer being considered for
active surveillance.
Methods or Background: This retrospective study included men with low-intermediate risk prostate cancer. Five readers with
varying levels of experience (<1 year to ≥5 years of experience), trained in three different countries, evaluated the initial bi-
parametric prostate MRI, with and without DL-assistance in a randomised design. PI-RADSv2.1 scores were recorded and compared
between reads and between readers, and against ground truth using metrics including area under the receiver operating
characteristics curve (AUC). Fleiss-Kappa analysis was performed for interreader agreement. Radiological ground-truth was
independent expert scoring/annotation of focal lesions. Histological ground-truth was the International Society of Urological
Pathologists grade group (GG) score.
Results or Findings: 100 men were included with mean age 61±7 years and mean PSA density 0.15±0.09 (SD). There were 23 ISUP
GG ≥2 cancers on histology. At an individual-reader level, for PI-RADS scoring compared to the radiological ground truth, AUC ranged
from 0.65 to 0.80. DL-assistance increased AUC, although the magnitude of benefit varied across the reader pool, AUC ranging from
0.69 to 0.82. Additionally, DL-assistance appeared to reduce interreader variability. Reader agreement (weighted kappa) ranged from
0.24 to 0.56 without DL-assistance, compared to 0.45-0.55 with DL-assistance.
Conclusion: AI can improve performance for PI-RADS scoring particularly for non-expert readers in this cohort, and may also reduce
variability in reader performance.
Limitations: This was a single centre, retrospective study.
Funding for this study: This study received no direct funding.
Ethics committee - additional information: This study was approved with the code: 18/NW/0297.
Assessing radiomic stability: impact of annotation variability on radiomics features consistency in different anatomical
regions (7 min)
Carmen Prieto-de-la-Lastra; Majadahonda / Spain
220
Author Block: C. Prieto-de-la-Lastra1, A. Jimenez-Pastor2, A. Picó Peris2, D. Veiga Canuto2, L. Marti-Bonmati2; 1Madrid/ES, 2Valencia/ES
Purpose: Radiomic features are calculated from delineated regions of interest (ROIs), characterising the patient and anatomical
region where they have been calculated. Therefore, the segmentation quality can substantially impact the power of radiomics. In this
study, the discrepancies among different segmentations of the same ROIs are compared to analyse radiomics stability in different
anatomical regions.
Methods or Background: Two datasets were inspected, each with different annotations: 100 MRI studies with the prostate gland
(central and peripheral) and seminal vesicles segmented; and 960 MRI scans with annotated neuroblastic lesions. The original
segmentations were modified with dilations and erosions of structuring element (SE) equal to 1, 2 or 3 voxels, simulating the
annotations from different radiologists. Therefore, 7 segmentations were generated for each case. 1015 radiomic features were
calculated from each mask. The distributions of characteristics across annotations were compared through the Wilcoxon-test, paired
and non-paired. The correlation among the different simulated annotators was analysed with the intraclass correlation index (ICC).
Finally, the most stable variables maintained across all the experiments were inspected.
Results or Findings: In the central prostate gland, the number of stable variables from Wilcoxon analysis was 165, 8 from Wilcoxon-
paired evaluation and 743 from ICC tests. According to the peripheral prostate gland, the results were 40, 0, and 406. In the seminal
vesicles analysis, the results were 107, 3, and 514. Finally, the neuroblastoma dataset resulted in 34, 0, and 637 stable variables from
each of the experiments, respectively. Furthermore, in both datasets, the number of stable features decreased as the size of the SE
increased.
Conclusion: Radiomics is less stable when annotations highly differ from the original ROIs, being more susceptible to sharper and
irregular shapes as the peripheral gland and cancer lesions.
Funding for this study: Funding was received from PRIMAGE (PRedictive In-silico Multiscale Analytics to support cancer
personalised diagnosis and prognosis, empowered by imaging biomarkers), a Horizon 2020|RIA project (Topic SC1-DTH-07-2018),
grant agreement no: 826494.
Ethics committee - additional information: Thi study was approved by the Institutional Review Board and written informed
consent was waived from all participants.
Deep learning-based risk estimation for personalised follow-up in low-risk prostate cancer surveillance (7 min)
Christian Roest; Groningen / Netherlands
Author Block: C. Roest1, T. Kwee1, P. Van Leeuwen2, H. Huisman3, D. Yakar1; 1Groningen/NL, 2Amsterdam/NL, 3Nijmegen/NL
Purpose: Timely follow-up in active surveillance of low-risk prostate cancer (PCa) is crucial for early detection of disease progression
and to minimise overuse of diagnostics. MRI-based deep learning (DL) may optimise follow-up timing by estimating progression risk.
Methods or Background: This multi-centre study included 1607 MRI scans of 1143 men undergoing MRI for suspicion of harbouring
clinically significant (cs, defined as International Society of Urological Pathology>1) PCa, who were negative for csPCa at the time of
the MRI scan. A novel DL model was developed, which used MRI and routine clinical parameters to predict the risk of PCa progression
(defined as csPCa at follow-up). The model was internally cross-validated in 829 exams, and externally validated in 778 exams. Cox-
regression assessed whether the model predicted risk of progression. Time-dependent receiver-operating characteristic curve
analysis was used to compare our proposed model to established risk estimation tools (European Randomised study of Screening for
Prostate Cancer [ERSPC], Prostate Cancer Prevention Trial risk calculators [PCPT]) and PI-RADS. The area-under-the-curve was
calculated five years after MRI. Optimized follow-up intervals were derived from Kaplan-Meier curves.
Results or Findings: DL scores predicted progression (internal: hazard-ratio [HR] 14.01, CI 6.61-30.65; p<0.001; external: HR 16.21,
CI 3.48-75.5; p<0.001). DL achieved the highest area-under-the-curve in internal (0.75, CI 0.66-0.85) and external cohorts (0.7, CI
0.64-0.76). Internally, DL outperformed ERSPC (p=0.002) and PI-RADS (p=0.006). Externally, DL outperformed ERSPC (p=0.02) and
PCPT scores (p<0.001). On internal validation, DL identified a 20% stratum of very-low-risk PCa with <10% risk of missed progression
after 3.5 year follow-up, and <15% risk on external validation.
Conclusion: Our proposed DL model provided more accurate risk estimations compared to established methods. DL risk scores may
help to personalise follow-up protocols for low-risk PCa.
Funding for this study: Funding was provided by a grant from Siemens-Healthineers.
Ethics committee - additional information: This was a retrospective study.
Development and external validation of PSMA-PET/MR based radiomics models to predict Gleason score in prostate
cancer (7 min)
Tianshuo Yang; Huai'an / China
221
Author Block: T. Yang1, W. Tao1, Y. Song1, J. Zhang2, X. Niu1, G. Fu1, G. Bai1; 1Huai'an/CN, 2Taizhou/CN
Purpose: This research aims to establish radiomics models based on PSMA PET/MR scans with external validation to predict GS of
PCa.
Methods or Background: A total of 192 PCa patients were enrolled in this study, including 160 patients in the internal validation set
(Centre A) and 32 ones in the external validation set (Centre B and Centre C). PET/MR scans were performed prior to clinical
treatment, and three kinds of radiopharmaceuticals ((18F)-PSMA-1007, (68Ga)Ga-PSMA-11 and (Al18F)-PSMA-BCH) were randomly
applied in the PSMA-targeted PET examinations. The patients were divided into the low-risk group (GS≤7) and the high-risk group
(GS>7). 1409 high-throughput features were extracted from each ROI and selected using the LASSO algorithm. Radiomics models
were constructed based on the above selected features through machine learning algorithm of LR, NB, RF, SVM, and XGBoost through
30 times 4-fold repeated cross-validation, respectively. The performance of every model was evaluated through the ROC curve. The
optimal algorithm and radiomics model were chosen according to the AUC value.
Results or Findings: 12-14 radiomics features and NB algorithm were selected to radiomics modelsʼ establishment. In the external
validation set, the models based on PSMA-PET, T2WI, and ADC maps exhibited stable predictive performance with AUC values of
0.762, 0.698, and 0.668 (75.0%, 65.6%, and 71.9% accuracy).
Conclusion: Our study demonstrated that PSMA-targeted PET-based radiomics model occupied better performance in the GS
prediction than those based on T2WI and ADC through the external cohort validation. Radiomics model of PSMA-targeted PET could be
utilised to predict PCa prognosis noninvasively and help clinicians make individualised treatment plans for patients.
Limitations: The number of patients in our study was small, and further large-scale data research from multiple centres will be
conducted in the future.
Funding for this study: This study received funding from the Huai’an Science and Technology Project (grant no. HAB202017 to WT).
Ethics committee - additional information: This study was approved by the Ethics Committee of The Affiliated Huaian No.1
People’s Hospital of Nanjing Medical University (Date 2021.12.22 / No YX-2021-113-01).
222
RPS 611 - Brain tumour imaging: advanced techniques
Categories: Molecular Imaging, Neuro, Oncologic Imaging, Research

CME Credits: 1
Moderator:
Jasmina Boban; Novi Sad / Serbia
Amide proton transfer-weighted imaging (APT-wi) for molecular and survival prediction of diffuse adult gliomas:
single-centre retrospective study (7 min)
Author Block: F. M. Doniselli, M. Moscatelli, M. Verri, R. Pascuzzo, M. Grisoli; Milan/IT

Purpose: The aim of this study was to observe the correlation between APT signal characteristics of low- and high-grade glial
neoplasms and the demographic characteristics, mutational status and grading, using first-order statistics.
Methods or Background: We retrospectively included patients treated for brain tumor in our centre between October 2020 and
October 2022 if they matched our set criteria, which were the presence of analysable pre-operative APT-weighted imaging (APT-wi),
or a diagnosis of a glioma of any grade. Volumetric ROI on the whole lesion (“tum_ROI”), on the whole lesion excluding the necrotic
component (“les_ROI”) and on necrotic component alone - if present (“nec_ROI”) - were placed; from each ROI, the mean, median,
standard deviation, 10th percentile, 90th percentile, skewness, and kurtosis of APT values were extracted. One-year survival, IDH,
grade, and MGMT promoter methylation were recorded.
Results or Findings: 52 patients (median age 56 years, IQR 25-76, 35 males; 5 astrocytoma grade 2, 8 grade 3 astrocytoma and 39
glioblastoma) were included. Median tumour volume was 40.7cm3 (0.38-167.6). Mean APT values in the les_ROI ranged
approximately between 1 and 3 while mean APT values in tum_ROI ranged between 1 and 7. Mean and 90th percentile APT values in
les_ROI was statistically different between grade 4 and LGG (p=0.042 and p=0.011 respectively) and for 1 year survival (p=0.011 and
p=0.012). No statistical significance was found between IDH-wild type and mutated gliomas nor between methylated and
unmethylated MGMT promoter tumours.
Conclusion: In our cohort of glioma patients, APT-wi provides useful information to discriminate between tumour gradings as well as
one-year survival.
Limitations: The study was limited by the limited cohort of patients, which occurred as it was a single-centre study.
Utilising the amide proton transfer technique to characterise diffuse gliomas based on the WHO 2021 classification of
CNS tumours (7 min)
Elena Filimonova; Novosibirsk / Russia
223
Author Block: E. Filimonova, A. Pashkov, N. Borisov, A. Kalinovsky, J. Rzaev; Novosibirsk/RU
Purpose: Diffuse gliomas present a significant challenge for healthcare systems globally. While brain MRI plays a vital role in
diagnosis, prognosis, and treatment monitoring, accurately characterising gliomas using conventional MRI techniques alone is
challenging. In this study, we explored the potential of utilising the amide proton transfer (APT) technique alone or in combination
with other quantitative MRI sequences to predict tumour grade and type based on the WHO 2021 Classification of CNS Tumours.
Methods or Background: Fourty-two adult patients with histopathologically confirmed brain gliomas were included in the study.
They underwent 3T MRI imaging, which involved APT, arterial spin labelling (ASL), and diffusion-weighted imaging sequences.
Multinomial and binary logistic regression models were employed to classify patients into clinically relevant groups based on MRI
findings and demographic variables.
Results or Findings: We found that the best model for tumour grade classification included patient age along with APT values. The
highest sensitivity (88%) was observed for Grade 4 tumours, while Grade 3 tumours showed the highest specificity (79%). For tumour
type classification, our model incorporated four predictors: APT values, necrosis, and the presence of haemorrhage. The glioblastoma
group had the highest sensitivity and specificity (87%), whereas balanced accuracy was the lowest for astrocytomas (0.73).
Conclusion: The APT technique shows great potential for noninvasive evaluation of diffuse gliomas. The changes in the classification
of gliomas as per the WHO 2021 version of the CNS Tumour Classification did not affect its usefulness in predicting tumour grade or
type.
Limitations: The study was limited by its small sample size, and lack of follow-up data.
Ethics committee - additional information: This study was approved by the local Ethics Committee of the Federal Center for
Neurosurgery, Novosibirsk, Russia (protocol No. 4 dated 02-08-2022).
Decoding malignant glioma heterogeneity by PET-MRI habitat analysis of HYpoxia, PERfusion and DIffusion imaging:
preliminary results of the HYPERDIrect study (7 min)
Antonella Castellano; Milan / Italy
Author Block: G. Nocera, N. Pecco, M. Bailo, M. Callea, P. V. Scifo, P. Della Rosa, F. Gagliardi, A. Falini, A. Castellano; Milan/IT
Purpose: Malignant gliomas are characterised by considerable intratumour heterogeneity, directly related to treatment failure. A
novel method for cancer detection involves identifying regions or habitats within tumours by assessing shared imaging
characteristics. This approach utilises quantitative analysis of conventional and advanced imaging data through mathematical
models, which effectively partition the tumour into voxel-based subregions exhibiting similar radiological features. The integration of
multiple images further refines the creation of distinct tumour habitats. In this study, habitat analysis has been applied on hybrid
PET/MR images to map hypoxia, neoangiogenesis, cellularity, and tumour metabolism.
Methods or Background: Twenty patients with suspected malignant glioma candidate for surgical resection or biopsy underwent
preoperative hybrid 3T PET/MRI acquisitions for assessing HYpoxia (using quantitative blood oxygenation level-dependent – q-BOLD
imaging), PErfusion (using Dynamic Contrast-Enhanced DCE-MRI), DIffusion (usng DTI), and methionine-PET for tumour metabolism.
Data obtained were processed to generate HYPERDIrect habitat maps. In FLAIR-derived tumour volumes, an automatic clustering
algorithm classified voxels of each quantitative map into two clusters (high and low intensity) . The combination of the clusters from
all maps identified eight distinct habitats. Maps were imported into the neuronavigational system to perform imaging-guided sampling
for histopathological correlation.
Results or Findings: Preliminary findings demonstrated high habitat imaging reproducibility and a reliable correlation between the
expected microenvironment of the different habitats and the actual histopathological characteristics: samples from more aggressive
habitats (with reduced diffusivity, high perfusion, and low hypoxia) histologically coincide with regions displaying elevated cell density
and increased microvascular proliferation. Samples from less aggressive habitats (high diffusivity, low perfusion, and low hypoxia)
correspond to glioma infiltrative areas.
Conclusion: Habitat imaging using the HYPERDIrect map approach might serve as a potential biomarker for non-invasively
characterizing tumour heterogeneity in vivo.
Funding for this study: Funding was received from the Italian Ministry of Health, grant number GR-2018-12365670.
Ethics committee - additional information: This study was approved by the Ethics Committee of Ospedale San Raffaele on March
9th, 2022 (code 19/INT/2022).
Vasari-based features nomogram to predict the tumour-infiltrating CD8+ T cell levels in glioblastoma (7 min)
Caiqiang Xue; Lanzhou / China
224
Author Block: C. Xue, J. Zhou; Lanzhou/CN
Purpose: Tumour-infiltrating CD8+ T cells play a key role in glioblastoma development, malignant progression, and recurrence. This
study sought to establish nomograms based on the Visually AcceSAble Rembrandt Images (VASARI) features of multiparametric
magnetic resonance imaging (MRI), to determine the expression levels of tumour-infiltrating CD8+ T cells in patients with
glioblastoma.c
Methods or Background: Pathological and imaging data of 140 patients with glioblastoma confirmed by surgery and pathology
were retrospectively analysed. The levels of tumour-infiltrating CD8+ T cells in tumour tissue samples obtained from patients were
quantified using immunohistochemical staining. Patients were divided into high and low CD8 expression groups. The MRI images of
patients with glioblastoma were analysed by two radiologists using the VASARI scoring system.
Results or Findings: A total of 25 MRI-based VASARI imaging features were evaluated by two neuroradiologists. The features with
the greatest predictive power for CD8 expression levels were cystic (OR, 3.063; 95% CI: 1.387, 6.766; P=0.006), haemorrhage (OR,
2.980; 95% CI: 1.172, 7.575; P=0.022), and ependymal extension (OR, 0.257; 95% CI: 0.114 0.581; P=0.001). A logistic regression
model based on these three features showed better sample predictive performance (AUC=0.745; 95% CI: 0.665, 0.825;
Sensitivity=0.527; Specificity=0.857).
Conclusion: The VASARI feature-based nomogram model shows promise in predicting the level of infiltrative CD8 expression in GB
tumours noninvasively for earlier tissue diagnosis and more aggressive treatment.
Limitations: This study was limited by the lack of data, all from one centre. Further studies were not conducted in combination with
multimodal MRI metrics. The predictive performance of the parameters included in this study was not high and may be improved in
future studies using artificial intelligence methods.
Funding for this study: Funding was received from the National Natural Science Foundation of China (grant number 82071872).
Ethics committee - additional information: This study was approved by the Lanzhou University Second Hospital (Approval No.
2020A-070).
Assessment of axonal fibre integrity using DWI MRI models in post-surgery low-grade glioma patients (7 min)
Chris W.J. Van der Weijden; Groningen / Netherlands
Author Block: A. Van Der Hoorn, H. L. Van der Weide, M. Kramer, J. Kłos, R. Borra, E. De Vries, C. W. Van der Weijden; Groningen/NL
Purpose: Diffusion-weighted imaging (DWI) using MRI is commonly used for characterising and monitoring patients with low-grade
glioma (LGG). Advanced DWI models can potentially provide more detailed microstructural information than traditional DWI
parameters. This study aims to establish the relationship between outcome parameters derived from advanced DWI models, and to
assess the parameter reliability of these models in post-surgery LGG-patients.
Methods or Background: The study involved 14 post-surgery LGG-patients and 6 healthy controls (HC) who underwent DWI and
T2w-FLAIR-MRI scans. Several DWI modeling approaches (diffusion tensor imaging, DTI; diffusion kurtosis imaging, DKI; white matter
tract integrity, WMTI; neurite orientation dispersion and density imaging, NODDI; fixel based analysis, FBA) were employed to extract
parameters like fractional anisotropy (FA), intracellular diffusivity (ICD), axonal water fraction (AWF), and fibre density (FD). These
parameters were chosen to assess axonal fibre integrity, which should be absent in the surgical cavity, affected in the perisurgical
cavity (hyperintensity on T2w-FLAIR-MRI), and unaffected in normal appearing white matter (NAWM) of LGG-patients and WM of HC.
Results or Findings: All parameters were significantly different between different tissues, with a gradient of low (DTI-FA, DKI-FA,
WMTI-ICD, WMTI-AWF) or no signal (NODDI-ICD, FBA-FD) in the surgical cavity, to medium signal in peri-surgical cavity, and high
signal in NAWM/WM. Spearman correlations revealed that all parameters corresponded well with each other (R2=0.69-0.99).
Conclusion: The results of WMTI-ICD and WMTI-AWF suggest the presence of intracellular diffusivity in the surgical cavity, probably
because the model does not account for cerebrospinal fluid. These WMTI-data and the high agreement between parameters derived
from different models suggest that if only diffusivity direction is required, DTI-FA might suffice. Should one would like to measure
intracellular water specifically, then NODDI-ICD would be the better alternative.
Ethics committee - additional information: This study was approved by the Medical Ethics Review Board (METc) of the University
Medical Centre, Groningen.
Predictive value of cellular metabolism parameters derived from perfusion MRI for the overall survival of patients with
glioblastoma (7 min)
Chris W.J. Van der Weijden; Groningen / Netherlands
225
Author Block: C. W. Van der Weijden, R. Borra, E. De Vries, A. Van Der Hoorn; Groningen/NL
Purpose: Perfusion weighted imaging (PWI) with MRI is used to characterise and monitor patients with glioblastoma during and after
treatment. Using advanced models, PWI MRI can provide information about metabolism. Metabolic parameters are altered upon
tumour growth, making them potential indicators of early tumour progression. This study aims to determine the predictive value of
advanced PWI parameters for the overall survival of glioblastoma patients.
Methods or Background: Dynamic susceptibility contrast (DSC) PWI and T1w post-contrast MRI was acquired in 7 patients with
post-surgery GBM. PWI was modelled to obtain the mean transit-time (MTT), relative cerebral blood volume (rCBV), relative cerebral
blood flow (rCBF), capillary transit-time heterogeneity (CTH), oxygen extraction fraction (OEF), and cerebral metabolic rate of oxygen
(CMRO2) for regions around the surgical cavity. Three perilesion regions were defined on the T1w post-contrast images, respectively
at distances of 0-0.5cm, 0.5-1.0cm, and 1.0-1.5cm from the surgical cavity. The outcome parameters were associated with overall
survival.
Results or Findings: Visual inspection of the data showed a gradient in the parameters from the surgical cavity outward, albeit not
statistically significant. The MTT for perilesion of 0-0.5cm, 0.5-1.0cm, and 1.0-1.5cm distance from the surgical cavity, was 4.2±1.2,
3.6±1.3, 3.5±1.3, rCBV was 2.0±0.6, 2.0±0.5, 2.3±0.7, rCBF was 33.9±11.2, 45.0±15.0, 50.7±7.2, CTH was 4.7±1.2, 4.3±1.5,
4.3±1.4, OEF was 0.4±0.1, 0.4±0.1, 0.4±0.1, and CMRO2 was 12.4±2.8, 14.4±2.7, 16.3±3.4, respectively. None of the parameters
were significantly correlated with overall survival.
Conclusion: Although no significant differences were observed yet, the PWI parameters seemed to normalise with greater distance
from the surgical cavity. This was even stronger for the currently hardly used CMRO2, although the sample size will be increased to be
able to draw firm conclusions.
Limitations: The study was limited by its sample size.
Ethics committee - additional information: This study was approved by the Medical Ethics Review Board (METc) at the University
Medical Centre, Groningen.
Selective vulnerability of cognitive networks in patients with glioma measured by resting-state fMRI (7 min)
Luca Pasquini; Rome / Italy
Author Block: L. Pasquini1, A. Napolitano2, M. Jenabi1, K. Peck1, M. Schmid2, A. I. Holodny1; 1New York City, NY/US, 2Rome/IT
Purpose: Gliomas affect the whole brain through structural or functional disconnection. This study investigates the impact of gliomas
on the cortical synchronisation of brain networks using resting-state fMRI (rs-fMRI). We hypothesised a different vulnerability of brain
networks depending on underlying function, tumour location and grade.
Methods or Background: We recruited 147 glioma patients (89M, 50.95±16.13y, 92 high-grade, 55 low-grade) and 200 healthy
controls (HCs) with rs-fMRI. Glioma segmentation was performed using 3D-slicer on FLAIR. Group independent component analysis
(ICA) was used to extract spatially independent components (IC), subsequently categorised into their respective networks using
NeuroMark fMRI 1.0 atlas. We computed cosine similarity (CS) and ran a permutation test to compare, for each IC, the CS of each
patient against the HCs distribution. A Chi-squared test was used to verify the significance of networks alterations, tumour location
,and genetics (p<0.05).
Results or Findings: Out of the 20 network components in HCs, 10.38±1.43 resulted altered in patients, including cognitive control
network (CCN); default mode network (DMN); subcortical network (SCN); sensorimotor network; visual network (VN). CCN showed
significant alterations with tumours in the temporal lobe (p=0.005), Broca's (p=0.01), and Wernicke's area (p=0.041). Tumours in
Wernicke's area also altered the DMN (p=0.04), SCN (p=0.038), and VN (p=0.003). Network alterations persisted with increased
distance from the tumour, and were more pronounced with higher WHO-grade (p<0.001).
Conclusion: These results indicate specific vulnerability of cognitive networks to tumour growth. Functional alterations extend
beyond tumour boundaries, and increase with WHO-grade. Tumour location in known eloquent areas exerts widespread effects on
brain networks.
Limitations: Limitations of this study include its retrospective design and lack of complete neuropsychological testing.
Funding for this study: Funding was received from the National Institutes of Health (NIH): NIH-NIBIB R01 EB022720, NIH-NCI R21
CA220144, NIH-NCI P30 CA008748; MSK Cancer Center Support Grant/Core Grant P30 CA008748.
Ethics committee - additional information: This study was approved by the local Institutional Research Board's ethics committee
with code: (16-360).
226
RPS 609 - New insights in interventional radiology above the neck
Categories: Head and Neck, Interventional Radiology

CME Credits: 1
Moderator:
Franca Wagner; Bern / Switzerland
Predictors of parenchymal haematoma after mechanical thrombectomy in patients with large ischeamic core due to a
large vessel occlusion in the anterior circulation (7 min)
Davide De Leoni; Rome / Italy
Author Block: D. De Leoni, A. M. Alexandre, A. Broccolini, A. Pedicelli; Rome/IT

Purpose: This study aimed to evaluate the predictors of any type of parenchymal haematoma according to ECASS II classification
and the predictors of type II parenchymal haematoma after MT in patients with an ASPECTS score 5 due to a LVO in the anterior
circulation.
Methods or Background: In this retrospective observational study, databases of 16 comprehensive stroke centres were screened
for consecutive patients with AIS due to LVO diagnosed between January 2016 and December 2021. The ASPECT score was calculated
from two standard axial CT slices.
Brain bleeding was confirmed with CT or MRI scan (24 and 72h after MT). Heidelberg classification was used to classify haematoma.
Presence of PH type 2 was selected as the primary outcome measure, due to its relevance in the overall clinical outcome of patients.
Results or Findings: 408 patients with LVO in the anterior circulation and an ASPECT score ≤5 that received MT were selected.
Higher number of passages in the second technique (OR 4.176, 95% CI 1.444 – 12.003, p=0.008) was predictive of PH type 2,
whereas procedure conducted under general anesthesia was associated with a lower risk of PH type 2 (OR 0.084, 95% CI 0.011 –
0.612, p=0.015). Higher age (OR 1.073, 95% CI 1.023 – 1.126, p=0.004) and number of passages during the second technique (OR
2.119, 95% CI 1.106 – 4.374, p=0.025) were an independent predictor of any PH.
Conclusion: Our study showed that an increased number of passages during MT in patients with large baseline brain infarct is a
predictor of post-procedural PH of any type. The choice of general anesthesia for MT should be considered as this is associated with a
lower risk of occurrence of PH type 2.
Limitations: This was a retrospective study.
Emergency carotid artery stenting in endovascular stroke treatment: a feasible and relatively safe procedure (7 min)
Ruth Kaufmann; The Hague / Netherlands
227
Author Block: R. Kaufmann, L. C. Van Dijk, W. Stomp, K. De Laat, S. F. De Bruijn, J. Piet, S. C. Klink, H. van Overhagen; The Hague/NL
Purpose: Emergency carotid artery stenting (eCAS) during endovascular stroke treatment (EVT) is considered controversial in the
literature. We performed a retrospective analysis in our institute to assess safety and efficacy.
Methods or Background: During 2015-2023, 533 patients underwent EVT. Seventy patients underwent 71 eCAS procedures
concurrently. Patient files were analysed on baseline characteristics, EVT procedure, and complication rates.
Results or Findings: There were 46 men and 24 women with a mean age of 73 years (range 53-99 years). Preprocedural NIHSS
scores ranged from 3 to 25 (median 14). Indications were significant carotid artery stenosis (n=36), occlusion (n=28), and dissection
(n=7; 3 of 7 occurred during EVT). Carotid revascularisation and stenting were performed prior to intracranial revascularisation in the
vast majority of patients. Stenting was performed in the left carotid artery (n=39) and in the right carotid artery (n=32), using 53
Carotid Wallstents, 15 Cordis Precise stents, and 3 X-pert stents. Additional endovascular treatment was performed in 55 patients;
carotid T (n=4), M1 (n=28), M1 and M2 (n=5), M2 (n=16), M3 (n=2), and A2 (n=1). All patients received a single intra-arterial dose of
5000 to 10000 IU heparin during the procedure. mTICI scores immediately postprocedural were either 2b (n=22) or 3 (n=49).
Complications occurred in 10 out of 71 patients (14%): intracranial bleeding (n=5 (7%)), access bleeding (n=4 (6%)), and vessel
occlusion (n=1) (1%)).
Conclusion: eCAS seems feasible and relatively safe with low intracranial bleeding rates in patients undergoing EVT. A single intra-
arterial bolus of heparin does not seem to increase the occurrence of intracranial bleeding.
Limitations: Our study is a retrospective, single-centre cohort study.
Ethics committee - additional information: Since it is a retrospective study, we received a waiver from our ethical committee.
Endovascular treatment of basilar artery occlusion: experience in a tertiary hospital (7 min)

Francisco Esteban Dianderas Gutiérrez; Sevilla / Spain
Author Block: F. E. Dianderas Gutiérrez, A. Alboniga-Chindurza Barroeta, E. Zapata Arriaza, M. Aguilar Perez, R. F. Ocete Pérez, A.
Zamora Piñero, A. González García; Sevilla/ES
Purpose: This study is a description and analysis of the experience in endovascular treatment of basilar artery occlusion (BAO) in a
tertiary hospital with a reference population of 2.5 million inhabitants, one of the largest in Europe.
Methods or Background: Patients with BAO in the period 2017-2021 registered in the Andalusian thrombectomy database
(ARTISTA) of the Huelva-Seville node were included. A statistical analysis of the demographic variables, recanalisation rate, and good
functional prognosis (modified Ranking scale (mRs≤2y≤3) as well as complications of the procedure was performed.
Results or Findings: Of 148 patients (7.8% of the ARTISTA database), median 70.6 years (63-80), pre-procedure NIHSS of 14
(6-24.7), 31.1% received fibrinolysis, 21.6% were awakening strokes, and 12.8% were in-hospital. The symptom onset-recanalisation
time was 358 minutes (256-631), with a median recanalisation puncture of 40 minutes (17.7-89). In 62.2% of cases there was BA
occlusion; in 23.9% and 9.9% of cases this also involved the vertebral and posterior cerebral arteries, respectively. In 61.1% of the
cases, the endovascular procedure was performed with local anesthesia (general: 34%, sedation: 4.9%). Distal aspiration was
performed in 82%, stent retriever in 6%, Solumbra in 5% and ATP and /or stent in 7%. The recanalisation rate was TICI ≥2b in 85,7%
of the cases. The MR at 90 days was 0-2 in 37.9% and 0-3 in 46.4%. 36.4% of the patients were Rank 6. In 9.3% of cases there were
some complications of the procedure (dissection in 2.2%). Symptomatic intracranial haemorrhage occurred in 5% of cases (PH1:
2.1%, PH2: 2.9%). 36.4% of cases required admission to the ICU; the rest of the patients went to the Stroke Unit.
Conclusion: Our results in the treatment of patients with OAB are similar to those described in the ATTENTION and BAOCHE recent
studies.
Ethics committee - additional information: This was a retrospective and educational study. All data was anonymised.
Outcome predictors in patients with large vessel occlusion and large ischaemic core undergoing effective mechanical
recanalisation: a retrospective multicentre study (7 min)
Arianna Camilli; Rome / Italy
228
Author Block: A. Camilli, A. M. Alexandre, A. Broccolini, A. Pedicelli; Rome/IT
Purpose: Mechanical thrombectomy, according to current guidelines, is not recommended for patients with a baseline large
ischaemic core, defined as an ASPECT score ≤5. The aim of our study was to evaluate the possible association of baseline clinical and
neuroradiological feature with clinical outcome in patients presenting with large ischaemic core.
Methods or Background: In this retrospective observational study, the prospective databases of 16 comprehensive stroke centers
were screened for consecutive patients with AIS due to LVO diagnosed between January 2016 and December 2021. Patients with an
ASPECT score ≤5 were included in this study. A 90-day mRS score of 0-2 was chosen as the primary clinical outcome. Secondary
clinical outcome measure was a 90-day mRS score of 0-3. Safety outcome measures were brain bleeding events and death of any
cause within 90 days.
Results or Findings: 432 patients with AIS due to LVO in the anterior circulation and a baseline ASPECT score ≤5 that were
subjected to MT were available for analysis. In univariate analysis of patients receiving efficient recanalisation after MT, lower age,
pre-event mRS and baseline NIHSS score as well as a higher ASPECT and Menon scores were associated with a 90-day mRS score 0-2.
In multivariate analysis that used variables with a p-value ≤0.1 in univariate analysis, lower age (OR 0.962, 95% CI 0.937-0.988,
p=0.004), baseline NIHSS (OR 0.908, 95% CI 0.837-0.986, p=0.021) and a higher Menon score (OR 1.773, 95% CI 1.278- 2.459,
p<0.001) were associated with a favourable 90-day clinical outcome.
Conclusion: In our series, a lower age, a lower baseline NIHSS score, and a higher Menon score were associated with favourable
clinical outcome at 90 days.
Limitations: This study was limited by being retrospective and observational.
Ethics committee - additional information: The study was approved by an Ethics Committee. The submitter provided no
additional information.
Chronic subdural haematoma: is endovascular treatment a safe and effective alternative? (7 min)
Isabel Bermudez-Coronel; Madrid / Spain
Author Block: E. Fandiño Benito, I. Bermudez-Coronel, J. C. Mendez; Madrid/ES

Purpose: The purpose of this study was to determine the effectiveness and safety of middle meningeal artery embolisation in chronic
subdural haematoma.
Methods or Background: A total of 34 patients underwent middle meningeal artery embolisation, with 3 patients treated in 2021,
12 in 2022, and 19 in 2023. Out of these, 11 of the 34 patients (32.35%) underwent bilateral embolisation. Prior to embolisation, 18
patients (52.94%) had previously undergone surgical drainage of the haematoma, while embolisation served as the primary
treatment for 12 patients (35.29%). In all cases, embolisation was carried out using embolic liquid agents.
Results or Findings: Patient follow-up was carried out and measurement and evolution of the haematoma was recorded over time.
On average, the thickness of the haematoma decreased by 70% compared to the pre-treatment CT scans. During the follow up
period, two patients died due to other underlying medical conditions. Additionally, one patient experienced sixth cranial nerve palsy,
which spontaneously resolved within 48 hours following the procedure. Notably, while general anesthesia was chosen for only 56.6%
of the total cases, it is increasingly becoming the preferred option over sedation anesthesia. Encouragingly, throughout the two-year
follow-up, no patient required surgical or endovascular reintervention.
Conclusion: Middle meningeal artery embolisation appears to be an effective and safe treatment for chronic subdural haematoma,
both, as a the primary option and as an adjuvant treatment to surgery, ultimately reducing the need for reinterventions.
Limitations: The primary limitation of this study lies in its observational nature, conducted at a single centre. To establish more
robust conclusions, a multicentre randomised controlled trial should be carried out.
Endovascular management of post-irradiated carotid blowout syndrome in patients with lower neck cancers (7 min)
Feng-Chi Chang; Taipei / Taiwan, Chinese Taipei
229
Author Block: F-C. Chang; Taipei/TW
Purpose: Lower neck cancers (LNCs) include specific tumour types and have vascular collaterals from other head-and-neck cancers
(HNCs). This prospective study evaluated the outcome of endovascular management of post-irradiated carotid blowout syndrome
(PIRCBS) in LNC patients by comparing reconstructive management (RE) and deconstructive management (DE).
Methods or Background: Between 2015 and 2021, LNC patients complicated with PIRCBS who underwent endovascular therapy
were prospectively included. The patients underwent DE by permanent embolisation of both internal carotid artery (ICA) and external
carotid artery (ECA) with coils (Nester, Cook; Concerto nylon detachable coils, ev3 Neurovascular) and/or adhesive agent (histoacryl,
Braun Surgical). ICA was occluded if pathological lesion was located at ICA only. Cross embolisation including proximal and distal to
the pathological lesion were done in all patients. Technical outcomes were evaluated by immediate haemostasis and neurological
complications. The haemostatic outcome was evaluated by the haemostatic result and haemostatic period.
Results or Findings: 59 patients (mean age: 58.45±10.25, 56 men) were enrolled, including 28 RE patients and 31 DE patients. The
results of RE versus DE were as follows: rebleeding events: 13/28 (46.4%) versus 10/31 (32.3%), p=0.265; haemostatic periods
(month): 9.4±14.0 versus 14.2±27.8, p=0.589; neurological complications: 4/28 (14.3%) versus 5/31 (16.1%), p=0.844; survival time
(months): 11.8±14.6 versus 15.1±27.5, p=0.605.
Conclusion: We observed that there was no difference of rebleeding risk and neurological complication between DE and RE groups.
RE could be used as a potential routine treatment for PIRCBS in patients with LNC.
Limitations: Several limitations were noted in this study, including variable entities and staging of malignancy in the lower neck
region. The treatment modalities of our patients with LNC were also heterogeneous, which might influence the outcome of CBS.
Funding for this study: This study received funding from Taipei Veterans General Hospital, Taiwan [V110C-037, V111C-028,
V112C-059, V112D67-002-MY3-1 (to FCC)], National Science and Technology Council, Taiwan [109-2314-B-075-036, 110-2314-
B-075-032 and 112-2314-B-075-066 (to FCC)].
Ethics committee - additional information: This study was approved by the Taipei Veterans General Hospital with code:
2022-02-001CU.
Efficacy and safety of use of flow diverting silk stents in endovascular treatment of intracranial aneurysms (7 min)
Elif Hazal Karlı; Istanbul / Turkey
Author Block: E. Sahibli, E. Kara, E. H. Karlı, M. Barburoğlu; Istanbul/TR

Purpose: The purpose of this study was to determine the feasibility, efficacy, and safety of flow converting silk stents in treating
intracranial aneurysms.
Methods or Background: The study included 75 patients (mean age 49.54 years), with at least six months of angiographic and
long-term follow-up imaging, with 95 aneurysms treated in 78 different procedures. The patients' pre-treatment and follow-up clinical
data, procedure results, demographic characteristics, aneurysm characteristics and localisations, occlusion rates, early and late
mortality, and permanent morbidity rates were analysed retrospectively.
Results or Findings: According to the morphology of the aneurysms, 89 (93.7%) were saccular, 2 were fusiform (2.1%), 2 (2.1%)
were blister, and 2 (2.1%) were dissected; according to size, 31 were (32.63%) small, 40 were (42.11%) medium, 11 were (11.58%)
large, and 13 were (13.58%) giant; according to localisation 87 were (91.58%) anterior circulation (AC) and 8 were (8.42%) posterior
circulation (PC); according to wall placement 91 (95.79%) were lateral and 4 (4.21%) were bifurcated; according to aspect ratio, 40
(42.11%) had a narrow neck and 52 (54.73%) had a wide neck. Two patients were treated for ruptured blister aneurysms. Aneurysm
occlusion rates were found to be 87.2% at 12 months, and 93.68% at long-term follow-up. A total of 9 patients (11.53%) developed
procedural complications. In angiographic long-term follow-ups, postprocedural late complications developed in 7 patients (8.9%). The
overall mortality rate was 5.33%, with postprocedural early 2.66% and late 2.66%. The permanent morbidity rate was 1.33%. The
mean follow-up period was 42.68±26.5 months, and the mean occlusion time was 8.76±8.3 months.
Conclusion: It was concluded that the use of flow-diverting silk stent was effective and safe in the endovascular treatment of
intracranial aneurysms. Technical success was worse in PC aneurysms compared to anterior, mortality and morbidity rates were lower
in anterior circulation.
Limitations: The study was retrospective.
Ethics committee - additional information: The study was approved by the Ethics Committee of the Medical School of Istanbul
University Istanbul on the 24th September 2021 with code number 17.
Contour device implantation for treatment of intracranial aneurysms in the basilar tip (7 min)
Karim Mostafa; Kiel / Germany
230
Author Block: K. Mostafa, F. Bueno Neves, F. Gärtner, T. Klintz, N. Larsen, S. Peters, J. Hensler, O. Jansen, F. Wodarg; Kiel/DE
Purpose: The purpose of this study was to assess and evaluate the feasibility and the periinterventional aspects of embolisation of
intracranial aneurysms located in the basilar tip using the Contour Neurovascular System (CNS).
Methods or Background: Treatment of basilar apex aneurysms remains challenging regarding the nobility of the parent vessel and
their often wide-necked configuration. Novel intrasaccular flow-disruption devices constitute an endovascular treatment alternative
besides coiling and stent-assisted procedures.
The presented study reports periinterventional and feasibility aspects of embolising basilar tip aneurysms with the CNS. Here, a
retrospective analysis of eight patients after CNS implantation into a basilar apex aneurysm was performed. Data on intervention
times, radiation dose, procedural success, and complications were gathered. All patients received follow-up digital subtraction
angiography after six months.
Results or Findings: CNS implantation was successful in all patients. Mean device instrumentation time was 18.8±7.7 minutes with
a mean full intervention time of 100±65.8 minutes. Mean full procedure radiation dose was 1917 (421-5107) cGy/cm2. Upon
interventional angiographic follow-up after six months, six aneurysms were adequately occluded. One showed constant slight
perfusion inside the device without the need for reintervention and one patient had undergone reintervention. No deaths were
reported.
Conclusion: CNS implantation is an alternative endovascular method for embolising intracranial aneurysms located in the basilar tip
with short intervention times and promising short- and medium-term follow-up data concerning aneurysm occlusion and
reinterventions.
Limitations: The number of patients included in this study was small due to the single-centre design as well as the novelty of the
CNS combined with the rarity of the specific aneurysm. Hence it can not be concluded that basilar apex aneurysm treatment using
the CNS is equal to conventional procedures. Additionally, further research is needed to evaluate long-term stability.
Ethics committee - additional information: This study was approved by the Ethikkomission der Christian-Albrechts-Universität
Kiel.
231
RPS 602 - The role of breast imaging in breast cancer treatment
Categories: Breast, Contrast Media, Imaging Methods, Interventional Oncologic Radiology, Molecular Imaging
CME Credits: 1
Moderator:
Nuala Healy; Cambridge / United Kingdom
Comparison of radiomics-based machine-learning classifiers for pretreatment prediction of pathologic complete

response to neoadjuvant therapy in breast cancer (7 min)
Xue Li; Beijing / China
Author Block: X. Li, C. Li, L. Jiang, M. Chen; Beijing/CN

Purpose: In recent years, machine learning (ML) classifiers have been used to establish high-performance predictive models for
pathological complete response (pCR) in breast cancer after neoadjuvant therapy (NAT). However, few studies have compared the
effectiveness of different ML classifiers. This study investigated the ability of radiomics models based on pre- and post-contrast first-
phase T1WI to predict breast cancer pCR after NAT and compared the performance of different ML classifiers.
Methods or Background: In this retrospective study, 300 patients from the Duke-Breast-Cancer-MRI dataset who underwent NAT
were included, including pCR (n=76) and non-pCR (n=224) cases. These patients were randomly divided into training and validation
groups at a ratio of 8:2. Radiomics features were extracted from pre- and post-contrast first-phase T1WI images of each patient. The
radiomics model was built using features selected through the Spearman correlation analysis and the LASSO algorithm after
normalisation. Seven ML classifiers were used to assess the predictive performance of the radiomics models.
Results or Findings: Out of the seven classifiers used, the LightGBM classifier performed best in predicting breast cancer pCR, with
an AUC of 0.813 in the validation group (accuracy 78.3%, sensitivity 46.7%, specificity 100.0%). During subgroup analysis, RF
achieved the highest AUC in pCR prediction in luminal breast cancers (0.859, accuracy 85.9%, sensitivity 68.8%, and specificity
83.3%), and DT yielded the highest AUC in pCR prediction in triple negative (TN) breast cancers (0.909, accuracy 88.2%, sensitivity
81.8%, and specificity 100%).
Conclusion: Overall, the LightGBM-based radiomics model demonstrated superior performance in predicting breast cancer pCR,
while RF and DT displayed promising results in predicting pCR for luminal and TN breast cancers, respectively, during subgroup
analysis.
Limitations: Our study included different NAT treatment regimens, and subgroup analysis based on treatment regimens was not
performed.
Ethics committee - additional information: Data obtained from the Cancer Imaging Archive (TCIA,
https://www.cancerimagingarchive.net/) did not require ethical approval; informed consent was waived since the TCIA dataset de-
identified patient information.
Ultrasound guided vacuum-assisted biopsy to assess pathological complete response to neoadjuvant therapy: a
paradigm shift in breast cancer treatment (7 min)
Elisa D’Ascoli; Milan / Italy
232
Author Block: E. D’Ascoli1, C. Depretto1, G. Della Pepa1, C. De Berardinis1, G. Irmici1, C. Cazzella2, D. Ballerini1, A. Bonanomi1, G. P.
Scaperrotta1; 1Milan/IT, 2Bergamo/IT
Purpose: The aim of this study was to evaluate the diagnostic performance of a pre-surgical mini-invasive ultrasound-guided biopsy
to predict pathological complete response (pCR) in breast cancer (BC) patients after neoadjuvant therapy (NAT) in order to assess the
possibility of omitting surgery in exceptional responders.
Methods or Background: We enrolled patients with histologically confirmed TN, Her2+ and Luminal B cT1-cT2-cT3 cN0-cN1
monofocal BC who received NAT with complete radiological remission on imaging (ultrasound, mammography, MRI and CEM) or with
breast residual tumor <1cm. Patients underwent ultrasound guided vacuum-assisted breast biopsy (VABB) and the results were
compared with the final histological results obtained after surgery.
Results or Findings: We enrolled 27 patients; 15/27 lesions were classified as TN, 6/27 Her2+ and 6/27 Luminal B. 22/27 cases
showed complete radiological response; 5/27 had residual tumor <1cm. In two cases, VABB identified residual disease and post-
surgical histological examination was negative. Overall, among the remaining 25 cases, in 22 (88%) there was concordance between
the VABB result and the final histological examination. Among TN and Her2+ concordance was observed in 17/19 (89.5%) cases.
VABB did not identify the presence of residual invasive disease in one case of Luminal B (16.7%) and in one case of TN (6.7%); in one
TN case (6.7%) VABB was negative but definitive histological examination showed the presence of carcinoma in situ.
Conclusion: pCR rates are rapidly improving, especially in TN and Her2+ BC patients. Reliably identifying patients with pCR could
lead to deescalation of locoregional therapy after NAT, allowing surgery to be omitted in exceptional responders; this would result in a
reduction of post-surgical complications and healthcare costs, and improvement in quality of life.
Limitations: The main limitations are the single-centre nature of the study and the limited number of patients.
Ethics committee - additional information: This study was approved by the Independent Ethics Committee at the IRCCS Istituto
Nazionale dei Tumori, Milano.
Unenhanced MRI for assessment of response to neoadjuvant therapy in patient with locally advanced breast cancer:
diagnostic value of DWI/ADC (7 min)
Marcella Pasculli; Rome / Italy
Author Block: M. Pasculli, F. Galati, V. Rizzo, G. Moffa, R. Maroncelli, F. Cicciarelli, F. Pediconi, C. Catalano; Rome/IT
Purpose: The aim of this study was to investigate the predictive value of DWI/ADC (DW-MRI) for the assessment of response to
neoadjuvant therapy (NT) in patients with locally advanced breast cancer (LABC).
Methods or Background: Patients with LABC candidate to NT, and who underwent pre-treatment breast MRI between March 2021
to March 2022, were retrospectively enrolled. MRI-based staging and DWI/ADC values (x10-3mm2/s) were analysed. According to
post-surgical outcomes, patients were classified as complete responders (pCR) and non-complete responders (non-pCR). Pre-
treatment ADC values were compared to the tumour's pathological outcome and post-treatment downstaging. The diagnostic
accuracy of DWI-ADC in differentiating between pCR and non-pCR groups was calculated with receiver operating characteristic (ROC)
analysis.
Results or Findings: 36 patients were evaluated (pCR, n=20; non-pCR, n=16). Pre-treatment lesion ADC values were significantly
different between the two groups (p=0.034), while no association was found between pre-NT tumour size and pathological response.
ADC values pre-teatment showed significant correlations with loco-regional downstaging after therapy (r=-0.537, p=0.022) and with
tumour volume reduction (r=-0.480, p=0.044). ADC values could differentiate pCR from non-pCR patients, with a sensitivity of 75%
and specificity of 70%.
Conclusion: ADC values on pre-treatment MRI were strongly associated with the outcome in patients with LABC, both in terms of
pathological response and loco-regional downstaging after NT, suggesting the use of Unenhanced DW-MRI as a potential predictive
tool of response to therapy.
Limitations: This was a single-centre study with a limited number of patients.
Ethics committee - additional information: Informed consent was waived due to the retrospective nature of this study, as
approved by our local Ethical Committee.
Performance of the node-RADS scoring system for standardised magnetic resonance imaging assessment of lymph
nodes in breast cancer (7 min)
Roberto Maroncelli; Rome / Italy
233
Author Block: R. Maroncelli, F. Pediconi, M. Pasculli, A. Marra, F. Cicciarelli, V. Rizzo, G. Moffa, F. Galati, C. Catalano; Rome/IT
Purpose: The Node-RADS score was recently introduced to offer a standardised and comprehensive evaluation of lymph node
invasion (LNI) based on a five-item Likert scale. We tested Node-RADS score diagnostic performance and assessed the applicability
and feasibility of the score among readers.
Methods or Background: A retrospective study was conducted on BC patients who underwent lymph node dissection between
January 2020 and January 2023. All patients underwent breast contrast-enhanced magnetic resonance imaging. Sensitivity,
specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for different Node-RADS cut-off values
(>1, >2, >3, >4) in predicting LNI. Pathologic results were considered the gold standard. Additionally, the overall diagnostic
performance was evaluated using ROC curves and the Area Under the Curve (AUC). Finally, Cohen’s K analysis was used to assess
interreader agreement.
Results or Findings: The final population includes 192 patients. By increasing the Node-RADS cut-off values, specificity and PPV
rose from 71.4% to 100% and 76.7% to 100%, respectively for Reader 1 and 69.4% to 100% and 74.6% to 100% for Reader 2. Node-
RADS >2 could be considered the best cut-off value due to its balanced performance. Node-RADS exhibited an AUC of 0.97 for Reader
1 and 0.93 for Reader 2. Node-RADS assigned on CE-MRI images independently predict LNI after adjusting for other variables in a
multivariable regression analysis (p<0.001 for both Readers). An excellent interreader agreement was found (K=0.834).
Conclusion: The current study establishes the groundwork for implementing Node-RADS as a method for assessing regional lymph
nodes in BC patients. The Node-RADS score has demonstrated moderate-to-high overall accuracy in identifying LNI, providing the
flexibility to establish different cut-off values based on specific clinical scenarios.
Limitations: This study was based on a relatively limited cohort.
Ethics committee - additional information: Informed consent was waived due to the retrospective nature of our study, and
approved by the local Ethical Committee.
Incorporating breast cancer molecular subtype rather than axillary disease extent on baseline 18F-FDG PET/CT in
axillary treatment strategies (7 min)
Thiemo van Nijnatten; Maastricht / Netherlands
Author Block: F. v. Amstel1, C. De Mooij1, J. Simons2, C. Mitea1, C. Van Der Pol3, E. Luiten4, L. Koppert2, M. Smidt1, T. van Nijnatten1;
1
Maastricht/NL, 2Rotterdam/NL, 3Leiderdorp/NL, 4Al Ain/AE
Purpose: In clinically node-positive (cN+) patients treated with neoadjuvant systemic therapy (NST), axillary disease extent on
baseline 18F-FDG PET/CT combined with pathologic axillary response to NST has been proposed to guide axillary treatment de-
escalation. This study aimed to assess how breast cancer molecular subtype would affect such a treatment strategy.
Methods or Background: Patients with cN+ breast cancer treated with NST in the RISAS trial (NCT02800317) who underwent 18F-
FDG PET/CT at baseline were included (period 2017-2019). Baseline 18F-FDG PET/CT exams were centrally reviewed to differentiate
between limited (1-3) and advanced axillary disease (≥4 hypermetabolic axillary lymph nodes). After NST, all patients underwent the
RISAS-procedure followed by completion axillary lymph node dissection. Axillary pathologic complete response (pCR) rates were
stratified by axillary disease extent on baseline 18F-FDG PET/CT, and subsequently by hormone receptor (HR)+/HER2-, HR+/HER2+,
HR-/HER2+ and triple negative (TN) molecular subtypes.
Results or Findings: A total of 185 patients were included, 62.7% with limited and 37.3% with advanced baseline axillary disease.
Overall axillary pCR rate was 29.7% (7% for HR+/HER2-, 52.6% for HR+/HER2+, 75% for HR-/HER2+, and 34.1% for TN; p<0.001).
Axillary pCR rates did not significantly differ between limited versus advanced baseline axillary disease within the molecular subtypes.
Breast molecular subtype showed to be a significant predictor of axillary response.
Conclusion: Axillary pCR rates between limited and advanced axillary disease on baseline 18F-FDG PET/CT were not significantly
different within each of the breast molecular subtypes. Breast molecular subtype is important here since it showed to be a significant
predictor of axillary response. Therefore, baseline axillary disease extent should be given a less important status while breast
molecular subtype should be considered important to guide axillary treatment strategies in cN+ patients treated with NST.
Limitations: No limitations were identified
Funding for this study: Funding was received from the University Fund Limburg (SWOL; project 20.048) and Dutch Cancer Society
(KWF – REFINE-trial; project 14055).
Ethics committee - additional information: Due to the retrospective design of the study, the necessity to obtain written informed
consent was waived by the local medical ethics committee.
MRI characteristics predicting recurrence/metastases in breast cancer patients receiving neoadjuvant chemotherapy (7
min)
Aisha Syed; Cardiff / United Kingdom
234
Author Block: A. Syed, J. Bansal, M. Wallace; Cardiff/UK
Purpose: The aim of this study was to evaluate MRI and tumour characteristics predicting recurrence or metastases in breast cancer
patients post neoadjuvant chemotherapy ( NACT).
Methods or Background: A retrospective evaluation of all breast MRIs for NACT monitoring between 2009 and 2018 was performed.
All patients were followed up for at least five years. Factors including patients’ age, tumour size, receptor status, number of lymph
nodes, and MRI characteristics were evaluated. SPSS was used for statistical evaluation, and p<0.05 was considered a significant
result. Binomial logistic regression was used to evaluate factors, controlling for other variables. The median age of patients was 45
years (range 25 to 73).
Results or Findings: Out of 135 patients, 114 had adequate data for evaluation in this study. Thirty-three (28.9%) patients showed
local recurrence or metastases. The median time to an event from the date of diagnosis was 35 months (range 0-144 months).
Compared to a non-mass enhancement, a mass-like enhancement was statistically associated with fewer events (p=0.011). The
factors most significantly associated with an event were triple negative (TN) status, a higher number of lymph nodes on baseline MRI,
and post-surgery (ypN). No significant association was found between T stage, tumour grade or MR response pattern (concentric
versus crumbling).
Conclusion: Of all breast cancer patients receiving NACT, 28.9% showed an adverse event at a median of 35 months. Factors
predicting an event in breast cancer patients receiving NACT were TN receptor status, non-mass enhancement on MRI, and higher
lymph node status.
Limitations: This was a small retrospective study.
Ethics committee - additional information: This was a retrospective study of imaging, therefore no ethics approval was required.
Baseline MRI BI-RADS and breast oedema score features as predictors for axillary lymph node response to neoadjuvant
chemotherapy in breast cancer (7 min)
Caroline Malhaire; Paris / France
Author Block: C. Malhaire1, O. Umay1, F. Frouin2; 1Paris/FR, 2Orsay/FR

Purpose: The aim of this study was to assess the association between pre-treatment breast MRI features and axillary lymph node
residual disease in women treated for node-positive breast cancer by neoadjuvant chemotherapy (NAC).
Methods or Background: In this single-centre, retrospective study, women with node-positive breast cancer who underwent NAC
and pre-treatment breast MRI between 2016 and 2021 were included. MRIs were evaluated using the standardised BI-RADS and T2-
weighted Breast Oedema Score. Univariate analysis and multivariate logistic regression analysis were conducted to evaluate
clinicopathological and MRI variables association with lymph node residual disease. A prediction model was developed from the
logistic regression analysis and evaluated on a randomly split train and test set (7:3 ratio).
Results or Findings: Of 142 breast cancers, 59% achieved post-NAC nodal response, varying by subtype: luminal 24%, HER2+ 69%,
and TN 75%. Factors associated with nodal response were TN and HER2+ subtypes, high Ki67, and tumour-infiltrating lymphocytes.
Univariate analysis identified MRI features like anterior third location, indicating the depth of the tumour within the breast, and
irregular shape as significant for residual axillary disease. In multivariate analysis, the anterior location and the absence of
intratumoural T2 hyperintensity remained significant. Adding MRI features to anatomopathological variables enhanced nodal residual
disease prediction models.
Conclusion: Luminal subtypes, low Ki67 levels, anterior third location, and lack of intratumoural T2 hyperintensity are independently
associated with axillary residual disease and provide additional predictive value to predict lymph node residual disease after NAC
Limitations: Study limitations include a single-centre design, a retrospective nature, and a limited sample size for histomolecular
subtype subgroup analysis.
Ethics committee - additional information: This study was approved by the relevant Institutional Reveiw Board.
235
OC - S.H.O.C.K.

Join us for this exceptional highlight of ECR 2024 as Congress President Carlo Catalano invites attendees to experience the unlimited
potential of Next Generation Radiology. In an ECR first, he will share the stage with a co-host, Ameca, the world's most advanced
robotic humanoid. Representing the forefront of human-robotics technology, Ameca will share its unique view on the evolution of
medical imaging from the perspective of an artificially intelligent being.
With many more extraordinary surprises awaiting attendees, the ECR 2024 Opening Ceremony S.H.O.C.K. will truly live up to its name.
Programme
Dance Performance
Zurcaroh
Opening Speech
Carlo Catalano & Ameca
Interlude: Act 1
The Next Generation of Radiologists

Nino Bogveradze
Recognition of Honorary Members
Interlude: Act 2
Acrobatic Performance
Recognition of Gold Medallists
Closing Words
Carlo Catalano & Ameca
Interlude: Act 3
Moderator:
236
ST 7 - Daily Wrap-up
Categories: General Radiology

Join our studio moderators as they look back on the day’s highlights and offer a glimpse of what’s still to come at ECR 2024.
Moderator:
Interview (30 min)
237

Moderators:
Ben Giese; Chicago / United States
Interview (30 min)

238
IF 7 - The collection of high-quality data to benefit the patient, the radiologist, and at a
multidisciplinary level
Categories: Audit, Education, Evidence-Based Imaging, Management/Leadership, Multidisciplinary

CME Credits: 1
Radiology is integral to most patients' journeys, from screening to diagnosis and treatment. More of these steps are guided by data -
clinical and imaging - that needs to be of the highest possible quality to ensure the best outcomes in more and more data-centric
healthcare. In this session, these topics will be discussed by the patients and from the clinicians' and radiologists' points of view.
Moderator:

The role of Specialist Centres of Excellence for patient benefit (15 min)
Steve Ebdon-Jackson; Reading / United Kingdom
Emanuele Neri; Pisa / Italy
Radiology and MDTs: experience and benefits for the patients (15 min)
Isabel Molwitz; Hamburg / Germany
Are we ready for novel screening programmes with advanced technology? (15 min)
Katharina Beyer; Rotterdam / Netherlands
Panel discussion: How to integrate patients' associations and technology innovation? (10 min)
239
RC 717 - Pros and Cons: dual-energy/spectral CT in an emergency is helpful
Categories: Emergency Imaging, Imaging Methods, Multidisciplinary

CME Credits: 1
Moderator:
Francesco Macri; Geneva / Switzerland

Francesco Macri; Geneva / Switzerland
Pro (15 min)

Savvas Nicolaou; Vancouver / Canada
This house believes that dual-energy/spectral CT in an emergency is helpful.
Con (15 min)

This house believes that dual-energy/spectral CT in an emergency is not helpful.
Panel discussion: Dual-energy/spectral CT: a step towards an expanding future of CT imaging (25 min)
240
OF 7T - Next-generation diagnosis: working across radiology-pathology boundaries
Categories: Education, General Radiology, Oncologic Imaging, Professional Issues, Students

CME Credits: 1
Moderator:

How imaging augments pathology diagnosis (30 min)

Dilani Lokuhetty; Lyon / France
Ian A. Cree; Badby / United Kingdom
1. To discuss how radiology has contributed to pathology from the WHO-IARC perspective.
2. To discuss how young radiologists can get involved in the next update of the WHO tumour classification.
How imaging augments pathology diagnosis (15 min)

1. To explain how imaging and pathology complement each other in the diagnostic pathway.
2. To appreciate how imaging and pathology may be synergistic through case examples.
Open forum discussion: New ways of working towards improving patient diagnosis (10 min)
241
RC 701 - Pancreatic IPMN: leave alone or when to be concerned?
Categories: Abdominal Viscera, Imaging Methods

CME Credits: 1
Moderator:
Celso Matos; Lisbon / Portugal

Celso Matos; Lisbon / Portugal
Current strategies in detection and follow-up of IPMN (15 min)

Giulia Zamboni; Verona / Italy
1. To list and describe the international guidelines that address diagnosis and follow-up of IPMN.
2. To analyse the guidelines and reflect on the impact on patient prognosis and health economics.
3. To reflect on the use of abbreviated protocols for screening and follow-up.
Worrisome and high-risk features of IPMN (15 min)

1. To identify worrisome and high-risk features in IPMN at cross-sectional imaging.
2. To consider possible pitfalls and differential diagnosis.
3. To reflect on clinical implications.
Indication and findings in EUS of IPMN (15 min)

Christian Jenssen; Strausberg / Germany
1. To become familiar with a 4-step algorithm for EUS-based diagnosis and risk assessment of IPMN.
2. To know the conspicuous EUS features of different types of IPMN.
3. To know when to be concerned: how to confirm high-risk stigmata of IPMN on EUS.
Panel discussion: What would be the justification for performing life-long follow-up? (10 min)
242
HW 7R - Advanced image post-processing and workflows
Categories: Imaging Methods, Radiographers

CME Credits: 1

1. To describe best practices, tips and tricks for advanced image post-processing and workflows in CT/MRI.
2. To demonstrate and apply best practices, tips and tricks for advanced image post-processing and workflows in CT/MRI.
Effective post-processing in radiology examinations: presentation, live software demonstration and Q&A (30 min)
Managing large multi-modality image datasets: presentation, live software demonstration, and Q&A (30 min)
243
E³ 25B - The post-interventional aorta
CME Credits: 1
Moderator:
Justus Erasmus Roos; Luzern-16 / Switzerland

Justus Erasmus Roos; Luzern-16 / Switzerland
Chest (23 min)

Nils Robrecht Planken; Amsterdam / Netherlands
1. To discuss the role of CT and MR in imaging the post-interventional thoracic aorta.

2. To show examples of common thoracic aortic post-operative findings.
3. To discuss the essential elements of the radiology report in patients with prior thoracic aortic interventions.
Abdomen (23 min)

1. To discuss the role of CT and MR in imaging the post-interventional abdominal aorta.

2. To show examples of common abdominal aortic post-operative findings.
3. To discuss the essential elements of the radiology report in patients with prior abdominal aortic interventions.
244
EIBALL 7 - Lung cancer screening: implementing quantitative imaging biomarker protocols
Categories: Chest, Imaging Methods, Oncologic Imaging, Physics in Medical Imaging, Radiographers
ETC Level: LEVEL II
CME Credits: 1
Moderator:

Status of lung cancer screening in Europe (15 min)

Torsten Gerriet Blum; Berlin / Germany
1. To learn about the status of the EU's lung cancer screening programme decision and plans to implement it in the 27 member
states.
2. To appreciate all the necessary contributors for successfully implementing national screening programmes.
3. To understand the scope and challenges that are coming with the implementation of lung cancer screening in Europe.
Protocol and quality control programmes in European lung cancer screening programmes (15 min)
Hans-Ulrich Kauczor; Heidelberg / Germany
1. To learn the specifics of imaging protocols and QA programmes necessary to benefit all European citizens eligible for lung cancer
screening.
2. To appreciate ESR's role in planning and implementing screening programmes in the respective member states.
3. To understand in detail the imaging solutions and challenges of the lung cancer screening QA programme.
QIBA profile on small lung nodule volume assessment and monitoring in low-dose CT (15 min)
Gudrun Zahlmann; Oak Brook / United States
1. To learn how to implement imaging procedures to make lung nodule volumetric assessments reliable and reproducible.
2. To appreciate imaging centres' critical role and responsibilities in performing low-dose CT imaging for screening.
3. To understand which devices and personnel are needed and what performance measures are required to achieve reliable and
reproducible nodule assessments.
Panel discussion: How can we accelerate the implementation of robust QA programmes in lung cancer screening? (10
min)
245
ISRRT - Global education perspectives
Categories: Education, Management/Leadership, Multidisciplinary, Professional Issues, Radiographers

CME Credits: 1
The session aims to showcase the appreciation of the benefits of multidisciplinary collaborative practice. The attendees will become
familiar with the current global status of education in the medical radiation science professions and be introduced to the ISRRT global
network of education and training for the medical radiation science professions.
Moderators:
Haakon H. Hjemly; Sorumsand / Norway

Haakon H. Hjemly; Sorumsand / Norway
Multidisciplinary collaboration: the experience of an Australian paediatric hospital (25 min)

Stephen Lacey; Parkville / Australia
Global network on education and training for radiographers and radiation therapists: ISRRT perspectives (25 min)
Napapong Pongnapang; Bangkok / Thailand
246
OF 7R - Keeping patients at the heart of radiography and radiotherapy practice
Categories: Evidence-Based Imaging, Imaging Methods, Management/Leadership, Professional Issues, Radiographers

CME Credits: 1
This session underscores the enduring commitment of radiographers and healthcare professionals to prioritise the well-being and
experience of patients in the fields of medical imaging and radiotherapy. It will comprise three distinctive talks, each offering valuable
insights and strategies that can be useful to achieving person-centred care in daily practice. In this session, radiographers and
healthcare professionals will be reminded about the importance of centring their focus on the patient, acknowledging that amidst the
ever-evolving landscape of radiography and radiotherapy, the human element remains paramount.
Moderator:
Laura Dolenc; Ljubljana / Slovenia

Laura Dolenc; Ljubljana / Slovenia
Opportunities and challenges for person-centred radiography (16 min)

Susanne Holm; Faaborg / Denmark
Heartfelt conversations: the art of compassionate communication (16 min)

Emma Hyde; White - British / United Kingdom
How technology and new devices can support care and practice in radiotherapy (16 min)
Claudio Votta; Roma / Italy
247
EIBIR 7 - Artificial intelligence (AI) for health imaging: pioneering cancer image repositories for
diagnosis and analysis
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Oncologic Imaging, Research
ETC Level: LEVEL II
CME Credits: 1
Moderator:

CHAIMELEON: accelerating the lab to market transition of AI tools for cancer management (12 min)
Alejandro Vergara; Valencia / Spain
1. To learn how to build pan-cancer imaging and multi-omics data registries in a public-private collaborative environment.
2. To understand how to overcome challenges related to image quality heterogeneity across European sites.
3. To appreciate the main challenges to be solved by the AI community in different types of solid tumours.
EuCanImage: towards a European cancer imaging platform for enhanced AI in oncology (12 min)
Maciej Bobowicz; Gdansk / Poland
1. To learn about European infrastructures for large-scale biomedical data.

2. To appreciate how these infrastructures can be leveraged to build secure cancer imaging repositories.
3. To understand how these cancer imaging repositories can enhance AI in cancer imaging.
INCISIVE: a federated data infrastructure enabling AI-supported cancer diagnosis and prediction (12 min)
Gianna Tsakou; Marousi / Greece
1. To elaborate on benefits and challenges related to decentralised data storage.

2. To elaborate on how health data sharing empowers AI-supported tools for cancer diagnosis and prediction.
3. To highlight the role of healthcare professionals in the development of AI tools.
ProCancer-I: AI models of prostate cancer diagnosis (12 min)

Nikolaos Papanikolaou; Lisbon / Portugal
1. To elaborate on challenges in a multicentric setup.

2. To present initial results focusing on AI-powered prostate cancer detection and characterisation.
3. To discuss safe translation to the clinics.
EUCAIM: integrating the AI for health imaging results into the European Cancer Imaging Initiative (6 min)
248
1. To learn how a pan-European digital federated infrastructure of cancer-related images and data can be used for the development of
AI tools toward precision medicine.
2. To appreciate the seamless access to de-identified, high-quality real-world data, to foster collaboration among clinicians,
researchers, and innovators.
3. To understand how AI data-driven decisions can be designed in diagnosis and treatment.
249
RC 712 - Bone marrow findings
Categories: Musculoskeletal, Paediatric

ETC Level: LEVEL II
CME Credits: 1
Moderator:
Maria Beatrice Beatrice Damasio; Genoa / Italy

Maria Beatrice Beatrice Damasio; Genoa / Italy
Frequent and less frequent findings in healthy children (15 min)
1. To describe the normal pattern of bone marrow conversion in the maturing skeleton.
2. To list the most common sites of residual red bone marrow.
3. To name at least three sites of physiological/self-limiting bone marrow changes in healthy children.
Optimal protocol: how to increase specificity and when to use extra sequences? (15 min)
Lil-Sofie Ording Müller; Oslo / Norway
1. To learn about the basic imaging protocol for bone-marrow imaging with a focus on whole-body MRI.
2. To describe the most relevant additional MRI sequences in equivocal bone marrow findings.
3. To discuss when and how to apply extra sequences in bone marrow imaging.
When to biopsy, when to follow, when to ignore (15 min)

Laura Tanturri De Horatio; Roma / Italy
1. To describe the main focal and diffuse bone marrow diseases in children.
2. To distinguish in which cases biopsy is necessary and in which follow-up is appropriate.
3. To recognise the bone marrow findings that do not require any monitoring.
Panel discussion: How to improve differential diagnosis between variants and pathology? (10 min)
250
RC 708 - Imaging of the face
Categories: Head and Neck, Imaging Methods

CME Credits: 1
Moderator:

Facial fractures (15 min)

Elizabeth Loney; Bradford / United Kingdom
1. To simplify reporting of complex maxillofacial trauma.

2. To review the principle of facial buttresses with respect to surgical outcomes.
3. To describe commonly used classification systems such as Le Fort.
Facial pain (15 min)

Christine Mary Glastonbury; San Francisco / United States
1. To understand the distinction between trigeminal neuropathy and trigeminal neuralgia.

2. To review imaging findings of neurovascular conflict.
3. To review cases that may manifest with trigeminal symptomatology.
Facial cosmetic implants and grafts (15 min)

Simonetta Gerevini; Cremona / Italy
1. To review the different types of cosmetic facial implants.

2. To present the expected imaging findings and complicated radiologic findings after implantation.
3. To discuss possible complications from facial implants
Panel discussion: Recommendations for image optimisation (10 min)
251
E³ 726 - Emerging non-surgical treatment complications in neuroradiology
Categories: Neuro, Oncologic Imaging

CME Credits: 1
Moderator:
Sven Haller; Geneve / Switzerland

Radiotherapy-related side effects on the healthy brain (15 min)

Jasmina Boban; Novi Sad / Serbia
1. To review the current knowledge of the mechanism for the radiotherapy-induced and radiotherapy-related side effects on the
healthy brain and to introduce the term "radiobrain".
2. To provide the overview of the typical imaging findings of the radiotherapy-related side effects.
3. To present the pitfalls and common mistakes in interpreting these effects.
4. To provide a short insight into current research perspectives in the field.
Neurologic complications of immunotherapy (15 min)

Teresa Santos Nunes; Lisboa / Portugal
1. To describe the neurological complications associated with immunotherapy.

2. To summarise the imaging features of neurologic immune-related adverse events.
3. To discuss the challenges in interpreting neuroimaging findings during immunotherapy.
Complications of new drug treatments in neurodegeneration (15 min)

Frederik Barkhof; Amsterdam / Netherlands
1. To become familiar with novel antibody treatments for Alzheimer's disease.

2. To recognise the features of Amyloid-Related Imaging Abnormalities (ARIA).
3. To be able to communicate ARIA findings in the setting of treatment monitoring.
Panel discussion: Challenges in imaging interpretation of treatment-related complications (10 min)
252
RC 715 - Imaging and interventional treatment of atherosclerotic lesions: old topic, new discoveries
Categories: Interventional Radiology, Multidisciplinary, Vascular

CME Credits: 1
Moderator:
David Laszlo Tarnoki; Budapest / Hungary

David Laszlo Tarnoki; Budapest / Hungary
Modern ultrasound imaging of atherosclerotic lesions (15 min)

Anita Hernyes; Budapest / Hungary
1. To summarise the modern US techniques: microvascular flow, automated IMT measurement, plaque categorisation, CEUS and
vascular stiffness assessment
2. To describe various atherosclerotic lesions with the US: IMT complex, increased IMT, plaque volume etc.
3. To learn what to include in your ultrasound report.
Role of CTA and MRA in atherosclerosis imaging (15 min)

1. To understand the technical protocol of CTA and MRA for carotid and lower extremities.
2. To understand the role of AI in CTA and MRA.
3. To understand the essential protocol for angio-CT and MRA in the context of atherosclerotic plaques.
4. To learn what to include in your CTA and MRA report in case of urgent and non-urgent examinations.
Management of carotid and femoral stenosis: the role of interventional radiology (15 min)
Christoph Johannes Zech; Basel / Switzerland
1. To name the relevant materials, technologies and infrastructure needed for the IR management of femoral stenoses and
occlusions.
2. To summarise and compare IR treatments' results.
3. To understand possible complications of the minimal-invasive treatment.
Panel discussion: The radiologists in the multidisciplinary management of atherosclerotic lesions: what should we
report, all plaques? How often should we control the non-significant stenosis? Should we screen atherosclerosis in the
high-risk population? (10 min)
253

CME Credits: 1


254
RC 716 - Breast imaging: back to basics
Categories: Breast, Imaging Methods, Oncologic Imaging

ETC Level: LEVEL I
CME Credits: 1
Moderator:

Breast MRI (15 min)

Isabelle Thomassin-Naggara; Paris / France
1. To understand indications for breast MRI

2. To learn about MRI dedicated protocol and technical advantages.
3. To investigate future opportunities for breast MRI.
CEM (15 min)

1. To understand indications for CEM

2. To understand the advantages of contrast-enhanced mammography.
3. To learn about the drawbacks of CEM.
BIRADS (15 min)

Paola Clauser; Vienna / Austria
1. To be familiar with the BIRADS lexicon.

2. To learn about classification categories of breast lesions in all modalities.
3. To investigate how to write a structured report.
Panel discussion: Future development of breast imaging (10 min)
255
RPS 709 - Novel treatment options for malformations and benign tumours
Categories: Abdominal Viscera, Head and Neck, Interventional Oncologic Radiology, Interventional Radiology, Paediatric, Vascular
CME Credits: 1
Moderator:
Rok Dežman; Ljubljana / Slovenia
Polidocanol sclerotherapy of venous malformations in children- efficacy, safety, and predictors of outcome and
complications (7 min)
Niclas Schmitt; Heidelberg / Germany
Author Block: N. Schmitt, J. Lorenz, S. Hohenstatt, F. Ruping, P. Günther, M. Bendszus, M. Möhlenbruch, D. F. Vollherbst;
Heidelberg/DE
Purpose: Venous malformations (VMs) represent the most common type of congenital vascular malformations and feature a wide
range in size, location, appearance, and clinical symptoms. Percutaneous sclerotherapy (PS) using polido-canol is an established
treatment, aiming at an improvement of the clinical symptoms. The influence of patient- and disease-specific factors on treatment
success and complications in children has not been systematically addressed so far.
Methods or Background: All children (<18 years) with VMs who were treated with PS using polidocanol between 01/2011 and
11/2021 in our institution were included. Demographics, clinical data and radiological features were analysed and the influence of
patient- and VM-related factors on the outcome and complications according to the CIRSE classification were investigated.
Results or Findings: A total of 83 children were included with a symptom improvement in 54.0%, stable symptoms in 34.9%, and
worsening in 11.1%. The overall complication rate was 13.3% with an absence of permanent complications. Complications were more
frequently in children after the treatment of extremity-located VMs (p=0.01; complication rate 12.2% for VMs of the extremities and
1.6% for VMs in other locations). This suggests that VMs located along body parts with a more intense movement may increase the
probability for periprocedural complications with a further impact by the increased venous pressure in the lower extremities. There
was no significant factor affecting the clinical outcome with a tendency for a worse outcome in children suffering from VM-related
swelling (p=0.11).
Conclusion: PS of VMs using polidocanol can be an effective treatment option in children with an acceptable safety profile.
Complications can be more frequently expected for VMs of the extremities and should be considered for therapy planning.
Limitations: This was a retrospective single-centre analysis, which included a disproportionate incidence of head and neck VMs and
assessed outcomes subjectively by the patients themselves.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Medical Faculty at
Heidelberg University.
Long-term outcomes of hand arteriovenous malformation management: single referral vascular anomalies centre over
25 years’ experience (7 min)
Yong Deok Lee; Seoul / Korea, Republic of
256
Author Block: Y. D. Lee, S. Y. Lee, K. B. Park, D-I. Kim, S. S. Yang, Y. S. Do; Seoul/KR
Purpose: Hand arteriovenous malformations (AVMs) are extremely difficult to manage for their functional importance and cosmetic
problems. A single centre retrospective study was conducted to identify long-term outcomes of multidisciplinary team management of
hand AVM.
Methods or Background: Retrospective vascular anomalies centre data was reviewed from 1995 to 2022. Patient’s demographics,
Schobinger’s AVM stage, sclerotherapy details, surgical history, and adverse events after sclerotherapy were reviewed.
Results or Findings: 150 hand AVM patients visited our hospital from 1995 to 2022. The mean age was 33 years (range, 1-75 years)
and 91 of 150 were female. 44 were Schobinger stage II and 106 were stage III. 101 patients (67%) received a total of 319 sessions of
percutaneous sclerotherapy. Angiographic devascularisation rates after sclerotherapy were as follows: 16 had 100%
devascularisation, 30 had over 90% devascularisation, 34 showed 50-90% devascularisation, 15 showed 0-50% of devascularisation,
and six showed aggravation. 123 of 319 of (39%) had sclerotherapy-related adverse events (112 were minor and 11 were major
adverse events). 15 of 101 (15%) patients eventually received amputation surgery after mean 1618 days (range, 3-5444 days) after
sclerotherapy (sclerotherapy-related necrosis (n=3) and delayed amputation (n=12)). 13 of 150 (9%) of patients received primary
surgical amputation for ulcer or bleeding (all Schobinger stage 3). The remaining 36 patients (34%) followed without any procedure.
Conclusion: In our study, 67% of hand AVM patients (101/150) were initially treated with sclerotherapy and 9% of patients (13/150)
by surgery. 80% of patients showed response to sclerotherapy (over 50% devascularisation rate) with a major complication rate of
7%. However, 15% of patients eventually received amputation surgery after sclerotherapy.
Limitations: The limitations of the study are its retrospective nature and single-centre experience.
Ethics committee - additional information: Ethical approval was not sought for the present study because this was a
retrospective observational study conducted at the Samsung Medical Centre.
Analysis of adverse events after peripheral arteriovenous malformation embolisation: single Korean referral vascular
anomalies centre experience (7 min)
Jiyoon Ban; Gangnam-gu, Seoul / Korea, Republic of
Author Block: J. Ban1, S. Y. Lee1, K. B. Park1, D-I. Kim1, S. S. Yang1, Y. S. Do2; 1Seoul/KR, 2Gyeonggi/KR
Purpose: Peripheral arteriovenous malformations (AVMs) are extremely difficult to manage for their functional importance and
cosmetic problems. This single-centre retrospective study was conducted to identify factors that contribute to adverse events (AEs)
after embolisation of AVMs using percutaneous sclerotherapy.
Methods or Background: A retrospective review of patients who underwent sclerotherapy of AVMs from 1997 to 2022 was
performed. 1273 procedures in 444 patients (median age: 28.0 years; interquartile range: 18.0-38.0 years) were evaluated. The AVM
characteristics and details of AEs were analysed. According to Society of Interventional Radiology guidelines, AEs were divided into
minor and major. Univariate and multivariate analyses were performed to find factors associated with AEs.
Results or Findings: Ethanol (94.2% of procedures, mean 26.2ml per session) and coils (33.4% of procedures, mean 31.4 coils)
were mainly used for sclerotherapy. Upper extremity AVMs were most common (39.6%), followed by lower extremity (33.9%),
abdomen-pelvis (22.4%), and thorax and neck (4.1%). 297 of 1273 procedures (23.3%) had sclerotherapy-related adverse events
(261 were minor and 36 were major adverse events). Skin damage and transient nerve injury were most frequent in AEs. Coil used
OR, 0.3672; 95% CI, 0.2607-0.5172, P<0.0001), diffuse AVMs (OR, 0.6837; 95% CI, 0.5095-0.9174; P=0.0112), body AVMs (OR, 0.522;
95% CI, 0.3685-0.7393; P=0.0003), and skin non-involvement (OR, 0.5741; 95% CI, 0.4268-0.7721; P=0.0012) were negatively
associated with AEs.
Conclusion: In our ethanol and coil based AVM sclerotherapy, 297 AEs were developed after 1273 sessions of AVM sclerotherapy
(23.3%). Most AEs were minor. Use of coil, diffuse involvement, body AVM and skin non-involvement showed protective effect on AEs.
Limitations: Our study was limited to a single centre and based on retrospective review of previous procedures.
Ethics committee - additional information: Ethical approval was not sought for the present study because this was a
retrospective study conducted at Samsung Medical Center with an observational nature.
Multicentred patient-reported outcome of bleomycin electrosclerotherapy in slow-flow malformations (7 min)

Vanessa Franziska Schmidt; Munich / Germany
257
Author Block: V. F. Schmidt1, L. Meyer2, R. Brill3, D. Puhr-Westerheide1, O. Öcal1, J. Ricke1, M. Seidensticker1, W. A. A. Wohlgemuth3,
M. Wildgruber1; 1Munich/DE, 2Eberswalde/DE, 3Halle a.d. Saale/DE
Purpose: The aim of this study was to evaluate the safety and clinical outcome of bleomycin electrosclerotherapy (BEST) for treating
extracranial slow-flow malformations.
Methods or Background: A multicentre cohort of 233 patients with symptomatic slow-flow malformations was retrospectively
investigated. Patient records were analysed with respect to procedural details and complications. A treatment-specific, patient-
reported questionnaire was additionally evaluated, obtained 3-12 months after the last treatment, to assess the subjective outcome
including mobility, aesthetic aspects, and pain (using a visual analogue scale VAS) as well as the occurrence of postprocedural skin
hyperpigmentation. All outcome parameters were compared according to patients' age.
Results or Findings: In total, 325 BEST treatments were performed with variable electrodes after an intravenous/ intralesional
Bleomycin injection. The mean number of procedures per patient was 1.4 (±0.7). The total complication rate was 10.8% (33/325)
including 29/352 (8.9%) major complications. Patient-reported mobility decreased in 10/133 (8.8%), was stable in 30/113 (26.5%),
improved in 48/113 (42.5%), and was rated symptom-free in 25/113 (22.1%) patients. Aesthetic aspects were rated impaired
compared to baseline in 19/113 (16.8%), stable in 21/133 (18.6%), improved in 62/113 (54.9%), and perfect in 11/133 (9.7%)
patients. Postprocedural skin hyperpigmentation occurred in 78/113 (69%) patients, remaining unchanged in 24/113 (30.8%), reduced
in 51/113 (65.5%), and completely resolved in 3/113 (3.8%) patients. The median VAS pain scale was 4.0 (0-10) preprocedural and
2.0 (0-9) postprocedural. Children/ adolescents (0-15 years) performed significantly better in all outcome parameters compared than
adults (≥16 years) (mobility, p=0.011; aesthetic aspects, p<0.001; pain, p<0.001).
Conclusion: BEST is an effective treatment for slow-flow vascular malformations, with few but potentially significant major
complications. Children seem to benefit more compared to older patients, suggesting that the therapy should not be restricted to
adults.
Limitations: The study was retrospective and included only a moderate response to the patient questionnaire.
Ethics committee - additional information: This study was approved by the local ethics Committee at the University Hospital,
LMU Munich, protocol No. 23-0035, 01/16/2021), and was performed following the 1964 Helsinki declaration and its later
amendments.
Ultrasound-guided ethanol ablation of medial and lateral neck cysts (7 min)

Katya M Duvivier; Amsterdam / Netherlands
Author Block: K. M. Duvivier, C. Peeters, S. Eerenstein, P. De Graaf; Amsterdam/NL

Purpose: The purpose of this study was to evaluate the treatment effect, cosmetic aspects, and the quality of life (QOL) of ethanol
ablation (EA) as a minimally invasive treatment of TGDC and BCC neck cysts.
Methods or Background: Between February 2022 and July 2022, 21 patients with either a TGDC or BCC neck cyst were
prospectively included in this study. EA was performed in 21 patients. Treatment was performed following a standard protocol
including ultrasound (US) guided cyst aspiration and rinsing with saline followed by injection of 96% ethanol. Patients received a
questionnaire both before EA and after three months of follow-up. Therapeutic outcomes, including volume reduction ratio,
therapeutic success rate, cosmetic results, and complications were evaluated.
Results or Findings: Twenty-one patients were included who underwent EA for TGDC (N=12, 57%) and BCC (N=9, 43%). The mean
age was 41.1 years old (IQR:33.5-51.5) with 6 males (29%) and 15 females (71%). Of the 21 eligible patients, 20 completed the
questionnaire pre-EA and post-EA. The mean volume of the cyst prior to treatment was 10.9cc (IQR: 17.01-2.79) and after EA-
treatment was 1.20cc (IQR: 0.11-1.42), wich results in a mean volume reduction of 81.86% (IQR:83.50-98.00). Two patients required a
second treatment. No significant complications occurred during follow-up. 89% (N=16/18) Of the patients show an overall
improvement of the QOL. The VAS-score during the procedure had a mean of 1.4 (N=21) and after the procedure of 0.7 (N=21) on a
0-10 pain scale.
Conclusion: EA is an effective, safe, and minimally invasive method of treating TGDC and BCC neck cysts . The procedure is well
tolerated and can be repeated, when necessary, without complications or hindering any surgery.
Limitations: The limitations of this current study are the small sample size and the short follow-up period.
Transarterial embolisation of renal angiomyolipomas: a prophylactic treatment for reducing myogenic components (7
min)
Lorenzo Braccischi; Bologna / Italy
258
Author Block: L. Braccischi1, L. Bellini2, F. Modestino1, L. Bartalena1, M. C. Galaverni1, M. Renzulli1, A. Cappelli1, L. Bianchi1, C.
Mosconi1; 1Bologna/IT, 2Rome/IT
Purpose: Renal angiomyolipomas (AMLs) are benign hamartomatous tumors containing elements of smooth muscle, fat and vascular
tissue. 70% of all AMLs are sporadic and tend to be solitary. The most concerning complication of AMLs is the risk of bleeding which is
linked to dimensions and can lead to retroperitoneal haemorrhage. Selective embolisation of the renal artery has been approved as a
prophylactic treatment option. Nevertheless, there is still a lack of literature. The aim of this study was to evaluate the short and long-
term efficacy of AML embolisation. The percentage of fat and myogenic component was calculated in each AML and compared in
volume before and after the embolisation of each component was evaluated.
Methods or Background: A total of 26 AMLs were treated between May 2015 and February 2023. All lesions were completely
embolised according to the angiographic criteria of vascular stasis and the absence of arterial feeders. Cyanoacrylate glue was the
most commonly used embolic agent.
Results or Findings: The efficacy of embolisation was determined over a mean follow-up period of 30.5 months. Mean lesion size at
diagnosis was 56.85mm with a pre-treatment mean percentage of myogenic component around 60%. Embolisation of the renal AMLs
was technically successful for all 26 lesions. Long-term follow-ups showed a mean diametre reduction of 23.57mm with a 50%
reduction in the volume of the myogenic component.
Conclusion: Transarterial embolisation is effective in reducing the volume of the myogenic component of AMLs, therefore lowering
the chance of haemorrhagic events while preserving renal function. No major complications were observed demonstrating the highly
safe profile of the procedure.
Long-term outcomes after percutaneous cryoablation of abdominal wall endometriosis (7 min)

Philippe Maas; Bordeaux / France
1 1 1 1 1 1 2 1 1
Author Block: P. Maas , E. Jambon , I. Molina Andreo , Y. Le Bras , N. Grenier , J-L. Brun , F. H. Cornelis , C. Marcelin ; Bordeaux/FR,
2
New York, NY/US
Purpose: The purpose of this study was to retrospectively evaluate the long term outcomes after percutaneous cryoablation of
abdominal wall endometriosis (AWE).
Methods or Background: In this IRB-approved retrospective observational study, fourty patients (median age, 37 years;
interquartile range [IQR], 32-40 years) presenting with 52 symptomatic AWE, treated with cryoablation between January 2013 and
May 2022 and with at least 12 month follow-up, were reviewed. Thirteen patients (32%) presented with deep endometriosis.
Outcomes were assessed using the visual analogue scale (VAS) and MR imaging. Pain-free survival rates were calculated using the
Kaplan-Meir method. Adverse events were analysed and graded according to the CIRSE classification.
Results or Findings: The median follow-up was 40.5 months (IQR, 26.5-47.2 months). Median VAS score before cryoablation was 8
(IQR, 7-9). Eight patients had residual pain after cryoablation (median 5, IQR 3-5.5, p=0.0008): 5 patients (12.5%) immediately, and 3
others (7.5%) recurred after 12 months. Five patients (12.5%) underwent a second cryoablation, one patient (2.5%) surgery, and two
patients (5%) did not wish to undergo further treatment as they considered the pain reduction adequate. The secondary efficacy of
cryoablation to control pain was 87.5% (35/40). Complete symptom relief was recorded in 80% (32/40) of patients from the first
cryoablation at the time of last follow up. The median pain-free survival rates were 89.2% (95% CI, 70.1-96.4) at 36 months, and
76.8% (55.3-83.8) after 60 months, respectively. No prognostic factor of failure according to patient and lesion characteristics was
identified.
Conclusion: Cryoablation is safe and effective in achieving long-term pain control of abdominal wall endometriosis.
Ethics committee - additional information: This study was approved by the Institutional Research Board at Bordeaux University.
Image guided cryoablation in extra abadominal desmoid tumors (7 min)

Faisal Syed; Mumbai / India
259
Author Block: F. Syed; Mumbai/IN
Purpose: The aim of this paper was to determine the efficacy and safety of cryoablation in patients unresectable with desmoid
tumors which were progressed on chemotherapy retrospectively at a single institution over a year-long period.
Methods or Background: A retrospective, single-institution review identified ten patients (ages 4-45) with extra-abdominal desmoid
tumors who received CT-guided percutaneous cryoablation salvage treatment, totalling fourteen sessions in 2021. Median maximal
lesion diameter was 6cm. Intent was palliative in all patients. Contrast-enhanced cross-sectional imaging was obtained before and
after treatment in addition to a routine clinical follow-up.
Results or Findings: Technical success was achieved in all patients. The median follow-up was 5 months. Symptomatic
improvement was demonstrated in 85% of patients. At 3 months, the average change in viable volume was -75% and response by
modified response evaluation criteria in solid tumors (mRECIST) was PR 70%, and SD 30% . No rapid postablation growth or track
seeding was observed. No major procedure-related complications were noted.
Conclusion: Cryoablation for desmoid tumors demonstrates a high degree of symptom improvement and local tumor control on early
follow-up imaging with relatively low morbidity.
Limitations: This study was retrospective.
260
RPS 711 - Stroke: treatment and outcome
Categories: Interventional Radiology, Neuro, Research, Vascular

CME Credits: 1
Moderator:
Mohamed T. El-Diasty; Jeddah / Saudi Arabia
Cost-effectiveness of mobile stroke units in Germany: a perspective on catchment areas, operating modes, and staffing
(7 min)
Fabian Tollens; Mannheim / Germany
Author Block: F. Tollens, C. Hoyer, K. Szabo, J. Rink; Mannheim/DE

Purpose: Estimating the cost-effectiveness of future mobile stroke unit (MSU) services with respect to local idiosyncrasies is essential
for optimising major program determinants and enabling large-scale implementation of MSU services. Therefore, the aim of this study
was to assess the cost-effectiveness of MSU services for varying urban German settings.
Methods or Background: Costs of different operating modes and hours including weekend and non-weekend coverage, including
different personnel configurations, were simulated for the German healthcare system. Ischaemic stroke incidence, circadian
distribution, rates of alternative diagnoses and stroke mimics, as well as missed cases, were incorporated to model case coverage
and patient-level costs of acute stroke care. Based on internationally reported stroke outcomes, a five-year Markov Model was applied
to analyse the cost-effectiveness outcomes for varying catchment zone populations.
Results or Findings: Compared to regular emergency medical services, stroke care by MSU achieved additional 0.06 QALYs on
average over a five-year time horizon. Assuming a catchment zone of 400,000 inhabitants and 12 hour/ seven day coverage resulted
in an incremental cost-effectiveness ratio (ICER) of €42,069 per QALY gained. Lower ICERs were possible when coverage was
expanded to 16 hour-service on 7 days per week. Sensitivity analyses revealed that the population size within the catchment zones
and the number of ischaemic strokes missed by MSU deteriorated economic performance of MSU significantly.
Conclusion: Major determinants of cost-effectiveness should be addressed when setting up novel MSU programs. For the German
healthcare system, the catchment zone should cover at least 400,000 inhabitants and MSU service should cover 12-16 hours per day,
including weekends.
Limitations: This study used a model-based economic evaluation of MSU program costs and stroke case estimates.
Ethics committee - additional information: There was no ethical approval necessary as no human data was analysed.
Set-up and planning of a MSU service in a medium-sized German urban area (7 min)
Johann Rink; Mannheim / Germany
261
Author Block: F. Tollens, J. Rink, C. Hoyer, K. Szabo; Mannheim/DE
Purpose: Mobile stroke units (MSU) equipped with mobile CT scanners have demonstrated major improvements in prehospital stroke
care. Due to the geographical, social, and structural characteristics of the German city of Mannheim within the Rhine-Neckar
Metropolitan Region, concepts of previously established MSU services cannot directly be transferred to Mannheim. The present
analysis aimed to identify major determinants that need to be considered when setting up an MSU service in Mannheim.
Methods or Background: Program costs consisting of hardware investments, CT scanner and contrast media injector, personnel,
and project management were estimated based on hospital accounting and vendors. Local stroke statistics from 2015 to 2021 were
analysed to calculate the circadian distribution of strokes and local incidence rates. Future MSU patient numbers and program costs
were analysed for varying operating modes, daytime coverage models, and staffing configurations. Case coverage and economic
determinants were assessed in sensitivity analyses.
Results or Findings: 54.3% of all stroke patients were admitted during a ten-hour time window on weekdays. When assuming that
about half of all incident stroke patients did not call the emergency centre or were missed at dispatch, an average of 0.8 stroke
patients could be expected in a 10-hour shift each day in Mannheim, which could potentially be increased by expanding the
catchment area. Overall estimated MSU costs amounted to €815,087 per annum for this operation mode. Teleneurological
assessment reduced overall costs by 11.7%.
Conclusion: This analysis provides a framework of determinants and considerations to address during the design process of a novel
MSU program to balance stroke care improvements with the sustainable use of scarce resources.
Limitations: The study is limited by using a model-based economic evaluation of MSU program costs and stroke case estimates.
Ethics committee - additional information: There was no ethical approval necessary as no human data was analysed.
Clinical scores and imaging-based thrombus biomarkers for predicting recanalisation after stent-retriever mechanical
thrombectomy (7 min)
Karina Janeth Gordillo Zabaleta; Barcelona / Spain
Author Block: K. J. Gordillo Zabaleta1, V. Cuba1, O. Chirife1, E. Ripoll1, S. Aixut1, L. Aja1, A. Nuñez1, M. Comas2, J. Puig2; 1Barcelona/ES,
2
Girona/ES
Purpose: Stent-retrievers (SR) are safe and effective devices for acute ischaemic stroke due to large vessel occlusion (LVO).
However, the efficacy of this technique in treating certain patients remains dubious. No imaging biomarkers are available to predict
the efficacy of these devices. The prediction of a fast and complete recanalisation after a single device pass (first-pass effect, FPE)
could improve and optimise decision-making in treating acute stroke. The objective of this study was to identify predictors of
recanalisation after SR mechanical thrombectomy.
Methods or Background: We performed a retrospective observational study with prospective inclusion to assess the efficacy and
safety of the Catchview Maxi stent-retriever device (BALT, Montmorency, France). We included 94 stroke patients with LVO. Location,
length, visual density, attenuation Hounsfield Unit (HU) coefficient, and permeability were the thrombus-related parameters analysed
on non-contrast CT and CT angiography. Collaterals and intracranial vascular calcification were also evaluated. FPE was defined as the
achievement of mTICI2c-3.
Results or Findings: FPE (26 patients, 28.3%) was associated with smoking (p=0.030), greater rates of baseline mRS 0-2 (p=0.008),
lower NIHSS score at baseline (p=0.040), 24-NIHSS score (p<0.001), greater rates of 90-day mRS 0-2 (p=0.003), shorter thrombus
length (p=0.05), and good collaterals (p=0.070). Patients with final mTICI 2c/3 (63 patients, 66%) were younger (median age, 74 vs
80 years; p<0.001), had lower 24-NIHSS score (16 vs 19; p=0.010), higher greater rates of good functional outcome at 90 days (mRS
0-2, 40.7% vs 7.1%; p=0.003), lower 24h-infarct volume (23.9 vs 98 mL; p=0.057), higher thrombus HU ratio (1.3 vs 1.2; p=0.034),
and higher thrombus permeability (enhancement percentage difference, 38.7 vs 21.9; p=0.016).
Conclusion: Combining clinical scores and imaging-based thrombus biomarkers is useful for predicting recanalisation after SR
mechanical thrombectomy.
Limitations: This was a monocentric retrospective study design.
Ethics committee - additional information: This study was approved by BITE-OUT (reference number of approval: PR237/23).
Haemodynamic consequences in endovascular treatment of symptomatic carotid stenosis: a comprehensive analysis (7

min)
José Rodríguez Castro; Oviedo / Spain
262
Author Block: J. Rodríguez Castro1, E. Murias Quintana1, S. Budiño Torres1, J. M. Jiménez Pérez1, M. García Ramos1, E. Uceda Andrés1,
P. García Martínez2, P. Vega Valdés1; 1Oviedo/ES, 2Salamanca/ES
Purpose: The purpose of this study was to assess the prevalence of bradycardia and hypotension during and immediately post-
endovascular intervention in patients with symptomatic carotid stenosis. Additionally, the research explores personal variables,
radiological stenosis characteristics, procedural techniques, and associated medical complications contributing to these symptoms.
Methods or Background: We conducted a retrospective observational cohort study, analysing a registry of 54 patients who
underwent angioplasty and stent placement for symptomatic carotid stenosis at our centre. A thorough collection of clinical and
analytical variables was performed for subsequent statistical analysis.
Results or Findings: Of the sampled patients, 34% (17) experienced haemodynamic depression post-procedure, with only four
presenting symptoms. Notably, three symptomatic patients were smokers, and two consumed over 40 standard drink units (SDUs) of
alcohol daily. The four patients with hypotension had significant contralateral internal carotid artery (ICA) stenosis. Among those with
bradycardia, 10 of 16 had stenosis near the carotid bifurcation. Interestingly, patients developing complications had shorter
hospitalisation times compared to those without.
Conclusion: The incidence of haemodynamic depression in patients with symptomatic carotid stenosis undergoing angioplasty and
stent placement is 34%. Elevated alcohol consumption (>40 SDUs/day) and significant contralateral ICA stenosis increase the risk of
post-treatment hypotension in our series. Characteristics of the stent, angioplasty balloon, use of corticosteroids, and procedural
duration showed no correlation with haemodynamic depression. Development of haemodynamic depression post-angioplasty does
not necessitate prolonged hospitalisation or increase in serious complications.
Limitations: While our study sheds light on significant associations, the retrospective nature and limited sample size may warrant
further investigation for a more comprehensive understanding.
Ethics committee - additional information: Written consent has been obtained from all patients to receive the described
treatment. Patient data has been anonymised. Due to the nature of this study involving a routine interventional procedure, specific
consent from the Hospital Ethics Committee was not required.
Post-stroke cerebral blood flow derived from arterial spin labelling as a biomarker of clinical outcome following
endovascular therapy (7 min)
Nico Sollmann; Langenau / Germany
Author Block: M. R. Hernandez Petzsche1, G. Hoffmann1, C. Zimmer1, C. Maegerlein1, T. Boeckh-Behrens1, S. Wunderlich1, S.

Kaczmarz1, M. T. Berndt1, N. Sollmann2; 1Munich/DE, 2Ulm/DE
Purpose: Arterial spin labelling (ASL) enables measuring cerebral blood flow (CBF), and it has been proposed to be sensitive to detect
perfusion changes following ischaemic stroke. Post-stroke infarct perfusion has been associated with higher risk of haemorrhagic
transformation (HT) and better clinical outcome. This study aimed to evaluate ASL-based CBF in the infarct core as a prognostic
biomarker for clinical outcome, and to investigate the relationship between CBF increase and HT risk.
Methods or Background: 111 patients (median age: 74 years, 50 men) underwent cerebral MRI (median 4 days after mechanical
thrombectomy for ischaemic stroke due to large vessel occlusion of the anterior circulation) including pseudo-continuous ASL,
diffusion-weighted imaging (DWI), and susceptibility-weighted imaging (SWI). %CBF increase was calculated within the segmented
DWI-positive infarct territory (in relationship to the unaffected contralateral side). Functional independence was defined as a modified
Rankin Scale (mRS) 0-2 at 90 days post-stroke. National Institutes of Health Stroke Scale (NIHSS) scores were determined at
admission.
Results or Findings: In univariate analysis, age (odds ratio (OR)=0.97, p=0.03), pre-stroke mRS (OR=0.36, p<0.001), NIHSS at
admission (OR=0.93, p=0.04), Alberta Stroke Program Early Computed Tomography Score (ASPECTS) at admission (OR=1.26,
p=0.03), complete recanalisation (OR=2.95, p=0.03), DWI-positive infarct volume on post-treatment MRI (OR=0.994, p=0.002), and
%CBF increase (OR=1.009, p=0.04) were associated with functional independence. In multivariate regression, %CBF increase
(OR=1.01, p=0.02), pre-stroke mRS (OR=0.30, p<0.001), and infarct volume (OR=0.99, p=0.001) were significantly associated with
functional independence. Post-stroke infarct %CBF increase was comparable between patients with and without HT (median %CBF
increase 20.6 in patients without HT and 9.8 in patients with HT, p=0.45).
Conclusion: ASL-derived %CBF increase from DWI-positive infarct territories may independently predict functional independence.
Yet, infarct %CBF increase was not significantly associated with increased risk of HT.
Ethics committee - additional information: This study was approved by the Technical University of Munich Ethics Committee.
Automated MRI-based basal ganglia and thalamus assessment at the acute-subacute phase after middle cerebral
cortical stroke can help predict the 3-month mRS shift (7 min)
Salim Zenkhri; Lausanne / Switzerland
263
Author Block: S. Zenkhri, M. Bénédicte, R. Corredor-Jerez, D. Strambo, P. Michel, V. Dunet; Lausanne/CH
Purpose: The purpose of this study was to evaluate the prognostic value of individual volumetric and ADC variations of the basal
ganglia (BG) and thalamus during the acute-subacute phases following a cortical middle cerebral artery (MCA) stroke to predict the
modified Rankin scale (mRS) at 3 months.
Methods or Background: In this retrospective single-centre study, 153 patients (69.3±15.3 years, 85 males) admitted for MCA
cortical stroke at the acute-subacute phase (<21 days) from January 2018 to October 2020 were included. MR performed on a 3T
(MAGNETOM Vida, Siemens) with DWI and T1-MP-RAGE sequences were used for analysis. T1-MP-RAGE images were processed with
the MorphoBox© research application to automatically segment BG and thalamus and compute Z-scores considering the confounding
effects of age and sex. ADC values were extracted from the co-registration between ADC map and segmented T1-MP-RAGE. The
differences between ipsilateral and contralateral values to stroke, and asymmetric index were computed. Stroke volume was
manually segmented on DWI. Multivariate regression analysis was performed to search for predictors of 3-months mRS.
Results or Findings: The mean stroke volume was 8.3±20.5ml. The mean delay between stroke onset and MRI evaluation was
56.5±78.5 hours. The Z-score difference between thalami correlated with the core volume (β=0.29, p<0.001) and delay from stroke
onset (β=0.18, p=0.038). The ADC mean value difference between thalami correlated with the delay from stroke onset (F=6.55,
p=0.0003). Asymmetric Z-score putamen index (F=5.50, p=0.021) and ADC thalamus index (F=5.49, p=0.021) independently
predicted mRS shift at 3 months.
Conclusion: At the acute-subacute phase of MCA cortical stroke, a reduced thalamic and BG volume and low ADC on the stroke side
correlate with core volume and delay from onset. They could additionally help to predict mRS shift at 3 months.
Limitations: This is a retrospective monocentric analysis.
Funding for this study: The authors of this research declare no conflict of interest, and no funding was received for this study.
Ethics committee - additional information: All procedures performed in studies involving human participants were in accordance
with the ethical standards of the institutional and/ or national research committee and with the 1964 Helsinki declaration and its later
amendments or comparable ethical standards. All collected data were anonymised to comply with national ethical guidelines and
laws. Therefore, patients’ consent was waived.
Dual-energy computed tomography can detect subclinical abnormalities of high density in the gyrus in patients after
neuro-endovascular treatment (7 min)
Sangil Suh; Seoul / Korea, Republic of
Author Block: S. Suh, P. Byeongsu, J. H. Han; Seoul/KR

Purpose: Abnormal high density in the gyrus (AHDG) frequently appears in non-contrast brain CT scans post-neuro-endovascular
procedures utilising iodinated contrast. The exact nature of AHDG remains elusive. This study employs Dual-Energy CT (DECT) to
explore whether AHDG could represent a 'minor' subarachnoid haemorrhage, confirming that it is iodine, not blood. We aimed to
elucidate the nature, incidence, and risk factors of AHDG observed in DECT scans post-embolisation of unruptured intracranial
aneurysms.
Methods or Background: A retrospective review of clinical data from 230 patients who underwent endovascular embolisation for
unruptured intracranial aneurysms at our hospital from October 2018 to March 2023 was conducted. Low- and high-energy images
were analysed using a 3-material decomposition algorithm focused on brain parenchyma, haemorrhage, and iodine. The osmolality of
the contrast agents used in the procedures was 290mOsm/kg, 515mOsm/kg, and 616mOsm/kg. Chi-squared tests and logistic
regression analyses were utilised to identify risk factors associated with AHDG.
Results or Findings: DECT images of 230 patients were visually inspected, and three physicians independently analysed them,
ensuring strong interobserver agreement. Consensus was reached through discussion in cases of discrepancy. Among the patients, 96
(41.7%) exhibited AHDG. Chi-squared tests revealed significant relationships between AHDG and variables like sex, aneurysm
location, and usage of higher osmolarity contrast agents. Logistic regression analyses demonstrated significant correlations between
AHDG and age (OR 1.03), procedure time (OR 1.02 per minute), aneurysm location (OR 10.15), and the usage of higher osmolarity
contrast agents (OR 2.70).
Conclusion: DECT effectively detects AHDG in patients with unruptured intracranial aneurysms post-embolisation. The findings
underscore the need for further research to unravel the complex relationships between the osmolarity of contrast agents and the
incidence of AHDG.
Ethics committee - additional information: Ethics committee approval is in progress.
Preoperative clinical and imaging characteristics associated with aneurysm recurrence after stent-assisted coil
embolisation (7 min)
Bing Tian; Shanghai / China
264
Author Block: B. Tian, Y. Hou, X. Tian, J-P. Lu; Shanghai/CN
Purpose: The purpose of this study was to explore the baseline clinical and imaging characteristics which can predict aneurysm
recurrence after stent-assisted coil embolisation.
Methods or Background: This retrospective study included 155 patients with intracranial aneurysms who underwent stent-assisted
coil embolisation in our hospital. All patients underwent preoperative 3D high-resolution MRI vessel wall imaging (HR-VWI) as well as
contrast enhancement MRA (CE-MRA). Regular radiographic follow-up was conducted to monitor for recurrence after embolisation.
Preoperative clinical (including sex, age, symptom, hypertension, and smoking) and imaging (location, maximum size, neck width,
aspect ratio, and aneurysm vessel wall enhancement on HR-VWI) characteristics were recorded. Aneurysm vessel wall enhancement
patterns were classified as circumferential AWE (CAWE), focal AWE (FAWE), and negative AWE (NAWE) groups. Univariable and
multivariable logistic analyses were used to test the clinical and imaging factors associated with aneurysm recurrence after stent-
assisted coil embolisation.
Results or Findings: Among the 155 patients, 26 patients (16.7%) suffered aneurysm recurrence after stent-assisted coil
embolisation during the 2-year follow-up. Aneurysm vessel wall enhancement patterns on preoperative HR-VWI were as follows: 62
CAWE (40%), 48 FAWE (31%), and 45 NAWE (29%). Multivariate logistic regression analysis revealed that the CAWE pattern (OR 3.87,
95% CI 1.2–10.4, p=0.02) and aspect ratio (OR 2.1, 95% CI 0.6-4.2, p=0.03) were independently associated with aneurysm recurrence
after stent-assisted coil embolisation.
Conclusion: In conclusion, the baseline 3D HR-MRI provides novel insights into the stability of aneurysms after endovascular
embolisation, and aneurysms with CAWE pattern and higher aspect ratio may be less stable after stent-assisted coil embolisation.
Limitations: This is a retrospective cohort study; larger-scale, perspective studies are needed in the future.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Changhai Hospital of
Shanghai, China.
265
RPS 705 - Radiomics applications in MRI
CME Credits: 1
Moderator:
Magnetic resonance radiomics-derived sphericity correlates with seizure in brain arteriovenous malformations (7 min)
Jih-Yuan Lin; Taipei / Taiwan, Chinese Taipei
Author Block: J-Y. Lin1, C. Lu2, Y-S. Hu3, J. K. Loo1, K. L. Lee1, C. Liao4, C-J. Lin1; 1Taipei/TW, 2Taipei City/TW, 3New Taipei/TW, 4Texas,
TX/US
Purpose: Angioarchitectural analysis of brain arteriovenous malformations (BAVMs) is qualitative and subject to interpretation. This
study quantified the morphology of and signal changes in the nidal and perinidal areas by using MR radiomics and compared the
performance of MR radiomics and angioarchitectural analysis in detecting epileptic BAVMs.
Methods or Background: From 2010 to 2020, a total of 111 patients with supratentorial BAVMs were retrospectively included and
grouped in accordance with the initial presentation of seizure. Patients' angiograms and MR imaging results were analysed to
determine the corresponding angioarchitecture. The BAVM nidus was contoured on time-of-flight MR angiography images. The
perinidal brain parenchyma was contoured on T2-weighted images, followed by radiomic analysis. Logistic regression analysis was
performed to determine the independent risk factors for seizure. ROC curve analysis, decision curve analysis (DCA), and calibration
curve were performed to compare the performance of angioarchitecture-based and radiomics-based models in diagnosing epileptic
BAVMs.
Results or Findings: In multivariate analyses, low sphericity (OR: 2012.07, p=0.04) and angiogenesis (OR: 5.30, p=0.01) were
independently associated with a high risk of seizure after adjustment for age, sex, temporal location, and nidal volume. The AUC for
the angioarchitecture-based, MR radiomics-based, and combined models was 0.672, 0.817, and 0.794, respectively. DCA confirmed
the clinical utility of the MR radiomics-based and combined models.
Conclusion: Low nidal sphericity and angiogenesis were associated with high seizure risk in patients with BAVMs. MR radiomics-
derived tools may be used for noninvasive and objective measurement for evaluating the risk of seizure due to BAVM.
Limitations: Although the dataset acquired from a single institution and machine may improve the homogeneity of image data
quality, an external validation dataset should be considered in future studies to improve the generalisability of prediction models.
Funding for this study: Funding was received from the Taipei Veterans General Hospital (grant number: V111C- 073) and Taiwan’s
Ministry of Science and Technology (grant number: MOST- 109–2628-B-0).
Ethics committee - additional information: This study was approved by the Third Institutional Review Board of Taipei Veterans
General Hospital. The protocol was implemented after review and approval by the Human Research Protection Center of TPEVGH (IRB
No.: 2020–06-005C).
MRI radiomics and machine learning: an innovative MRI radiomics and machine learning-based method to predict
treatment response in MRI-guided HIFU ablation of bone metastasis (7 min)
Valerio D'Agostino; Parete / Italy
266
Author Block: V. D'Agostino1, M. P. Aparisi Gomez2, R. Sassi1, A. G. Morganti1, M. Buwenge1, A. Bazzocchi1; 1Bologna/IT, 2Auckland/NZ
Purpose: Pain management of bone metastases is performed with systemic and local therapies. External Beam radiotherapy is
currently the gold standard for treatment of painful metastases, however MRI-guided high-intensity focused ultrasound (MRgHIFU) has
shown great results in pain relief. To date, a reliable imaging method to predict the success of the treatment is yet to be defined. This
work aims to investigate the potential role of a radiomics-based machine-learning (ML) algorithm applied to pre- and post-treatment
MR T1w and T2w-images for the prediction of a clinical success (reduction of ≥6 point in the numerical rating pain scale) of MRgHIFU
treatment.
Methods or Background: 188 patients (112 females, 76 males) with 200 bone metastases, who underwent MRgHIFU ablation were
retrospectively selected and classified into two groups, on the basis of clinical success of the treatment. Two-dimensional
segmentations were manually drawn by an MSK-expert radiologist on axial pre- and post-treatment T1w and T2w sequences.
Radiomic feature extraction was performed using PyRadiomics. To reduce dimensionality, variance and intercorrelation analysis were
used. Subsequently, a LogitBoost classifier was trained with stratified cross-validation, tested and validated within our population.
Results or Findings: Group A (Responders) reported 112 lesions; group B (Not-responders) reported 88 lesions. 3567 radiomics
features were extracted, of which 2864 were discarded due to high intercorrelation (>0.8). The feature selection process identified
ten features to build the ML classifier, which was able to correctly classify 94% of instances on the training set and 85% on the testing
set. Weighted average precision and recall were 0.90 and 0.92 respectively, while the AUROC curve was 0.88. The performance was
similar within the validation set.
Conclusion: An ML-classifier powered by MRI radiomics might be a feasible tool to predict bone metastases pain response to
MRgHIFU.
Limitations: The study was limited by a lack of external validation, mild class imbalance, and its retrospective nature.
Reproducible radiomics features from multi-MRI-scanner test-retest-study: influence on performance and

generalisability of radiomics models (7 min)
Markus Wennmann; Heidelberg / Germany
Author Block: M. Wennmann, L. Rotkopf, F. Bauer, H. Goldschmidt, T. F. Weber, H-P. Schlemmer, S. Delorme, K. H. Maier-Hein, P.
Neher; Heidelberg/DE
Purpose: The aim of this study was to evaluate the influence of using only a subset of reproducible radiomics features, defined in a
prior in-vivo multi-MRI-scanner test-retest-study, on the generalisability and external performance of radiomics models.
Methods or Background: This retrospective study used data acquired between 2015 and 2021. The task for the radiomics models
was to predict bone marrow plasma cell infiltration from MRI in myeloma patients. Different machine learning (ML) models were
trained on data from Centre 1, using either all radiomics features, or using only reproducible radiomics features defined by a prior in-
vivo multi-MRI-scanner test-retest study. Models were tested on an internal and a multicentric external data set. Pearson correlation
coefficient r and mean absolute error (MAE) between predicted and actual plasma cell infiltration were used to quantify the model
performance. The difference between performance on the internal and external test set was calculated to measure generalisability.
Results or Findings: 302 MRIs from 300 patients from 8 centres were included. When using only reproducible features compared to
all features, for all ML models the generalisability improved. However, for the best model, a random forest regressor, the model using
all features still outperformed the model using only reproducible features on the external test set (r of 0.44 vs. 0.33 and MAE of 20.5
vs 21.9). When comparing the external performance across all combinations of ML models and feature selection methods, a random
forest regressor using all features (r=0.44, MAE=20.5) showed the best external performance.
Conclusion: A radiomics feature selection based on in-vivo reproducibility experiments between different MRI scanners improves the
generalisability of radiomics models, however, does not necessarily lead to an improvement of the external performance of the
overall best radiomics model.
Limitations: The study is limited by its retrospective nature.
Ethics committee - additional information: This study was approved by the local Institutional Research Board with code:
S-537/2020.
MRI-based radiomics analysis for prediction of treatment response to neoadjuvant chemoradiotherapy in patients with
locally advanced rectal cancer: a large multicentric study (7 min)
Yaru Feng; Shanghai / China
267
Author Block: T. Hu1, J. Gong1, Y. Sun1, M. Li1, C. Cai1, Y. Cui2, X. Zhang3, T. Tong1, Y. Feng1; 1Shanghai/CN, 2Taiyuan/CN, 3Beijing/CN
Purpose: The aim of this study was to investigate the ability of the MRI-based radiomics models for pretreatment prediction of good
response (GR) to neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer (LARC).
Methods or Background: A total of 921 patients with LARC were retrospectively recruited from 3 hospitals, including a training
dataset (TD) (n=508) and external validation datasets 1 (EVD1) (n=242) and 2 (EVD2) (n=171). Radiomics features were extracted
from the T2WI and ADC images. Three classifications, including logistic regression (LR), random forest (RF), and support vector
machine (SVM) were applied to construct radiomics models for predicting GR. The clinical-MRI model was constructed with significant
clinical characteristics and MRI morphological features by using the logistic regression analysis. The prediction performance was
evaluated using the area under the curve (AUC) and decision curve analysis (DCA).
Results or Findings: Two clinical-MRI features and ten radiomic features were selected for the GR prediction. Compared to models
from other classifiers and the clinical-MRI model, the model obtained with SVM showed promising discrimination of GR to nCRT with
AUCs of 0.798 (95% CI, 0.758-0.837), 0.790 (95% CI, 0.725-0.856) and 0.743 (95% CI, 0.666-0.821) in the training and two external
validation datasets respectively. Decision curve analysis confirmed that the radiomics models were clinically useful.
Conclusion: The MRI-based radiomics model exhibited better performance for response prediction to nCRT in LARC patients than the
clinical-MRI model, and also provided value for prognosis prediction.
Limitations: Selection bias may have been introduced by excluding patients with clinical complete response. The ROI did not include
lymph nodes. Manual segmentation of ROIs is a time-consuming procedure and requires accurate identification of MRI lesions.
Funding for this study: Funding for this study was received from the National Natural Science Foundation of China (No.82001776,
81971687, 82271946),Shanghai Natural Science Foundation (No.20ZR1412700).
Ethics committee - additional information: The study was approved by the Institutional Review Boards of all participating
centres, and the requirement for informed consent was waived due to the retrospective nature.
Prediction of the Ki-67 expression level for head and neck squamous cell carcinoma with machine learning-based
multiparametric MRI radiomics (7 min)
Weiyue Chen; Lishui / China
Author Block: W. Chen, G. Lin, J. Ji; Lishui/CN

Purpose: The aim of this study was to develop and validate a machine learning-based fusion model to preoperatively predict Ki-67
expression levels in patients with head and neck squamous cell carcinoma (HNSCC) using multiparametric magnetic resonance
imaging (MRI) images.
Methods or Background: A total of 152 patients with pathologically proven HNSCC were retrospectively enrolled and divided into
training (n=106) and validation (n=46) cohorts. Features were extracted from T2-weighted imaging fat suppression and contrast-
enhanced T1-weighted images and screened using the least absolute shrinkage and selection operator (LASSO) regression. Seven
machine learning classifiers, including k-nearest neighbors (KNN), support vector machine (SVM), logistic regression (LR), random
forest (RF), linear discriminant analysis (LDA), naive Bayes (NB), and eXtreme Gradient Boosting (XGBoost) were trained. The best
classifier was evaluated to calculate radiomics (Rad)-scores and combined clinical factors to build a fusion model, which was
visualised as a nomogram. Performance was evaluated by the area under the receiver operating characteristic curve (AUC),
calibration curve, and decision curve analysis (DCA).
Results or Findings: The SVM classifier showed the best performance, with an AUC of 0.862 in the validation cohort. The fusion
model incorporating SVM-based Rad-scores, clinical T stage and MR-reported LN status was constructed and achieved AUCs of 0.915
(0.864–0.967) and 0.896 (0.770–0.966) in the training and validation cohorts, with accuracies of 91.98% and 84.78%, respectively.
The calibration curves demonstrated a good model fit, and DCA showed the clinical benefits of the fusion model.
Conclusion: The machine learning-based fusion model based on multiparametric MRI can predict the expression of Ki-67 in HNSCC
patients, which might be helpful for prognosis evaluation and clinical decision-making.
Limitations: This is a retrospective single-centre study with unavoidable bias and a limited sample size.
Funding for this study: Funding for this study was received from the National Natural Science Foundation of China (No.82072026).
with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later
amendments or comparable ethical standards. This study was approved by the Institutional Review Board and Human Ethics
Committee of the Fifth Affiliated Hospital of Wenzhou Medical University (No. 2023-523), with the requirement for patient informed
consent being waived due to its retrospective nature. All patients’ information was anonymised prior to the analysis.
Texture analysis and rectal cancer: correlation with histology and prognosis in patients with advanced rectal cancer (7
min)
Ilaria Mariani; Lissone / Italy
268
Author Block: I. Mariani, C. Maino, T. P. Giandola, C. Talei Franzesi, D. Ippolito; Monza/IT
Purpose: The aim of this study was to retrospectively collect radiomic data from preoperative rectal MR and determine the possible
relationships between texture analysis and response to neoadjuvant treatment.
Methods or Background: 88 patients with biopsy-proven advanced rectal adenocarcinoma, staging MR and RAR after neoadjuvant
treatmente were enrolled. Based on tumour regression grade, we considered TRG 1-2 patients as responders and TRG 3-5 patients as
non-responders. Texture analysis was conducted by using LIFex software, where T2-weighted oblique axial MR sequences were
uploaded; a region-of-interest (ROI) was manually drawn on a single slice. Features with a Spearman correlation index >0.5 have
been discarded and a LASSO feature selection has been applied. Selected features were trained using bootstrapping.
Results or Findings: According to TRG classes 49 patients (55.8%) were considered responders while 39 (44.2%) as non-
responders. Two features were associated with responders’ classes: GLCM_Homogeneity and Discretised Histo Entropy log 2.
Regarding GLCM_Homogeneity, the area under the receiver operating characteristic curve (AUC), sensitivity (sens), specificity (spec),
positive predictive value (PPV), and negative predictive value (NPV) were: 0.779 (95% CIs=0.771-0.816), 86% (80-90), 67% (60-71%),
81% (76-84), and 88% (84-90), respectively. Regarding Discretised Histo Entropy log 2, diagnostic values were as follows: AUC=0.775
(0.700-0.801), sens=80% (74-83), spec=63% (58-69%), PPV=77% (70-81), and NPV=82% (80-85). By combing both radiomics
features the radiomics signature diagnostic accuracy increased (AUC=0.844, p<0.05). Finally, the AUC of 1000 bootstraps was 0.810.
Conclusion: Texture analysis can be considered an advanced complementary diagnostic tool to determine a possible correlation
between pre-surgical MR data and response to neoadjuvant therapy.
Limitations: Considering its low robustness, further studies with a larger cohort of patients should aim to validate these preliminary
data.
Prediction of overall survival in paediatric neuroblastoma patients through machine learning in the large multi-
institutional PRIMAGE cohort (7 min)
Jose Lozano; Valencia / Spain
Author Block: J. Lozano1, A. Jimenez-Pastor1, G. Weiss2, D. Veiga Canuto1, B. Martínez De Las Heras1, A. Cañete Nieto1, B. Hero3, R.
Ladenstein4, L. Marti-Bonmati1; 1Valencia/ES, 2Boston, MA/US, 3Cologne/DE, 4Vienna/AT
Purpose: Neuroblastoma (NB) is the most frequent and highly aggressive solid cancer in childhood, in which imaging plays a pivotal
role at every step of the patient's journey. This study sought to develop a machine learning model using clinical, molecular, and
magnetic resonance (MR) radiomics features to predict patient’s overall survival (OS) and aid in their stratification.
Methods or Background: A database of 513 patients was used for model training, validation, and testing. Furthermore, 22
additional patients from hospitals not originally in the database were utilised as an external test. Manual tumour segmentations of the
NB were conducted on the corresponding T2-weighted MR images to segment the primary tumour by an experienced radiologist. In
total, 107 radiomics features were extracted and subsequently harmonised across manufacturers and magnetic field strengths using
the nested ComBat methodology. Finally, radiomic features were combined with the clinical and molecular data to serve as input for
the models. A nested cross-validation approach was used as training methodology to select the best preprocessing and model
configuration.
Results or Findings: A C-index of 0.788±0.049 was achieved in the test, being a random survival forest the model showing the best
performance. For the additional 22 patients, a C-index of 0.934 was obtained. The model exhibited superior predictive performance
and patient stratification compared to the standard risk group INRG. Interpretability analysis revealed the significance of clinical
variables, with radiomics features related to lesion heterogeneity and size playing an important role in prediction.
Conclusion: The OS predictive model demonstrated high performance and alignment with established clinical variables, highlighting
the importance of radiomics features. It presents new evidence for enhancing patient care and clinical decision-making.
Limitations: Greater sample sizes are required in the external test to confirm the results.
Funding for this study: Funding was received from PRIMAGE (PRedictive In-silico Multiscale Analytics to support cancer
personalized diagnosis and prognosis, empowered by imaging biomarkers), a Horizon 2020|RIA project (Topic SC1-DTH-07-2018),
grant agreement no: 826494.
Ethics committee - additional information: This study was approved by an Institutional Review Board and written informed
consent was obtained from all participant centres.
mp-MRI radiomic model predicts peri-tumour tertiary lymphoid structures in hepatocellular carcinoma: a multi-centre
study (7 min)
Shichao Long; Changsha / China
269
Author Block: S. Long, J. Chen, L. Zhong, W. Liu, Y. Pei, W. Li; Changsha/CN
Purpose: The massive presence of peri-tumour tertiary lymphoid structure (mpTLS) questions whether HCC patients can benefit from
immunotherapy. However, it has been identified only by pathological examination. This study aims to develop a noninvasive tool
using preoperative multiple parameter MR imaging (mpMRI) radiomic for predicting mpTLS.
Methods or Background: 584 consecutive HCC patients (mpTLS+:154; mpTLS-:430) were retrospectively recruited from four
independent institutes and were divided into training (n=307) in one institute, and a validation cohort (n=277) in the other three
institute. 76 HCC participants (mpTLS+:21; mpTLS-:55) were also enrolled as prospective cohort, including the relapse-free survival
(RFS) and overall survival (OS). All subjects underwent preoperative mpMRI. Three different models (Model 1: peri-tumour model;
Model 2: intra-tumour model; Model 3: combined models 1 and 2) were constructed to stratify mpTLS+. The optimal model was
decided by the maximum area under of curve (AUC) in the training set and validated in both the validation and the prospective
cohort, which was further used to predict the RFS and OS in the prospective data.
Results or Findings: For retrospective data, Model 3 (AUC:0.92) was the optimal model for diagnosing mpTLS+ than model 2
(AUC:0.87) and model 1 (AUC:0.85) in training cohorts (all P<0.001), which was validated in validation cohort (AUC: 0.91 vs 0.85 vs
0.84; all P<0.001), and prospective corhort (AUC:0.91 vs 0.84 vs 0.77; all P<0.001). For prospective data, Model 3 could predict the
RFS (P < 0.001) and OS (P<0.001) based on mpTLS+.
Conclusion: Model 3 (the combined model) is a reliable and noninvasive tool for predicting mpTLS and further can forecast OS and
RFS, which is helpful in deciding immunotherapy for HCC patients.
Funding for this study: Funding was received from the National Natural Science Foundation of China [82071895 and 82271984] to
W.Z.L.; Hunan Provincial Science and Technology Department [2023JJ30903 and 2022JJ30950] to W.Z.L. and Y.G.P. National Geriatric
Disease Clinical Medical Research Center Foundation [2022LNJJ08] to Y.G.P. Youth Project of Natural Science Foundation of Hunan
Province to W.G.L.
Ethics committee - additional information: This study was approved with the Ethics Committee number: 2018111101.
270
RPS 710 - Sports imaging
Categories: Emergency Imaging, Imaging Methods, Musculoskeletal

CME Credits: 1
Moderator:
Philip Robinson; Leeds / United Kingdom
Assessing cartilage regeneration in a minipig model: cutting-edge MR techniques challenge the gold standard (7 min)
Miriam Frenken; Düsseldorf / Germany
Author Block: K. L. Radke, B. Valentin, A. Müller-Lutz, M. Frenken; Düsseldorf/DE

Purpose: The purpose of this study was to examine the potential of differently composed MR sequences, including T1, T2, T2* and
T1ρ, in assessing cartilage regeneration. These MR techniques were compared to dGEMRIC, providing valuable insight into non-
invasive methods for monitoring and assessing cartilage repair after surgery.
Methods or Background: In orthopaedics, assessing articular cartilage regeneration safely and effectively remains a challenge,
despite advancements in compositional MRI. We aimed to investigate the efficacy of various compositional MR sequences in
quantitatively assessing cartilage and compare them to the gold standard, delayed gadolinium-enhanced MR imaging of cartilage
(dGEMRIC).
In a preclinical minipig model, we induced standardised osteochondral defects in the proximal femur of 14 animals, divided into four
groups: porcine collagen scaffolds with autologous adipose stromal cells (ASC), autologous bone marrow stromal cells (BMSC),
unpopulated scaffolds (US), and an untreated defect group. After six months, we employed different MR composition methods,
including dGEMRIC, T1, T2, T2*, and T1ρ, using a clinical 3T MR scanner.
Results or Findings: Results showed significant differences in the untreated defect group, with lower dGEMRIC values (404.86±64.2
ms, P=0.018) and higher T2 times (44.24±2.75 ms, P<0.001) in the defect region. Conversely, there were no significant differences in
dGEMRIC values among the three treatment groups (ASC, BMSC, US), indicating successful cartilage reconstruction.
Conclusion: Notably, dGEMRIC proved effective for monitoring cartilage regeneration. Interestingly, T2 imaging emerged as a
reliable alternative, offering non-contrast cartilage imaging for future in vivo studies exploring different treatment modalities'
cartilage regeneration potential.
Limitations: These results have not yet been translated from animal to human.
Funding for this study: Funding was received from the University of Düsseldorf, Deutsche Arthrose-Hilfe.
Ethics committee - additional information: The study was approved by the Ethics Committee of the University of Düsseldorf.
Limited magnetic resonance imaging in acute upper ankle injuries (7 min)

Aynur Gökduman; Frankfurt a. Main / Germany
271
Author Block: A. Gökduman; Frankfurt a. Main/DE
Purpose: The purpose of this study was to investigate the diagnostic accuracy of limited magnetic resonance imaging for acute ankle
injuries in patients with negative plain radiographs.
Methods or Background: Patients that had undergone a routinely performed 3-T MRI scan after acute trauma of the upper ankle
with unremarkable plain radiographs (within a two-week interval) were eligible for study inclusion in this retrospective study. For
reference, standard definition five radiologists blinded to clinical information evaluated the full MRI study (PD-weighted fat-saturated
axial, coronal, and sagittal sequences, and T1-weighted coronal/sagittal sequences) in consensus reading sessions. After six weeks,
the studies were independently reevaluated by utilising only PD-weighted fat-saturated coronal sequences. Sensitivity, specificity, and
accuracy for injuries of the most relevant bone structures (tibia, fibula, and talus), ligaments (syndesmotic, medial, and lateral
ligaments), and tendons (tibialis and peroneal tendons) assessed by application of a binary classification (0=no injury, 1=injury) were
the primary metrics of diagnostic performance. Diagnostic confidence was rated by the usage of a 5-point Likert scale (1=low;
5=high).
Results or Findings: 137 patients were enrolled in this study. A total of 108 bone injuries, 126 ligament injuries, and 5 tendon
injuries were defined by the consensus. Limited MRI showed overall high diagnostic accuracy for assessing bone (sensitivity: 97%,
specificity: 93%, accuracy: 95%), ligament (sensitivity: 92%, specificity: 94%, accuracy: 93%), and tendon (sensitivity: 80%,
specificity: 100%, accuracy: 90%) injuries in comparison to full MRI (p<0.01). Diagnostic confidence was comparable (scores: 4.37/5
vs 4.41/5; p=0.12) and interreader agreement for limited MRI was high (K=0.83).
Conclusion: Limited MRI provides high diagnostic accuracy and confidence for assessing acute trauma injuries of the upper ankle.
Ethics committee - additional information: This study was approved by vote through the relevant Ethics Board.
Surgical correlation of peroneal compartment pathology in 2D and 3D ankle MRIs with interreader reliability (7 min)
Muhammad Rubeel Akram; Dallas / United States
Author Block: M. R. Akram, N. Rajamohan, A. Chhabra, G. Liu, M. VanPelt, Y. Xi, F. Duarte Silva; Dallas, TX/US
Purpose: Peroneal compartment pathology (PCP) is a common yet underdiagnosed source of lateral ankle pain, often requiring
clinical examination and imaging for diagnosis. Current literature lacks research on 3D MRI accuracy for PCP. We investigated
interreader reliability (IRR) for PCP using parallel 2D and 3D ankle MRIs, hypothesising higher IRR and accuracy with this combined
analysis than with 2D alone.
Methods or Background: We retrospectively screened 634 patients with 3-Tesla ankle MRIs, including multiplanar proton density
DIXON and isotropic 3D fast-spin echo acquisitions (3DFSE), and excluded patients with unretrievable operative reports or prior
peroneal compartment surgery. Our final sample comprised 27 scans from 26 patients who underwent peroneal surgery at least six
months post-MRI. Two radiologists (14- and 1-year post-musculoskeletal fellowship) analysed scans for peroneus brevis (PB) and
peroneus longus (PL) tenosynovitis, tendinopathy, and tears. We evaluated IRR using intraclass correlation coefficient (ICC) and
assessed sensitivity (Se) and specificity (Sp) against surgical results.
Results or Findings: IRR was good for PB/PL tendons (ICC=0.735-0.724, respectively) and common peroneal tenosynovitis
(ICC=0.630), and was fair for PL/PB tenosynovitis (ICC=0.573-0.541, respectively). Surgical reports identified PB partial/split tears in
51.9% (MRI Se=63.6-83.3% and Sp=66.7-83.3%) and complete tears in 7.4% (MRI Se=25-100% and Sp=88.5-100%) of cases.
Regarding the PL tendon, surgical reports identified 7.4% partial/split tears (MRI Se=0-28.6% and Sp=80-92%) and 3.7% complete
tears (MRI Se=33.3-100% and Sp=93.2-100%). The readers identified an average of 31.5% PL tenosynovitis, 27.8% PB tenosynovitis,
and 46.3% common peroneal tenosynovitis cases.
Conclusion: MR identification of PCP had moderate IRR with higher specificity than sensitivity. Additional 3D sequencing did not
improve accuracy compared to prior literature.
Limitations: Surgical reports prioritised tears over tendinopathy details, which led to the merger of normal and tendinopathy
subgroups and prevented us from analysing accuracy for tendinopathy.
Ethics committee - additional information: The study was retrospective.
Shear-wave elastography for the evaluation of tendinopathies: a systematic review and meta-analysis (7 min)
Mariachiara Basile; Catania / Italy
272
Author Block: M. Basile1, S. Gitto1, C. Messina1, Ž. Snoj2, S. GIanola1, S. Bargeri1, G. Castellini1, L. M. Sconfienza1, D. Albano1; 1Milan/IT,
2
Ljubljana/SI
Purpose: The purpose of this study was to compare pathologic and healthy tendons using shear-wave elastography (SWE).
Methods or Background: A systematic review with meta-analysis was conducted by searching Pubmed and EMBASE up to
September 2022. Prospective, retrospective, and cross-sectional studies that used SWE in the assessment of pathologic tendons
versus control were included. Our primary outcomes were SWE velocity (m/s) and stiffness (kPa). Methodological quality was assessed
by the methodological index for non-randomized studies (MINORS). We used the mean difference (MD) with corresponding 95%
confidence intervals (CIs) to quantify effects between groups. We performed sensitivity analysis in case of high heterogeneity, after
excluding poor quality studies according to MINORS assessment. We used Grades of Recommendation, Assessment, Development
and Evaluation to evaluate the certainty of evidence (CoE).
Results or Findings: Overall, 16 studies with 676 pathologic tendons (188 Achilles, 142 patellar, 96 supraspinatus, 250 mixed) and
723 control tendons (484 healthy; 239 contralateral tendon) were included. Five studies (31.3%) were judged as having poor
methodological quality. Shear-wave velocity and stiffness meta-analyses showed high heterogeneity. According to a sensitivity
analysis, pathologic tendons had a lower shear wave velocity (MD −1.69m/s; 95% CI -1.85; -1.52; n=274; I2 50%) compared to
healthy tendons with very low CoE. Sensitivity analysis on stiffness still showed high heterogeneity.
Conclusion: Pathological tendons may have reduced SWE velocity compared to controls. Future robust high-quality longitudinal
studies and clear technical indications on the use of this tool are needed.
Limitations: We included different a study design in order to obtain more literature, however this choice has incorporated some
methodological heterogeneity.
Ethics committee - additional information: Institutional Review Board approval was not required because no patient data was
used for this study.
Muscle mapping evaluation in athletes with myocardial inflammation (7 min)

Mattia Tordin; Padova / Italy
Author Block: A. Lupi, D. Creazzo, M. Tordin, S. Zinato, E. Quaia, A. Pepe; Padua/IT

Purpose: In the context of magnetic resonance imaging (MRI), modern-day mapping techniques allow for precise and quantitative
tissue characterisation and are increasingly widespread. In the musculoskeletal field, MRI finds extensive application thanks to its high
soft tissue contrast. In specific clinical scenarios, T1/T2 mapping can be used as a valuable tool to provide skeletal muscle tissue
characterisation. To the best of our knowledge, correlation between myocardial and skeletal muscle involvement has never been
explored. The aim of our study is to evaluate muscle mapping in athletes with myocardial inflammation.
Methods or Background: We included competitive athletes consecutively referred to our center for a Cardiovascular Magnetic
Resonance (CMR) (1.5T Siemens), that showed myocardial inflammation. T1 and T2 mapping values were extracted from single
Regions of Interest (ROIs) manually traced always in the same back muscles included in the Field of View (latissimus dorsi or serratus
anterior), both on MOLLI and T2p-SSFP images, and compared with reference values obtained in our Institute from fifty healthy
volunteers stratified by age and sex. Image analysis was performed using cvi42.
Results or Findings: Fifteen patients; twelve males, mean age 50.1±14.7 years, with myocardial inflammation assessed with
parametric and/ or nonparametric techniques, were included. Normal T1 and T2 mapping muscle values were found in four (26.6%)
patients (mean T1 852.3±10.1ms and T2 38.8±1.5ms). In one female patient (6.7%) with peri-myocarditis by CMR, T2 mapping
values turned out to be pathological (44.2ms, n.v. 30-42ms) and in ten male patients (66.6%) T1 mapping values turned out to be
pathological (n.v. 762-853ms).
Conclusion: These preliminary results show that muscle oedema could be found in subjects with myocardial inflammation. Further
studies on larger population and prognostic data are needed to better understand increased T1 mapping values significance.
Ethics committee - additional information: This study was approved by the Azienda Ospedale Università Padova.
Rethinking the anatomical variants role in peroneal tendinopathy (7 min)

Pamela Soledad Perez; Lomas de Zamora / Argentina
273
Author Block: O. D. Rodriguez Martinez, P. S. Perez, M. E. Scherer, J. E. Prieto Terán, N. E. Alanis, I. Bidart, S. Centofante, E. Piedra,
O. Montaña; Buenos Aires/AR
Purpose: The study purpose was to assess the anatomical variants prevalence of peroneal tendon tract in patients with external
retromalleolar pain and define their association with the development of tendinopathies.
Methods or Background: A retrospective review of 1929 ankle MRI scans (Philips 1.5T) was performed between January and
December 2022. We found 117 patients with peroneal tendon pathology in the context of external ankle pain. In these, the
prevalence of convex/ flat retromalleolar sulcus (CFRS), hypertrophic peroneal tubercle of the calcaneus greater than 5mm (HPTC),
peroneus quartus muscle (PQM), and os peroneum presence were evaluated as anatomical variants that could be predisposed to the
development of pathology tendinous. Cases with fractures in the area of interest, people under 15 years old and over 80 years old,
and studies with suboptimal imagen quality and/ or with acquisition artifacts such as magic angle were excluded from the study.
Results or Findings: In the 117 patients studied, it was found that 69% (n=81) had at least one anatomical variant in the peroneus
tendons tract up to their distal insertions, with the loss of the concavity of the retromalleolar sulcus being the most frequent (65
cases). HPTC was the second most common variant, with eleven cases, and finally four cases of PQM were found. Only one case of os
peroneum was found in analysed patients, probably due to the method used (MRI).
Conclusion: We found a high prevalence of SRCP and TPCH in association with peroneal tendinopathies. This demonstrates the
importance of a systematic evaluation of these anatomical variants as possible causes of external retromalleolar pain, and its search
and mention in routine ankle study is recommended.
Funding for this study: This study was funded by DIM.
Ethics committee - additional information: The study is retrospective, so approval is not needed.
Patients at risk of bone non-union in distal radius fractures: value of dual-energy CT-derived metrics (7 min)
Leon David Grünewald; Frankfurt a. Main / Germany
Author Block: L. D. Grünewald, V. Koch, C. Booz, S. Martin, S. Mahmoudi, T. Vogl; Frankfurt a. Main/DE
Purpose: Distal radius fractures (DRF) are among the most diagnosed fractures in emergency departments, and bone non-union can
lead to persistent pain and functional limitations. Despite the clinical implications, there is currently no established approach to
identify patients at risk of bone non-union. The purpose of this study was to evaluate various metrics derived from routine CT scans of
the distal radius to identify patients at risk of bone non-union.
Methods or Background: The distal radius of patients who underwent dual-energy CT (DECT) between 01/2016 and 08/2021 was
retrospectively analysed. Cortical HU, trabecular HU, cortical thickness, and DECT-based bone mineral density (BMD) were obtained
from all examinations. Patient files and follow-up images of patients were examined for the occurrence of bone non-union. Receiver-
operating characteristic (ROC) analysis identified AUC values for BMD, HU values, and cortical thickness, and logistic regression
models were used to evaluate their associations with the occurrence of bone non-union.
Results or Findings: 263 patients (median age 52 years; interquartile range 36–64) were included in this study. ROC curve analysis
demonstrated a significantly higher AUC value for DECT-derived BMD compared to cortical HU, trabecular HU and cortical thickness
(0.83 vs. 0.63, 0.60 and 0.56, respectively; p<0.01). Logistic regression models confirmed a significant association of lower DECT-
derived BMD with the occurrence of bone non-union (Odds Ratio, 0.93; p<0.001), but not of cortical HU, trabecular HU or cortical
thickness (p>0.05 for all values, respectively).
Conclusion: CT examinations of the distal radius obtained in clinical routine may serve as a useful tool in identifying patients at risk
of developing bone non-union. Dual-energy CT-derived bone mineral density is a better predictor of bone non-union compared to
cortical HU, trabecular HU, and cortical thickness.
Limitations: The study is limited by the possibility of preselection bias, as well as the limiting of research to dual-energy CT scans.
Ethics committee - additional information: Written consent was waived due to the retrospective nature of the study.
Diagnostic accuracy of a fast deep learning-based MRI protocol in diagnosing meniscal and ligament tears of the knee:
comparison with standard MRI protocol (7 min)
Fabio Lombardo; Negrar / Italy
274
Author Block: F. Lombardo, C. Longo, E. Oliboni, S. Crosara, G. Sala, T. Mignolli, G. Foti; Negrar/IT
Purpose: The purpose of this study was to evaluate the diagnostic accuracy of a deep learning- (DL) based protocol in identifying
meniscal and ligament tears of the knee in comparison to standard magnetic resonance imaging (MRI) protocol.
Methods or Background: We included 88 consecutive patients (46 males and 42 females, mean age of 55.2) between April and
May 2023. All patients underwent standard MRI protocol, including multiplanar 3 mm T1W, STIR and DP fat-saturated sequences
(acquisition time 18 minutes) and a fast protocol acquired with DL sequences (acquisition time 5 minutes) the same day. Four
radiologists (21, 16, 13 and 5 years of experience, respectively), blinded to standard MRI protocol, evaluated the presence of meniscal
and ligament tears on DL protocol images. Standard MRI images served as standard of reference (consensus reading of two additional
experienced MSK radiologists). Diagnostic accuracy values of DL protocol (qualitative assessment) and inter-observer agreement
were calculated.
Results or Findings: Standard MRI revealed the presence of meniscal tear in 45/88 patients (51.1%) and ligament tears in 37/88
cases (42.0%). Sensitivity, specificity and overall accuracy of DL protocol for meniscal and ligaments tear were 97.7% (44/45), 100%
(43/43) and 98.9% (87/88), and 100% (37/37), 98.0% (50/51) and 98.9% (87/88), respectively. The diagnostic accuracy values of DL
protocol were similar to those of standard protocol (p=0.43). Conversely, the acquisition time for DL protocol was significantly lower if
compared to that of standard protocol (p=0.001).
The inter-observer agreement was near perfect (k=0.96).
Conclusion: Despite a significant reduction of acquisition time, the DL MRI protocol of the knee showed similar diagnostic accuracy
value compared to standard MRI protocol.
Limitations: The study is limited by its small sample size.
Ethics committee - additional information: This study was approved by the relevant Institutional Research Board.
275
MD 1 - Musculoskeletal disease: new developments in imaging and treatment - recommendations for

clinical practice and directions for the future
Categories: General Radiology, Imaging Methods, Multidisciplinary, Musculoskeletal, Research

CME Credits: 1
Moderators:
Mario Maas; Amsterdam / Netherlands
Welcome and introduction to connAction (2 min)


Mario Maas; Amsterdam / Netherlands
1. To highlight the innovative DECT tool of bone marrow oedema detection.

2. To critically review the protocol provided by vendors - plug and play.
3. To provide insight into acceptance in clinical practice.
Bone marrow oedema detection with dual-energy CT: a role for radiologist or computer only? (12 min)
Mikael Boesen; Copenhagen / Denmark
Oedema and CT scanning: a clinical perspective (12 min)

Frank Bloemers; Amsterdam / Netherlands
Expert panel discussion (32 min)
276
CUBE 7 - Heat or cold: drug-eluting beads or radioembolization – what is what

Oncologic IR Day - Topic Coordinator: Prof. Dr. Ralf-Thorsten Hoffmann
in the audience.
Moderator:
Ralf-Thorsten Hoffmann; Dresden / Germany

Heat or cold: drug-eluting beads or radioembolization – what is what (28 min)

Tobias Jakobs; München / Germany
1. To learn about differences in heat and cold-based ablation systems.

2. To learn about technical differences in drug-eluting particles.
3. To learn about technical and physical differences in radioactive embolising particles.
277
ESR eHealth 8 - Is there a future for structured reporting?
Categories: Imaging Informatics, Management/Leadership, Professional Issues

ETC Level: LEVEL II
CME Credits: 1
This session should provide an overview on past developments, current trends and potential future developments in structured
radiology reporting. Following up on the currently published Update on Structured Reporting by the ESR, the speakers will discuss
where and how structured reporting is already being used and what constitutes a state-of-the-art radiological report that provides the
most relevant items to the referring physicians in a usable and actionable format. Lastly, current developments in Natural Language
Processing and Large Language Models are discussed as how those technologies will impact how we report and extract structured
data from unstructured reports.
Moderator:
Merel Huisman; Nijmegen / Netherlands

Past and present of structured reporting (15 min)

Tobias Jorg; Mainz / Germany
Steps to success for a state-of-the-art radiology report (15 min)

Structuring free-text reports with NLP and large language models (15 min)
Keno K. Bressem; Berlin / Germany
Panel discussion: When will we be able to produce reusable data in clinical routine? (10 min)
278
EU 8 - What is appropriate image quality and how can or should it be used for optimisation?
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Physics in Medical Imaging

ETC Level: LEVEL II
CME Credits: 1.5
Moderator:

Determine image quality based on reader studies scientifically (15 min)

Anders Tingberg; Malmö / Sweden
1. To learn about the standard methodology and tools for reader studies.
2. To appreciate the science behind it.
3. To understand the conditions of such reader studies and potential drawbacks.
Determine image quality based on physical parameters (15 min)

Ehsan Samei; Durham, NC / United States
1. To learn about physics-based metrics for image quality assessment.

2. To appreciate the problems with phantom-based measurements.
3. To understand the potential of image quality assessment in patient images.
Determine image quality using AI-based methods (15 min)

1. To learn about image quality assessment using AI-based methods.

2. To appreciate the problems associated with such approaches.
3. To understand the requirements for such assessment.
The concept of appropriate image quality as addressed by the ESI/AAPM paper (10 min)
1. To learn to differentiate between best image quality and appropriate image quality.
2. To appreciate the possibilities of optimisation using appropriate image quality and corresponding metrics.
3. To understand the concept of appropriate image quality.
The concept of appropriate image quality assessment from the radiologists' point of view (10 min)
Franz Kainberger; Vienna / Austria
1. To learn how to evaluate appropriate image quality efficiently.

2. To appreciate the use of such approaches in daily clinical routine.
3. To understand the benefits of the approach for safe imaging.
279
What to learn from appropriate image quality concept? (10 min)
Ehsan Samei; Durham, NC / United States
1. To learn about potential applications of the concept.

2. To appreciate the remaining questions.
3. To understand the necessary next steps regarding research and application.
Panel discussion: How to move forward and how to implement the concept in daily clinical practice for optimised safe
imaging (10 min)
280
OF 8R - Making strides towards person-centred imaging
Categories: Education, Evidence-Based Imaging, Imaging Methods, Management/Leadership, Professional Issues, Radiographers
CME Credits: 1
This session embodies the essence of modern healthcare and highlights a commitment to focussing on the unique needs, comfort,
and dignity of individuals undergoing imaging procedures. This session will include three compelling talks that collectively illuminate
the path towards achieving person-centred care in the realm of medical imaging. Indeed, apart from addressing the pressing need for
greater inclusivity in imaging, the speakers will share innovative approaches and practical tools that can help empower radiographers
and healthcare professionals to embrace diversity, foster inclusivity, and prioritise the unique needs of each individual, ultimately
advancing the cause of compassionate and effective medical imaging.
Moderator:

Towards autism-friendly MRI (16 min)

Nikolaos Stogiannos; Corfu / Greece
A conceptual toolkit for radiographers to care for persons experiencing psychosis (16 min)
Krzysztof M. Skuza; Lausanne / Switzerland
Delivering person-centred care to minimise discomfort and maximise safety in radiography (16 min)
Giuseppe Roberto Bonfitto; Milan / Italy
281
ESR/EFRS - Teleoperations in radiology
Categories: Education, Evidence-Based Imaging, Professional Issues, Radiographers, Research

CME Credits: 1.5
Opportunities for teleoperations are rapidly expanding within radiology. The aim of this session is to highlight the concept of
teleoperations and to explore a range of real-world examples within clinical practice and education.
Moderators:
Tiina Pauliina Nousiainen; Jyväskylä / Finland
Borut Marincek; Kilchberg / Switzerland
Chairpersons' introduction: what is the concept behind the teleoperations? (5 min)

Tiina Pauliina Nousiainen; Jyväskylä / Finland
Borut Marincek; Kilchberg / Switzerland
Initial experiences of remote MRI scanning (15 min)

Anton Sheahan Quinsten; Bochum / Germany
Teleoperations in ultrasound: where are we now? (15 min)

Teleoperations and methods to support the military (15 min)

David Muchena; Colchester / United Kingdom
Remote imaging as a means to support radiographer and radiologist education (15 min)
Ricardo Silva Teresa Ribeiro; Lausanne / Switzerland
Experiences of implementing a televisits system for cancer patients during the COVID-19 pandemic (15 min)
Patrizia Cornacchione; Rome / Italy
Panel discussion: How can we use teleoperation to maximise patient outcomes? (10 min)
282
TC 827 - The basics of neuroimaging in neurodegeneration
Categories: Hybrid Imaging, Molecular Imaging, Multidisciplinary, Neuro, Nuclear Medicine

CME Credits: 1
Structural and molecular imaging techniques can demonstrate the onset of neurodegenerative and dementing disorders years before
the onset of symptoms. This session will discuss the structural MRI changes that occur with normal aging, how to differentiate normal
from abnormal and which MR imaging sequences to use for diagnostic assessment in the memory clinic. The basics of PET imaging in
neurodegeneration will be discussed, including how/which tracers can dissect the molecular pathology associated with
neurodegenerative diseases. The role of imaging biomarkers in clinical routine as well as in trials for neurodegenerative diseases is
being demonstrated. At the end of the course, the participants will better understand available imaging techniques and biomarkers in
diagnosis of neurodegenerative diseases.
Moderators:
Meike W. Vernooij; Rotterdam / Netherlands

The basics of MR imaging in neurodegeneration (19 min)

Yoshimi Anzai; Salt Lake City / United States
The basics of molecular imaging in neurodegeneration (19 min)

Javier Arbizu; Pamplona / Spain
Role of imaging biomarkers for dementia in clinical routine and trials (19 min)
283
US 8 - Advances in ultrasound imaging of transplantation
Categories: Education, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods, Paediatric
CME Credits: 1.5
Moderators:
Thomas Fischer; Berlin / Germany

Renal transplantation (20 min)

1. To learn the monitoring strategy of the early phase after kidney transplantation using MPUS.
2. To work out the benefits of elastography, new broadband Doppler techniques and CEUS in the early phase after transplantation.
3. To understand characteristic features of late-phase complications, such as tumours of the shrunken kidneys and vascular problems.
Liver transplantation (20 min)

Gibran Timothy Yusuf; London / United Kingdom
1. To understand the role of ultrasound and Doppler ultrasound assessment of liver transplants.
2. To appreciate the role of advanced ultrasound techniques in liver transplant assessment, particularly CEUS.
3. To understand when complications need further radiological assessment for appropriate management.
Combined kidney and pancreas transplantation (20 min)

Jose Ángel Jiménez-Lasanta; Barcelona / Spain
1. To understand the anatomy of combined renal/pancreas transplantation.

2. To understand the role of ultrasound, including Doppler, in assessing renal/pancreas transplants.
3. To understand when other imaging modalities are required for assessment.
Paediatric liver and multiorgan transplantation (20 min)

Annamaria Deganello; London / United Kingdom
1. To understand the main surgical techniques and post-surgical complications of paediatric liver and multi-visceral transplantation.
2. To review the role of ultrasound in the post-operative monitoring of paediatric transplant recipients.
3. To learn how the versatility of ultrasound, including advanced techniques, can assist in the management of post-operative
complications.
284
SA 8 - Imaging the anxious child
Categories: Imaging Methods, Multidisciplinary, Paediatric

ETC Level: LEVEL I
CME Credits: 1.5
Moderator:

Effective strategies for easing anxiety in paediatric radiology: tried and proven approaches (18 min)
Nanko De Graaf; Rotterdam / Netherlands
1. To learn strategies on how to put the child at ease for the imaging study.
2. To understand the various roles of caregivers in imaging studies.
3. To learn how to cultivate a positive memory of the imaging study.
Increase your success rate for MRI in children without sedation (18 min)
Bac Nguyen; Oslo / Norway
1. To prepare the patient before entering the MRI suite.

2. To explain the mode of operation during the examination.
3. To explain how to handle the child after the scan (post-festum).
Using digital technology in interventional radiology: how to aid child co-operation in imaging (18 min)
Elisa Aguirre Pascual; Madrid / Spain
1. To learn strategies on how to put the child at ease for interventional procedures.
2. To familiarise with different types of digital technologies used in interventional radiology procedures.
3. To use of VR glasses to downgrade or avoid sedation in interventional procedures.
When and how to use sedation? (18 min)

lgnacio Malagon; Nijmegen / Netherlands
1. To name and recognise absolute contraindications for sedation in remote locations.

2. To familiarise with most of the drugs available for sedation and their antidotes.
3. To be able to discuss and answer the question: which professionals should be involved in the care of these patients?
Panel discussion: Can we really leave children and parents with a positive memory of their experience in the radiology
department? (13 min)
285
E³ 821d - Family cancer syndromes
Categories: Abdominal Viscera, Breast, Imaging Methods, Oncologic Imaging

CME Credits: 1.5
HBOC syndrome and beyond (45 min)

1. To illustrate the spectrum of genetic mutations responsible for HBOC.

2. To become familiar with the appropriate imaging studies in HBOC.
3. To learn how to integrate different imaging techniques.
Multiple endocrine neoplasia (MEN) (45 min)

1. To become familiar with the spectrum of manifestations of MEN syndromes.

2. To illustrate the imaging features of MEN.
3. To correlate the imaging findings with the pathologic diagnosis.
286
E³ 818 - Pre- and post-procedural cardiac imaging
Categories: Cardiac, Evidence-Based Imaging, Imaging Methods

CME Credits: 1.5
Moderator:

Known coronary disease: to revascularise or not? (28 min)

Maja Pirnat; Maribor / Slovenia
1. To determine which imaging modality can be applied to answer the clinical question.
2. To assess the important imaging features in assessing the severity of coronary disease.
3. To judge if the coronary disease is likely to benefit from revascularisation or not.
Imaging before cardiac valve replacement (28 min)

Jonathan R. Weir-Mccall; London / United Kingdom
1. To describe the indications for cardiac valve replacement.

2. To evaluate the important imaging features to assess prior to valve replacement.
3. To describe the measurements to perform as part of CT/MR imaging.
After valve replacement: do-not-miss findings (28 min)

Ricardo P. J. Budde; Rotterdam / Netherlands
1. To describe the indications for radiological imaging after cardiac valve replacement.
2. To evaluate the important imaging features after valve replacement.
3. To distinguish potential complications after valve replacement on radiological imaging.
287
E³ 819 - X-ray: still alive?
Categories: Artificial Intelligence & Machine Learning, Imaging Methods

CME Credits: 1.5
Moderator:
Ernst J. Rummeny; Munich / Germany

Ernst J. Rummeny; Munich / Germany
Dark-field radiography: an update (25 min)

Franz Pfeiffer; Garching / Germany
1. To explain the technical basis of dark-field radiography.

2. To describe results in pre-clinical and clinical studies.
3. To analyse advantages in clinical practice.
Breast tomosynthesis: current technology and applications (25 min)

Francesca Galati; Rome / Italy
1. To explain the technical basis and recent implementations of breast tomosynthesis.

2. To describe the advantages of breast tomosynthesis in both clinical and screening settings.
3. To analyse the advantages when using contrast-enhanced mammography.
AI and x-ray: the perfect wedding? (25 min)

Erik R. Ranschaert; Turnhout / Belgium
1. To summarise the current applications of AI in x-ray examinations.

2. To describe major benefits for radiologists and the diagnostic workflow.
3. To examine the most important barriers to successful implementation.
Discussion (10 min)
288
E³ 821c - Choice and use of contrast media in different clinical settings
Categories: Contrast Media, Emergency Imaging, General Radiology, Genitourinary, Medico-legal

ETC Level: LEVEL II
CME Credits: 1.5
Use of sonographic contrast media (30 min)

Orlando Catalano; Napoli / Italy
1. To illustrate indications, advantages and limitations of sonographic contrast media.

2. To explain the injection technique and image acquisition using easy and difficult cases.
Choice and injection parameters using iodinated contrast media in CT (30 min)
Wolfgang Schima; Vienna / Austria
1. To discuss the choice and injection parameters in the acute setting, in first examinations, in daily practice and in follow-up
examinations.
2. To learn more about dose reduction in fragile and oncologic patients.
Choice and injection parameters of gadolinium-based contrast media (30 min)

Luigi Grazioli; Mairano / Italy
1. To understand the choice and injection parameters in different clinical settings.

2. To show easy and difficult cases facing liver MRI.
289
E³ 821b - Imaging evaluation of bone and soft tissue tumours: the most important findings
Categories: Abdominal Viscera, Imaging Informatics, Imaging Methods, Musculoskeletal, Oncologic Imaging
CME Credits: 1.5
Bone tumours (45 min)

Radhesh Krishna Lalam; Pant / United Kingdom
1. To understand the most important findings in the diagnosis of bone tumours with different imaging techniques.
2. To learn the most common differential diagnosis (especially facing "don't touch lesions").
Soft tissue tumours (45 min)

Ana Navas Cañete; Leiden / Netherlands
1. To understand the most important findings in the diagnosis of soft tissue tumours with different imaging techniques.
2. To learn the most common differential diagnosis.
290
RPS 801 - Liver steatosis and fibrosis quantification
Categories: Abdominal Viscera, General Radiology, Imaging Methods

CME Credits: 1.5
Moderator:
Tommaso Vincenzo Bartolotta; Palermo / Italy
Gravitational 2D/3D MR-elastography vs pneumatic 2D MR-elastography: prospective evaluation of a novel system (7

min)
Vitali Koch; Frankfurt a. Main / Germany
Author Block: V. Koch1, L. D. Grünewald1, J. Gotta1, S. Mahmoudi1, C. Booz1, R. Hammerstingl1, O. Darwish2, T. Vogl1, R. Sinkus3;
1
Frankfurt a. Main/DE, 2London/UK, 3Paris/FR
Purpose: Magnetic resonance elastography (MRE) can quantify tissue biomechanics non-invasively and represents a promising
technique for assessing fibrosis in nonalcoholic fatty liver disease (NAFLD). This research presents the preliminary results of a
prospective study designed to investigate the value of 2D/3D-MRE utilising the gravitational concept compared to the current product
solution (2D-MRE Resoundant).
Methods or Background: Thirty-two participants with different stages of NAFLD were examined twice at 60Hz (Aera, 1.5T, Siemens
Healthineers, Germany): firstly, using the Resoundant system (2D-MRE, SE-EPI sequence, 11secs BH) and secondly, using the
gravitational transducer approach (2D-MRE and 3D-MRE, GRE sequence, TE=9.2ms (in-phase) and fractional motion encoding at
30mT/m, 14secs BH). While 2D-MRE provides solely the magnitude of the complex shear modulus |G*|, 3D-MRE allows for the
additional quantification of both real and imaginary parts of G*. Data extraction and analysis was performed twice by two readers.
Results or Findings: We found a high level of correlation between the 2D-MRE acquisitions obtained using both the Resoundant
MRE system and the gravitational transducer approach (r≥0.83 (95% CI, 0.69 to 0.92), p<0.001), although there was a clear bias with
2D-MRE overestimating stiffness values. For markers indicative of liver damage, GOT levels were below 30 for |G*|<4kPa, while for
|G*|>4kPa, GOT levels were above 30 in a binary fashion. Interestingly, the 3D-MRE approach showed different correlations
depending on the level of GOT expression: for GOT<30 wave attenuation showed a clear trend wrt GOT (r=-0.27, P=0.37), while for
GOT>30 viscosity correlated very well to GOT (r=0.61, P=0.047). Viscosity for GOT<30 clustered at low values, similar to attenuation
for GOT>30.
Conclusion: Gravitational MRE represents a novel method for the sensitive characterisation of patients with NAFLD, with 3D-MRE
carrying the potential to provide imaging biomarkers for liver damage.
Limitations: This research is based on preliminary data.
Ethics committee - additional information: This study received approval from the relevant local Ethics Committee.
Real-world risk-stratification of patients with chronic liver disease using quantitative magnetic resonance imaging (7
min)
Nabih Nakrour; Watertown / United States
291
Author Block: N. Nakrour1, E. Shumbayawonda2, M. Pansini3, M. Harisinghani1; 1Boston, MA/US, 2Oxford/UK, 3Lugano/CH
Purpose: The objective of this study was to assess the comparative diagnostic value of multiparametric MRI (mpMRI) and magnetic
resonance elastography (MRE) in real-world clinical practice for managing suspected chronic liver disease.
Methods or Background: A retrospective analysis of the prospective MR exams of 77 patients referred to tertiary chronic liver
disease practices. Patients underwent MRE and mpMRI (LiverMultiScan) as a part of their routine clinical care. MRE measures liver
fibrosis with liver stiffness (kPa). LiverMultiScan quantifies liver disease activity (iron-corrected T1, cT1), fat (proton density fat
fraction, PDFF), and iron content (T2*). MRE≥3kPa indicates any level of fibrosis; cT1≥800ms and ≥875ms indicate active and high-
risk disease, respectively.
Results or Findings: 55% (42) of patients were diagnosed with MASLD/MASH and 45% with mixed chronic liver diseases, including
alcoholic liver disease, viral hepatitis, hemochromatosis, high ferritin, among others. The majority, 71% (55), had normal liver
stiffness (≤3kPa); however, 29% (22) of these had active disease (cT1>800ms), with 14% (11) having elevated cT1 indicative of high-
risk disease (cT1>875ms). There was a linear significant correlation between MRE and cT1 (r=0.411, p=0.0004), and those with
elevated MRE (MRE>3kPa) had cT1 864±74ms. cT1 correlated with PDFF (r=0.5, p<0.001), but MRE was not (r=-0.055, p=0.65). cT1
was successful in 99% (76) of patients, while MRE was successful in 90% (69).
Conclusion: MRE and cT1 provide clinically complementary information on liver health state, with cT1 identifying patients with
underlying liver disease activity who are at risk of worse outcomes but have normal liver stiffness. These patients would benefit from
more intense clinical management or surveillance to prevent clinical outcomes. Caution should be taken in utilising MRE in patients
with suspected high liver iron.
Limitations: MRE and cT1 had an unreliable result in one patient. MRE technical failures were in patients with elevated liver iron.
Funding for this study: This study was partially funded.
Ethics committee - additional information: This study was approved with code: Protocol #2020P000706.
Changes in the prevalence of steatotic liver disease under the new nomenclature using MRI-derived proton density fat
fraction (7 min)
Hee Jun Park; Seoul, South Korea / Korea, Republic of
Author Block: H. J. Park, S. Lee, J. S. Lee; Seoul/KR

Purpose: New nomenclatures, MAFLD and MASLD, with revised diagnostic criteria, were proposed in 2020 and 2023 to replace the
term NAFLD. The purpose of this study was to investigate the changes in the prevalence and clinical characteristics of non-alcoholic
fatty liver disease (NAFLD), metabolic dysfunction-associated fatty liver disease (MAFLD), and steatotic liver disease (SLD) in a health
check-up using magnetic resonance imaging-derived proton density fat fraction (MRI-PDFF).
Methods or Background: We included 844 participants who underwent liver MRI-PDFF at our health check-up clinic between
January 2020 and November 2022. Hepatic steatosis was defined as MRI-PDFF≥5%. Participants were categorised according to
NAFLD, MAFLD, MASLD, and sub-classifications of SLD.
Results or Findings: The prevalence of NAFLD, MAFLD, and MASLD were 25.9%, 29.5%, and 25.2%, respectively. 30.5% of the
participants were categorised as SLD. The prevalence of the SLD sub-categories were 25.2% for MASLD, 3.7% for MASLD and alcohol-
associated liver disease (MetALD), 0.1% for alcohol-associated liver disease, 1.3% for specific etiology SLD, and 0.1% for cryptogenic
SLD. Compared with patients in the MASLD group, those in the MetALD group were younger, predominantly male, and exhibited
higher levels of serum aspartate aminotransferase, gamma-glutamyl transpeptidase, and triglycerides.
Conclusion: The prevalence of NAFLD and MASLD assessed using MRI-PDFF was similar, with MASLD accounting for 97.3% of the
patients with NAFLD. MetALD group may have clinical characteristics that differ from MASLD group.
Limitations: The study was limited by the potential selection bias of participant recruitment, which included a relatively expensive
health check-up including a liver MRI. The sample size for the younger age group was also relatively small. Furthermore, information
on excessive alcohol consumption and comorbidities was based on self-reported questionnaires, which may be subject to recall bias.
Finally, access to information on the use of steatogenic medications by the study population was limited.
Funding for this study: Funding for this study was received from the Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education (grant number: RS-2023-00244520).
Ethics committee - additional information: This study was approved by the relevant Institutional Review Board. The requirement
for written informed consent was waived due to the retrospective nature of the analysis.
MRI hepatic steatosis and sarcopenia in metabolic patients: a correlation prospective study (7 min)
Benedetta Masci; Rome / Italy
292
Author Block: B. Masci, M. Zerunian, S. Nardacci, F. Pucciarelli, M. Polici, I. Nacci, D. Principessa, D. Caruso, A. Laghi; Rome/IT
Purpose: The purpose of this study was to evaluate the possible correlation between hepatic steatosis and sarcopenia assessed with
skeletal muscle index (SMI) at unenhanced MRI of the abdomen in patients with multiple metabolic risk factors.
Methods or Background: 105 patients with metabolic syndrome were prospectively enrolled between October 2022 and June 2023.
Each patient underwent 1.5T upper abdomen MRI examination, acquisition protocol included axial proton density fat fraction (PDFF),
magnetic resonance elastography (MRE) and axial T1 weighted gradient-echo sequences targeted at the third lumbar vertebra (L3)
level. A single radiologist performed quantitative image analysis on a dedicated workstation. Steatosis and myosteatosis estimation
expressed as a percentage and liver stiffness (kPa) were collected and grading of steatosis and fibrosis was assigned. The presence of
sarcopenia was assessed by segmenting the L3 dual-echo images with ImageJ pixel analysis software and calculating the SMI.
Correlations among parameters were assessed using Spearman’s rank coefficient using a dedicated software and p<0.05 was
considered significant.
Results or Findings: PDFF values of hepatic steatosis were found in 65% of patients (6.7±1.8%); liver stiffness resulted in higher
compatibility with inflammation values in 35% of patients (2.6kPa±0.2kPa).
SMI average values were compatible with sarcopenia in 71% of patients (Males: 41.5±6.8cm^2/m^2; Females:
33.5±3.2cm^2/m^2).The statistical analysis showed significant correlation between hepatic steatosis and sarcopenia (P<0.03, rho
0.17). Moreover, by stratifying the cohort for sex, a significant correlation between the steatosis grade and sarcopenia in female
patients (P<0.02, rho -0.38) emerges.
Conclusion: Patients with metabolic syndrome and hepatic steatosis are correlated with lower SMI values and sarcopenia, and the
correlation is stronger in female patients. This method might be a non-invasive, radiation-free and repeatable method for a
comprehensive metabolic assessment of the patients at diagnosis and at follow-up.
Limitations: We analysed a small population sample and performed a manual segmentation, which was often time consuming. A
further limitation is the single vendor technology analysis approach taken.
Ethics committee - additional information: This study was approved by the Institutional Review Board, and written informed
consent was obtained for all patients.
Does meal or water intake affect ultrasound attenuation coefficient estimate of liver fat content? (7 min)
Richard G. Barr; Canfield / United States
Author Block: R. G. Barr1, G. Ferraioli2; 1Youngstown, OH/US, 2Pavia/IT

Purpose: The purpose of this study was to assess whether meal or water intake may affect the measurement of the ultrasound
attenuation coefficient imaging, a parameter that is directly related to liver fat content.
Methods or Background: The study was performed in two centres (in Italy and in the USA). Attenuation coefficients were obtained
using the ATI algorithm implemented in the Aplio i-series ultrasound systems (Canon Medical Systems, Japan) by one operator at each
centre. Measurements were performed at baseline and 5, 15, 30, and 45 minutes after drinking 500ml of water (group1), or 30, 45,
60, 90, and 120 minutes after eating a meal of about 600 kilocalories (group2). Multilevel generalised estimating equations for
repeated measures were used for the statistical analysis to consider the clustered nature of the data.
Results or Findings: Twenty-one individuals were enrolled: 11 (10 females; aged 43.7±12.5 years) in Italy and 10 (7 females; aged
59.7±7.1 years) in the USA. At the B-mode ultrasound, six (28.6%) had liver steatosis. The baseline attenuation coefficient values, in
dB/cm/mHz were 0.62 (0.11) in group1 and 0.66 (0.12) in group2. There was not any significant difference in attenuation coefficient
values at every timing either in group1 or group2. This result did not change even when controlling for sex, age, and skin-to-liver
capsules.
Conclusion: The measurement of the attenuation coefficient, which is a biomarker of liver steatosis, does not require a fasting state
and drinking water does not affect the result.
Limitations: Although the findings of this study were statistically significant, the number of participants was small.
Ethics committee - additional information: This study was approved by the Western Reserve Health IRB, Youngstown, Ohio, USA,
and the Fondazione IRCCS Policlinico S. Matteo, Pavia, Italy.
Non-invasive assessment of steatosis and fibrosis in patients with risk factors for NAFLD: agreement among
quantitative US and multiparametric MRI (7 min)
Ilaria Nacci; Rome / Italy
293
Author Block: I. Nacci, F. Pucciarelli, B. Masci, M. Zerunian, D. De Santis, D. Caruso, A. Laghi; Rome/IT
Purpose: The purpose of this study was to assess ultrasound attenuation coefficient (AC) for quantifying liver fat deposit and to
evaluate two-dimensional shear wave elastography (2D-SWE) for quantifying liver fibrosis using MRI-proton density fat fraction (MRI-
PDFF) and magnetic resonance elastography (MRE) as the references, respectively, in patients with risk factors for NAFLD.
Methods or Background: This prospective study included patients with risk factors for NAFLD (diabetes or metabolic syndrome).
Patients with secondary causes of fat deposition were excluded. All participants underwent liver QUS and MRI on the same day, and
laboratory tests within 30 days. Based on MRI-PDFF and MRE, we obtained normal liver, liver steatosis and liver fibrosis groups. We
examined diagnostic performance of AC and 2D-SWE for detecting liver fat content and stiffness using area under receiver operating
characteristic curve (AUC). We also analysed correlations of QUS biomarkers to MRI using Spearman correlation coefficient.
Results or Findings: A total of 54 participants were included. Of these participants, 21 (38.9%) had MRI-PDFF≥5%, and 10 (18.5%)
had MRE≥2.9kPa. AUC of AC for determining greater than and equal mild steatotic livers was 0.80 (95% confidence interval [CI]:
0.67-0.94). AUC of 2D-SWE for determining greater than and equal F1 liver fibrosis was 0.59 (95% CI: 0.40-0.79). AC had a sensitivity
of 71.4% and a specificity of 87.9%, 2D-SWE had a sensitivity of 60% and a specificity of 52.3%. AC did not correlate well with MRI-
PDFF in assessing hepatic steatosis (r=0.43), and 2D-SWE did not correlate with MRE in evaluating liver fibrosis (r=0.39).
Conclusion: Our preliminary results show that QUS biomarkers are more accurate for screening mild hepatic steatosis than fibrosis in
patients with risk factors for NAFLD, but do not accurately discriminate different degrees of steatosis and fibrosis.
Limitations: The sample size was small, and the study was single-centred.
Ethics committee - additional information: The study was approved by the relevant Institutional Review Board, and all patients
gave informed consent.
Ultrasound-based steatosis grading system using 2D-attenuation imaging: an individual patient data meta-analysis
with external validation (7 min)
Marco Dioguardi Burgio; Levallois Perret / France
Author Block: C. Hobeika1, M. Ronot1, D. Valla1, J. M. Correas2, V. Vilgrain1, M. Dioguardi Burgio1; 1Clichy/FR, 2Paris/FR
Purpose: Non-invasive tools assessing steatosis, such as ultrasonography-based 2D-attenuation imaging (ATI), are needed to tackle
the worldwide burden of NAFLD. This one-stage individual patient data (IPD) meta-analysis aimed to create an ATI-based steatosis
grading system.
Methods or Background: A systematic review (EMBASE+MEDLINE, 2018-2022) identified studies, including patients with
histologically or MRI-PDFF-verified ATI values for grading steatosis (S0 to S3). One-stage IPD meta-analyses were conducted using
generalised mixed models with a random study-specific intercept. Created ATI-based steatosis grading system (aS0 to aS3) was
externally validated on a prospective cohort of patients with type 2 diabetes and NAFLD (n=174, histologically and MRI-PDFF verified
steatosis).
Results or Findings: Eleven enrolled studies included 1374 patients, classified into S0, S1, S2, and S3 in 45.4%, 35.0%, 9.3%, and
10.3% of the cases. ATI was correlated with histologically (r=0.60; 95%CI: 0.52,0.67; p<0.001), and MRI-PDFF (r=0.70; 95%CI:
0.66,0.73; p<0.001) quantified steatosis while uncorrelated with liver stiffness (r=0.03; 95%CI: -0.04,0.11, p=0.343). Steatosis grade
(Coefficient: 0.27; 95%CI: 0.07,0.47; p=0.008) was the only independent factor associated with ATI, while age, sex, BMI, chronic
hepatitis, and alcohol consumption were not. ATI marginal means within S0, S1, S2, and S3 subpopulations were 0.59 (95%CI:
0.56,0.61), 0.69 (95%CI: 0.65,0.72), 0.77 (95%CI: 0.73,0.81), and 0.84 (95%CI: 0.80, 0.89) dB/cm/MHz; all contrasts between grades
were significant (p<0.0001). Three ATI thresholds were calibrated to create a new ATI-based steatosis grading system (aS0 to aS3,
cut-offs: 0.66, 0.73, and 0.81dB/cm/MHz). Its external validation showed Obuchowski measures (to interpret as AUC values) of
0.84±0.01 and 0.82±0.02 with histologically- and MRI-PDFF-based references.
Conclusion: ATI is a reliable non-invasive marker of steatosis. This validated ATI-based steatosis grading system could be valuable in
assessing NAFLD patients.
Limitations: There is a difference in steatosis prevalence among the included studies.
Ethics committee - additional information: This study was approved by the local Institutional Research Board.
Depth dependence of the backscatter coefficient measurement for ultrasound fat quantification (7 min)
Giovanna Ferraioli; Pavia / Italy
294
Author Block: G. Ferraioli1, A. De Silvestri1, R. G. Barr2; 1Pavia/IT, 2Rootstown, OH/US
Purpose: It has been reported that the estimate of the ultrasound attenuation coefficient (AC) for liver fat quantification is affected
by the depth of measurements, with a linear decrease of values with depth. The purpose of this paper was to determine whether the
backscatter coefficient (BSC) has the same behavior.
Methods or Background: This retrospective study was performed with the Sequoia ultrasound system equipped with the ultrasound
derived fat fraction (UDFF) algorithm (Siemens Healthineers, Germany) that combines BSC with AC. UDFF was obtained positioning
the upper edge of the region of interest (3x3 cm) at 1.5, 2, 3, 4, 5 cm below liver capsule. BSC data were extracted from UDFF offline.
The median value of five acquisitions was used for statistical analysis. A fractional polynomial regression, which selects the best
model considering the polynomial development of the variables of interest, was used. The covariates included were age, sex, skin-to-
liver-capsule distance, and stiffness. The distance was included as linear factor or with a power of -2;-1;-0.5;0;0.5;1;2;3 or each
possible pair of them. Best fitting models was chosen according to partial F test. Body mass index (BMI) was not included because of
collinearity with skin-to-liver capsule distance.
Results or Findings: Fifty-one individuals (25 females, mean age: 61±13 years; mean BMI: 27±6kg/m2; skin-to-liver-capsule
distance: 2.0±0.5cm; liver stiffness: 8±6kiloPascal) were studied. Best fitting model included depth as square root (beta -40; 95%
confidence intervals from -64 to -15) and linear factor (beta 11; 95% confidence intervals from 4 to 18). Skin-to-liver-capsule distance
and stiffness also were independent predictors of BSC.
Conclusion: There is a depth dependence in liver BSC measurement that can substantially affect results. A standardised acquisition
protocol is needed to compare results and to reliably assess changes in serial measurements.
Artificial intelligence-calculated hepatorenal index for quantitative evaluation of hepatic steatosis in metabolic
dysfunction-associated steatotic liver disease (7 min)
Pál N. Kaposi-Novák; Budapest / Hungary
Author Block: Z. Zsombor, A. D. Rónaszéki, B. Csongrády, R. Stollmayer, B. K. Budai, V. Bérczi, P. Maurovich-Horvat, K. Hagymási, P.
N. Kaposi-Novák; Budapest/HU
Purpose: The goal of our study was to evaluate artificial intelligence-calculated hepatorenal index (AI-HRI) as a diagnostic method for
hepatic steatosis.
Methods or Background: We prospectively enrolled 103 patients with clinically suspected metabolic dysfunction-associated
steatotic liver disease (MASLD). All patients had a quantitative ultrasound (QUS), including AI-HRI, ultrasound attenuation coefficient
(AC), and ultrasound backscatter-distribution coefficient (SC) measurements. The ultrasonographic fatty liver indicator (US-FLI) score
was also calculated. The magnetic resonance imaging fat fraction (MRI-PDFF) was the reference to classify patients into four grades of
steatosis: none <5%, mild 5-10%, moderate 10-20%, and severe ≥20%. We determined the agreement between AI-HRI by two
examiners using the intraclass correlation coefficient (ICC) of 68 cases.
Results or Findings: The AI-HRI was significantly different between groups without (1.480 ±0.607, p<0.003) and with mild steatosis
(2.155 ±0.776), as well as between mild and moderate steatosis (2.777 ±0.923, p<0.018). AI-HRI showed a moderate correlation with
AC (Spearman's r=0.597), SC (rs=0.473), US-FLI (r=0.5), and MRI-PDFF (r=0.528). The agreement in AI-HRI was good between the
two examiners (ICC=0.635, 95% confidence interval (CI)=0.411–0.774, p<0.001). The AI-HRI could detect mild steatosis (AUC=0.758,
95% CI=0.621–0.894) with fair and moderate/severe steatosis (AUC=0.803, 95% CI=0.721–0.885) with good accuracy. However, the
performance of AI-HRI was not significantly different (p<0.578) between the two diagnostic tasks.
Conclusion: AI-HRI is an efficient, reproducible, and accurate QUS method to diagnose mild and moderate hepatic steatosis.
Limitations: This is a single-centre study with a limited number of patients.
Ethics committee - additional information: This study was approved by the Regional and Institutional Science and Research
Ethics Committee at Semmelweis University (Protocol number: SE RKEB 140/2020, 16-07-2020).
Fatty liver disease in the adult population: new multiparametric ultrasound compared to magnetic resonance (7 min)
Carmen Solito; Rome / Italy
295
Author Block: C. Solito, C. Di Bella, P. Pacini, G. T. Lucarelli, M. Renda, V. Dolcetti, G. Del Gaudio, V. Cantisani; Rome/IT
Purpose: Metabolic dysfunction-associated fatty liver disease (MAFLD) is the most frequent chronic liver disease in the adult
population and liver biopsy diagnosis is the gold standard. The purpose of this study is to evaluate the diagnostic accuracy of the new
Hepato-Renal Index with Automated ROI Recommendation (EzHRI™), Tissue Attenuation Imaging (TAI™), and Tissue Scatter
distribution Imaging (TSI™) ultrasound software for hepatic steatosis quantification, comparing with MRI.
Methods or Background: 92 patients underwent multiparametric ultrasound evaluation and 3T MRI evaluation with two different
techniques- 3D T2* with Dixon pulse multiple-echo sequence (IDEAL IQ; MR-PDFF- and MR spectroscopy). The best QUS cut-offs have
been identified and the QUS results are compared with MRI proton density fat fraction for diagnosis and quantification of liver fat
content. Inter-observational variability was also tested between two operators.
Results or Findings: The best cut-off values for MAFLD diagnosis are EzHRI>1.21; TAI>0.63 and TSI>92 with a sensitivity of 97.5%,
88.8% and 98.8% respectively and a specificity of 83.3%, 100% and 66.7% respectively. Comparing QUS with MRI in detection of liver
steatosis, sensitivity and specificity were 88.8% and 100% for TAI, 98.8% and 66.7% for TSI and 97.5% and 83.3% for EzHRI. In the
quantification of liver fat content, TAI sensitivity and specificity were 80%-100%, 100%-100% and 100%-100% for mild, moderate and
severe steatosis respectively; TSI sensitivity and specificity were 97.8%-66.7%, 100%-66.7% and 100%-66.7% for mild, moderate and
severe steatosis respectively; EzHRI sensitivity and specificity were 95.6%-83.3%, 100%-83.3% and 100%-83.3% for mild, moderate
and severe steatosis respectively.
Conclusion: QUS is a useful tool for screening and monitoring patients with MAFLD and can assist in avoiding unnecessary biopsies.
Limitations: This study is limited by interoperator variability, which diminishes its reproducibility. Furthermore, there is reduced
accuracy in patients with obesity or renal disease.
Ethics committee - additional information: This study was approved by Sapienza-Roma. No additional information was provided
by the submitter.
Fully automated MRI-based convolutional neural network for noninvasive diagnosis of cirrhosis (7 min)
Tianying Zheng; Chengdu / China
Author Block: T. Zheng1, Y. Zhu2, Y. Chen1, Y. Qu1, Y. Chen2, B. Song1; 1Chengdu/CN, 2Shanghai/CN
Purpose: This study aimed to develop a fully automated diagnostic convolutional neural network (CNN) model for cirrhosis based on
liver MRI and serum biomarkers.
Methods or Background: This single-centre retrospective study enrolled consecutive patients receiving pathological evaluation of
liver fibrosis stage and contrast-enhanced liver MRI between March 2010 and May 2021. On the training set, an MRI-based CNN model
was constructed for cirrhosis against pathology, and then a combined model was developed integrating the CNN model and serum
biomarkers. On the testing set, the area under the receiver operating characteristic curve (AUC) was computed to compare the
diagnostic performance of the combined model with that of aminotransferase-to-platelet ratio index (APRI), fibrosis-4 index (FIB-4),
and radiologists. The influence of potential confounders on the diagnostic performance was evaluated by subgroup analyses.
Results or Findings: A total of 934 patients (median age, 53 years; 762 men; training, n=840) were enrolled, 601 (64%) with
pathological cirrhosis. The CNN model was constructed on pre-contrast T1- and T2-weighted imaging, and the combined model was
developed integrating the CNN model, age, and eight serum biomarkers. On the testing dataset, the combined model achieved an
AUC of 0.89, which outperformed FIB-4, APRI and two radiologists (AUC: 0.71 to 0.78, all p<0.05). Subgroup analyses revealed
comparable diagnostic performances of the combined model in patients with different sizes of focal liver lesions.
Conclusion: Based on pre-contrast T1- and T2-weighted imaging, age, and serum biomarkers, the combined model allowed accurate
diagnosis of cirrhosis, independent of size of focal liver lesions.
Limitations: The limitations of the study are that it is a single-centre retrospective study, lack of a head-to-head comparison with
elastography techniques, and unknown model performance compared with a combination of radiologists and serum biomarkers .
Funding for this study: Funding was received from the National Natural Science Foundation of China (Grant No. U22A20343,
82101997), National Health Commission Capacity Building and Continuing Education Center (Grant No. YXFSC2022JJSJ007), the 1.3.5
project for disciplines of excellence, West China Hospital, Sichuan University (Grant No. ZYJC21012), Med-X Center for Informatics,
Sichuan University (NO.YGJC007), and the Science and Technology Department of Sichuan Province (No. 2022YFS0071).
Ethics committee - additional information: This study was approved by the Ethics Committee on Biomedical Research, West
China Hospital of Sichuan University (2021-1370).
Noninvasive assessment of liver segmental volumes and its relationship with 5-year prognostication (7 min)
Damiano Catucci; Bern / Switzerland
296
Author Block: D. Catucci, J. Hrycyk, N. F. Lange, V. Obmann, A. Berzigotti, L. Ebner, A. Christe, J. T. Heverhagen, A. T. Huber;
Berne/CH
Purpose: This study aimed to analyse the performance of caudate to right lobe ratio (CRL-R), liver segmental volume ratio (LSVR)
and liver segmental volume and attenuation ratio (LSVAR) to screen for chronic liver disease (CLD) on routine abdominal CT scans
and to predict the 5-year probability of transplant-free survival and first hepatic decompensation.
Methods or Background: This retrospective study included 108 patients without CLD (noCLD-group; n=108), as well as 98 patients
with CLD and liver biopsy. All patients underwent abdominal CT scans between 03/2015 and 08/2017. Patients with CLD were further
divided into three groups based on their liver fibrosis degree: early CLD (F0-F2; eCLD-group; n=40), advanced CLD (F3-F4; aCLD-
group; n=20), and aCLD with clinically significant portal hypertension according to the BAVENO VII consensus (aCLDPH-group; n=38).
CRL-R, LSVR, and LSVAR were measured in all patients. The study analysed the 5-year outcomes of each patient, including death or
liver transplantation and first hepatic decompensation. Statistical analysis included the Kruskal-Wallis test, ROC curve analysis, and
the Kaplan-Meier curve.
Results or Findings: CRL-R, LSVR and LSVAR differed significantly between all groups (p<0.001). A CRL-R cutoff-value of > 0.93
proofed best to detect patients with CLD (sensitivity of 69%, specificity of 78%). Patients with both CRL-R >0.99 and LSVR >0.37 had
the lowest probability of 5-year transplant-free survival (46%) and the lowest probability of a decompensation-free 5-year course
(75%).
Conclusion: CRL-R, LSVR and LSVAR allow screening for CLD and prognostication of 5-year transplant-free survival and occurrence of
first hepatic decompensation.
Limitations: The limitations of this study were its retrospective design and the lack of invasive measurement of the hepatovenous
pressure gradient, which was not available due to ethical concerns.
Funding for this study: This study received funding by the Swiss National Science Foundation (SNF), Grant number 188591.
Ethics committee - additional information: This retrospective study was approved by the Institutional Review Board.
297
E³ 821a - Advances in imaging for prostate cancer

CME Credits: 1.5
PI-RADS update: what is next? (45 min)

Anwar R. Padhani; Northwood / United Kingdom
1. To become familiar with the strengths and limitations of the PI-RADS v2.1 system.
2. To recognise imaging pitfalls and mimickers in prostate cancer.
3. To illustrate the modification of the PI-RADS system aimed to improve diagnostic performance.
AI in prostate imaging (45 min)

1. To become familiar with the applications of AI in the diagnosis and staging of prostate cancer.
2. To learn about the use of AI information to guide biopsy.
3. To discuss the potential impact of AI methods on clinical decision-making and patient outcome.
298
RPS 811 - Foetal and neonatal neuroimaging: epilepsy and the senses
Categories: Forensic Imaging, Imaging Methods, Multidisciplinary, Neuro, Paediatric, Research

CME Credits: 1.5
Moderator:
Cornelius Deuschl; Essen / Germany
Feed and wrap (F&W) technique vs under sedation: quantitative and qualitative analysis of neonatal MRI brain images
(7 min)
Francesca Lisi; Alatri / Italy
Author Block: G. Varcasia, C. Schiarelli, F. Lisi, A. Infante, A. Perna, T. Verdolotti, M. Bottalico, S. Gaudino; Rome/IT
Purpose: The F&W technique refers to the use of feeding and swaddling to induce natural sleep in infants during the MRI scans,
without sedation. The aim of our study was quantitative and qualitative evaluation of neonatal MRI brain images with the F&W
technique vs under sedation.
Methods or Background: In this retrospective, observational, single centre study, neonatal brain MR imaging data performed with
clinical suspicion of hypoxic-ischaemic encephalopathy (HIE) were reviewed between February 2018 and February 2020. Quantitative
evaluation was defined as analysis of measurement of the SNR (signal to noise ratio) and CNR (contrast to noise ratio) in the T2, T1
SE, 3D-TFE, DWI (b1000) sequences with the use of regions of interest (ROI) in the white matter, in the basal ganglia and in the air
space surrounding the skull. Qualitative evaluation was defined as assessment of the quality of the whole images, visualisation of the
cortex and of the basal ganglia with the use of Likert Scale (range 0-2).
Results or Findings: Overall, 210 scans were analysed; 74% performed under sedation, and 26% with F&W technique. Qualitative
analysis showed that 88% of F&W were partially or fully diagnostic against a predictable diagnostic effectiveness of 99.3% of MRI
scans during sedation. Quantitative analysis showed comparable SNR and CNR between the two populations in DWI and T2w, while
3D-T1 showed lower CNR and SNR values in the FW scans. Only T1-SE CNR was comparable between F&W scans and under sedation.
Interobserver agreement was good.
Conclusion: The F&W technique has a high rate of success with high-quality brain imaging in non-sedated infants and can be used
effectively to detect diagnostic features of HIE. Furthermore, it is not burdened by adverse events that may occur during and after
anesthesiological procedures.
One-minute full brain multi-contrast MRI vs brain CT in paediatric patients in the elective setting: a prospective
feasibility study (7 min)
Francesca De Luca; Solna / Sweden
299
Author Block: F. De Luca, A. Kits, D. Martin Muñoz, A. Aspelin, O. Kvist, Y. Osterman, S. Diaz Ruiz, S. Skare, A. Falk Delgado;
Stockholm/SE
Purpose: The radiological examination of paediatric patients with suspected cerebral pathology is performed at first instance with
brain CT when MRI resources and general anaesthesia are unavailable. The study aimed to evaluate the tolerance to a complete
diagnostic fast brain MRI (EPIMix) by paediatric patients without general anaesthesia and to compare the diagnostic accuracy of
EPIMix and brain CT.
Methods or Background: Paediatric patients with a referral for elective brain CT between March 2019 and March 2020 were
prospectively included and underwent EPIMix and brain CT without anaesthesia. Three readers evaluated EPIMix and CT image
datasets separately, blinded from any clinical information. The study outcomes were to assess the tolerance to a complete diagnostic
EPIMix sequence without general anaesthesia and the diagnostic performance in classifying scans as non-pathological or pathological.
The two image modalities were compared using descriptive statistics and Fisher’s exact test. Furthermore, a side-by-side evaluation
of EPIMIx and CT image datasets was performed by two additional unblinded readers and one previously blinded reader.
Results or Findings: All included paediatric patients (n=15) tolerated EPIMix well. The EPIMIx and CT scans were reported as non-
pathological in 13 and 12 cases, respectively, by the three readers, while two cases of EPIMix and three cases of CT were classified as
pathological by one reader (Fisher’s exact test, reader 1–3 p=1.00). Side-by-side evaluation reported all scans as non-pathological on
both EPIMix and CT image datasets.
Conclusion: Paediatric patients tolerated the diagnostic fast brain MRI EPIMix without the need for general anaesthesia. Further,
EPIMix showed a comparable diagnostic performance to brain CT.
An article from this study has been published in BMC Medical imaging. The Version of Record of this article is published in BMC
Medical Imaging and is available online at https://doi.org/10.1186/s12880-024-01196-6
Limitations: The three blinded readers evaluating the images had varied experiences. However, there were no significant
differences in scan classification between EPIMix and CT, which enhances the applicability of our findings.
Funding for this study: Funding for the authors of this study was received from Stockholms Läns Landsting. Funding was also
received from the Svenska Sällskapet för Medicinsk Forskning and Karolinska Institutet.
Ethics committee - additional information: This study was approved by the Swedish Ethical Review Authority.
Diffusion tensor magnetic resonance neuroimaging metrics of neonates with intraventricular haemorrhages predict
neurodevelopmental outcome (7 min)
Selina Seeliger; Vienna / Austria
Author Block: S. Seeliger, P. Kienast, V. Schmidbauer, K. Goeral, J. Elis, D. Prayer, G. Kasprian; Vienna/AT
Purpose: Intraventricular haemorrhages (IVH) are a major cause of neurodevelopmental impairment in premature neonates.
Approximately 15-20% of neonates delivered prior to reaching 32 weeks of gestation (GW) suffer from IVH. This study aims to identify
new noninvasive markers based on diffusion tensor magnetic resonance imaging (DTI) to accurately predict neurodevelopmental
outcomes in preterm infants with IVH.
Methods or Background: Neuroimaging was performed on 41 (n=15, 37% female) preterm neonates (mean GW 26,34 ± 2.6 weeks
SD at the time of birth) with IVH using DTI. Regions of interest (ROIs) were placed in early myelinating structures. Fractional
anisotropy (FA) and apparent diffusion coefficient (ADC) were measured in the set ROIs. Neonatal outcome scores were calculated
using the Bayley Scales of Infant and Toddler Development conducted at 12 months and 24 months corrected age. Statistical
analyses, including Pearson's correlations and ANOVA, were performed to establish correlations between DTI metrics and
neurodevelopmental outcomes.
Results or Findings: FA of the medulla oblongata showed a highly significant correlation with the one-year outcome (r=-0.559,
p=0.004). FA of the right internal capsule showed a significant correlation between one-year outcome (r=-0.404, p=0.045) and two-
year language outcome (r=-0.658, p=0.020). FA of the left internal capsule showed highly significant correlations between cognitive
(r=-0.525, p=0.007), motor (r=-0.637, p<0.001), and motor outcome (r=-0.527, p=0.007) after 12 months corrected age.
Conclusion: DTI-based MRI provides valuable prognostic information on future neurodevelopmental outcomes after IVH. DTI
enhances the sensitivity of neonatal MR neuroimaging and could support clinical decision-making for these vulnerable patients.
Funding for this study: Funding for the authors of this study, Patric Kienast and Julia Elis, was provided by the Vienna Science and
Technology Fund (WWTF; LS20-030, PIMIENTO).
Ethics committee - additional information: The study was approved with the approval code: EK 2455/2020.
Quantitative synthetic MRI- vs MR-fingerprinting-based relaxometry of foetal brainstem myelination: a post-mortem

imaging study (7 min)
Jakob Malik; Vienna / Austria
300
Author Block: V. Schmidbauer, J. Malik, I-V. A. Malla Houech, J. Binder, E. Gelpi, D. Prayer, G. Kasprian; Vienna/AT
Purpose: Novel quantitative magnetic resonance imaging (MRI) modalities provide relaxometric properties that are linked to
myelinogenesis. The aim of this work was to investigate ongoing myelin development in foetuses based on a quantitative post-
mortem imaging approach using Synthetic MRI- and MR-Fingerprinting (MRF). Furthermore, the results of both modalities were
compared.
Methods or Background: In 26 cases, quantitative post-mortem foetal MR data were available. Relaxometric measurements (T1-
/T2-relexation times (T1R/T2R)) were determined in the medulla oblongata and the midbrain using Synthetic MRI-/MRF-specific post-
processing procedures. Pearson’s correlations were applied to detect relationships between T1R/T2R metrics and gestational age (GA)
at MRI. Intra-class correlation coefficients were calculated to proof the consistency of the results provided by both modalities.
Results or Findings: Both modalities provided quantitative data that revealed negative correlations with GA at MRI: Synthetic MRI-
derived T1R (medulla oblongata (r=-0.459; p=0.021); midbrain (r=-0.413; p=0.040)) and T2R (medulla oblongata (r=-0.625;
p<0.001); midbrain (r=-0.571; p=0.003)) versus MRF-derived T1R (medulla oblongata (r=-0.433; p=0.035); midbrain (r=-0.386;
p=0.062)) and T2R (medulla oblongata (r=-0.883; p<0.001); midbrain (r=-0.890; p<0.001)). Results consistency between both MR
approaches ranged between 0.661; confidence interval (CI: 0.351 – 0.841) (T2R: medulla oblongata) and 0.920 (CI: 0.821 – 0.965)
(T1R: midbrain).
Conclusion: Synthetic MRI and MRF provide consistent results. Both modalities hold high sensitivity to quantify ongoing
myelinogenesis prenatally. Therefore, quantitative MRI holds promising potential for the assessment of biochemical aspects of brain
maturation in the second half of pregnancy.
Limitations: The study was limited by its small sample size.
Ethics committee - additional information: This study received ethical approval with the code: 1585/2023.
The visibility of the indusium griseum on foetal MRI postmortem and in vivo (7 min)
Ivana Pogledic; Vienna / Austria
Author Block: I. Pogledic1, M. Bobić-Rasonja2, C. Mitter1, A. Štajduhar2, E. Schwartz1, G. Kasprian1, M. Judaš2, D. Prayer1, N. Jovanov-
Milošević2; 1Vienna/AT, 2Zagreb/HR
Purpose: The study aimed to assess the visibility of the indusium griseum (IG) on MR scans of the foetal human brain and to evaluate
its reliability as an imaging biomarker of the regularity of brain midline development.
Methods or Background: The retrospective observational study encompassed T2-w 3T MR images postmortem from ninety foetal
brains and immunohistochemical sections from 41 foetal brains with regular brain development. Three raters independently
evaluated the visibility of IG, and weighted kappa statistics and regression analysis were used for visibility evaluation.
Results or Findings: The visibility of the IG was the highest between the 25 and 30 GW period (inter-rater kappa 0.623-0.709) and
excellent intra-rater variability (kappa 0.81 – 0.93). The immunochemical analysis of the IG discloses the expression of highly
hydrated extracellular molecules in IG as the substrate of higher signal intensity and the best visibility of IG during the mid-foetal
period. The IG anatomic position above the callosal trunk could influence the measurements of CC thickness when imaging methods
are used at its developmental peak. Super-resolution imaging modalities may improve the visibility of the IG in in vivo MRI in the near
future, strengthening the value of IG as a biomarker.
Conclusion: The knowledge of developmental brain histology and foetal age allows us to predict the IG-visibility in MRI and use it as
a biomarker to evaluate the morphogenesis of the brain midline. IG is particularly interesting as a biomarker in postmortem
pathological examination by MRI.
Limitations: The number of encompassed cases due to the unavailability of MRI scans without brain pathologies is the main
limitation of this study; therefore multicentric foetal MR studies, with more cases included, could elucidate the role of IG in health and
disease and its prognostic value.
Funding for this study: Funding was received from the Croatian Science funds: IP-2019-04-3182, DOK-01-3771, DOK-02-5988);
University of Zagreb School of Medicine Funds: (10106-22-3116; 10106-23-2487), and by the European Regional Development Fund:
“Experimental and clinical research of hypoxic-ischemic damage in perinatal and adult brain” (GA KK01.1.1.01.0007).
Ethics committee - additional information: Approval for the histological part of the study was given by the Ethics Committee of
the University of Zagreb, School of Medicine, following the Declaration of Helsinki (No.: 380-59-10106-19-111/210
class:641-01/19-02/01). The MRI part of the study was approved by the Ethics Committee of the Medical School of the University of
Vienna (EK 1211/2019).
Regional cortical thinning and aberrant cortical folding in sleep-related hypermotor epilepsy (7 min)
Huaxia Pu; Chengdu / China
301
Author Block: H. Pu, X. Su, S. Zhang, Q. Yue, Q. Gong; Chengdu/CN
Purpose: Sleep-related hypermotor epilepsy (SHE) is a focal epilepsy whose neurobiological factors such as cortical surface structure
alterations remain poorly understood. The objective of this study was to investigate the neuromorphometric abnormalities and their
relationship with clinical characteristics in SHE patients using structural magnetic resonance imaging (MRI).
Methods or Background: 69 patients with SHE and 69 healthy controls (HCs) were prospectively recruited and underwent high-
resolution T1-weighted MRI scanning. The surface-based morphometry (SBM) analysis based on Computational Anatomy Toolbox
(CAT12) was used to evaluate structural parameters such as cortical thickness, fractal dimension, gyrification, and sulcal depth. The
two groups were compared with two-sample t-tests (p<0.05, family-wise error (FWE) corrected at cluster level). The association
between the cortical morphological changes in local brain regions and the epilepsy duration or the seizure frequency was explored
with correlation analysis.
Results or Findings: SHE patients exhibited significantly decreased cortical thickness in the left inferior-parietal, right superior-
parietal, bilateral lateral-occipital cortices, left fusiform, and lingual gyri (p=0.00001 - 0.00004), and decreased sulcal depth in the
right precentral and postcentral gyri (p≤0.00001) compared to HCs . However, there were no statistically significant differences in the
fractal dimension or gyrification between SHE and HCs (all P>0.05). Correlation analysis showed a negative correlation between the
left inferior-parietal cortical thickness alteration and the duration of SHE (r=-0.325, p=0.006).
Conclusion: Our study revealed that cortical thinning and aberrant folding of local cerebral cortices existed in SHE and may be
associated with clinical hypermotor semiology. These findings might provide a foundation for understanding the epilepsy network
disturbances and the underlying neuropathologic mechanism of SHE.
Limitations: This study used a cross-sectional design with a relatively small sample size. Some patients received anti-epileptic drugs,
which may impact the results.
Funding for this study: Funding was received from the National Natural Science Foundation of China (Grant No. 82271961).
Ethics committee - additional information: This study was approved by the local Ethical Committee of West China Hospital,
Sichuan University.
Task-negative brain activity as an indicator of task engagement in preoperative temporal lobe epilepsy patients: a
functional MRI study (7 min)
Radheshyam Stepponat; Vienna / Austria
Author Block: R. Stepponat, M. S. Yildirim, J. Leinkauf, F. Fischmeister, S. Bonelli, G. Kasprian; Vienna/AT

Purpose: This study aimed to analyse the relationship between task-negative and task-positive brain activity, focusing on the
stability of the default mode network during verb-generation and sentence-comprehension and its relationship to language
lateralisation and task-engagement.
Methods or Background: In clinical practice, fMRI is mainly used for preoperative language lateralisation assessment, commonly
employing verb-generation (VG) and sentence-comprehension (SC) tasks. Language lateralisation is determined by calculating the
lateralisation index (LI) for task-positive brain activity, but variability in task performance can lead to uncertain results. Twenty-three
patients (12 male, 11 female), averaging 28.6 years, with lesional (n=13) and MR-negative (n=10) temporal lobe epilepsy underwent
preoperative fMRI. Standard preprocessing was conducted. The LI toolbox was used to assess temporal lobe lateralisation for the task-
active contrast. Region of interest (ROI) analysis of the posterior cingulate cortex (PCC, Brodmann-area 23 and 31) at a threshold of
p≤0.01 was conducted for the task-negative contrast.
Results or Findings: Sixteen patients displayed consistent temporal lobe lateralisation: 13 left and 3 bilateral lateralization, while
seven patients exhibited varying lateralisation between tasks.
In VG task-negative ROI analysis, four patients lacked PCC activation, while others showed left (n=3), right (n=3), or bilateral (n=13)
activation. In SC all but one patient displayed bilateral PCC activation. Divergent lateralisation between tasks was significantly
associated with PCC inactivity in VG task-negative ROI analysis using X2 (1, N=23)= 4.5, p=0.033.
Conclusion: The absence of PCC activation during VG may be indicative of incompliance, epilepsy-associated DMN changes, and
cognitive impairment due to epilepsy, all of which may limit the clinical usability of the data. Further functional connectome subgroup
analysis is underway.
Limitations: ROI analysis conducted in the study was limited to PCC. The study's sample size is a further limitation; a larger sample
size may yield more significant correlations. Task-compliance was also only partially documented.
Funding for this study: No funding was received for this study, which was conducted as part of a PhD-thesis.
Ethics committee - additional information: This study was approved by the Medical University of Vienna, EK: 1141/2023.
Altered neurovascular coupling in thyroid-associated ophthalmopathy: a combined resting-state fMRI and arterial spin
labelling study (7 min)
Wen Chen; Suzhou / China
302
Author Block: W. Chen1, H. Hu2, X-Q. Xu2, F-Y. Wu2; 1Suzhou/CN, 2Nanjing/CN
Purpose: Besides the well-documented ophthalmic manifestations, thyroid-associated ophthalmopathy (TAO) is believed to be
related to emotional and psychological abnormalities. Given the previous neuroimaging evidence, we hypothesised that TAO patients
would have altered neurovascular coupling associated with clinical-psychiatric disturbances. The purpose of this study was to
investigate neurovascular coupling changes in TAO by combining resting-state functional magnetic resonance imaging (rs-fMRI) and
arterial spin labelling (ASL) techniques.
Methods or Background: Amplitude of low-frequency fluctuation (ALFF) was calculated from rs-fMRI (evaluating neuronal activity),
and cerebral blood flow (CBF) was computed from ASL (evaluating vascular response) in 37 TAO patients and 21 healthy controls
(HCs). Global neurovascular coupling was assessed by across-voxel CBF-ALFF correlation, and regional neurovascular coupling was
evaluated by CBF/ALFF ratio. Auxiliary analyses were performed by using fractional amplitude of low-frequency fluctuation (fALFF) and
regional homogeneity (ReHo) as rs-fMRI measures.
Results or Findings: TAO patients showed significantly reduced global CBF-ALFF coupling compared with HCs. Moreover, TAO
patients exhibited decreased CBF/ALFF ratio in the left lingual gyrus (LG)/fusiform gyrus (FFG), and increased CBF/ALFF ratio in the
bilateral precuneus (PCu). In TAOs, CBF/ALFF ratio in the left LG/FFG was positively correlated with visual acuity, whilst CBF/ALFF ratio
in the bilateral PCu was negatively correlated with MoCA score. The auxiliary analyses showed trends of reduced global neurovascular
coupling (i.e. CBF-fALFF correlation and CBF-ReHo correlation), as well as significant altered regional neurovascular coupling (i.e.
CBF/fALFF ratio and CBF/ReHo ratio) in several brain regions.
Conclusion: TAO patients had altered neurovascular coupling in the visual and higher order cognitive cortices. The neurovascular
decoupling might be a possible neuropathological mechanism of TAO.
Limitations: The sample size was relatively small, and the cross-sectional design may limit the evaluations of brain changes along
with the development of the disease.
Funding for this study: Funding was received from the National Natural Science Foundation of China (NSFC) (81801659 to Hao Hu)
and Clinical Capability Promotion Project of Jiangsu Province Hospital (to Xiao-Quan Xu).
Ethics committee - additional information: This study was approved by the Ethics Committee of the First Affiliated Hospital of
Nanjing Medical University (ref: 2021-SRFA-024).
Structural and functional brain changes after glucocorticoid therapy in thyroid-associated ophthalmopathy (7 min)
Qian Wu; Nanjing / China
Author Block: Q. Wu, J. Zhou, W-H. Jiang, X-Y. Pu, X-Q. Xu, H. Hu, F-Y. Wu; Nanjing/CN
Purpose: The purpose of this study was to investigate the brain structural and functional alterations in patients with thyroid-
associated ophthalmopathy (TAO) before and after glucocorticoid therapy, using voxel-based morphometry (VBM) as well as resting-
state functional magnetic resonance imaging (MRI) with amplitude of low-frequency fluctuation (ALFF) and regional homogeneity
(ReHo).
Methods or Background: Between 2019 and 2022, 32 patients with TAO and 23 healthy controls were recruited to undergo pre-
therapy MRI scans. Intravenous glucocorticoid therapy was administered to all patients. Twenty-six of the patients were available for
rescanned MRI three months after the end of therapy. The VBM, ALFF and ReHo methods were used to evaluate the brain structural
and functional differences.
Results or Findings: Before therapy, TAO patients showed significantly decreased grey matter volume in the left orbital part of
superior frontal gyrus (ORBsup) and the medial superior frontal gyrus (SFGmed) than healthy controls. Patients had higher ALFF
values in bilateral gyrus rectus and olfactory cortex as well as lower values in bilateral cuneus. The patients also showed decreased
ReHo values in bilateral lingual gyrus. After therapy, increased grey matter volume in left anterior cingulate gyrus and SFGmed,
increased ALFF values in bilateral cuneus and superior occipital gyrus, as well as increased ReHo values in bilateral SFGmed were
found in TAO patients compared to the pre-therapy cohort. Compared to controls, decreased grey matter volume in the left ORBsup
was observed in post-therapy TAO patients.
Conclusion: Our results indicated that TAO might cause functional and structural deficits in the visual and emotional regions of the
brain, with recovery in the former and partial restoration in the latter after effective glucocorticoid therapy. These findings may lead
to a deeper understanding of the pathophysiological mechanism behind TAO.
Limitations: The sample size was relatively small.
Funding for this study: Funding was received from the National and Natural Science Foundation of China (NSFC) (81801659 to Hao
Hu), Jiangsu Province Capability Improvement Project through Science, Technology and Education (JSDW202243 to Fei-Yun Wu), and
Jiangsu Province Hospital (the First Affiliated Hospital, Nanjing Medical University) Clinical Capability Enhancement Project (JSPH-
MC-2021-8 to Xiao-Quan Xu).
Ethics committee - additional information: This study was approved by the Review Board of the First Affiliated Hospital, Nanjing
Medical University.
The neural correlates of hearing loss-related distress: an fMRI study (7 min)

Faten Aldhafeeri; Hafar al-Batin / Saudi Arabia
303
Author Block: F. Aldhafeeri; Hafar Albatin/SA
Purpose: Hearing loss (HL) is associated with a decline in the ability to hear high-frequency sounds, resulting in impaired speech
comprehension, particularly under adverse listening conditions. The presence of hearing loss in older adults has been found to have a
substantial influence on their overall quality of life and general well-being. Additionally, there is evidence suggesting a potential
correlation between HL and the development of depressive symptoms as well as feelings of social isolation. The purpose of this study
was to examine how hearing impairment affects the brain's emotional processing network.
Methods or Background: A total of 16 subjects with hearing loss and 15 age and gender-matched healthy controls were recruited
in this study. All participants underwent functional magnetic resonance imaging (fMRI) during which they were asked to listen to
emotionally evocative sound clips. These were rated as pleasant unpleasant or neutral. The sounds were obtained from the
International Affective Digitised Sounds (IADS). To examine the impact of hearing loss severity on neuronal activity, a correlation
analysis between the mean fMRI blood oxygen level-dependent (BOLD) signal and the hearing thresholds in both groups were
performed.
Results or Findings: Compared to the healthy control group, HL subjects demonstrated exaggerated brain activation in auditory,
limbic and frontal regions. HL group exhibited a positive correlation between the mean fMRI BOLD signal and hearing threshold in the
right superior temporal gyrus and anterior cingulate.
Conclusion: Current findings indicate that hearing loss leads to neuronal reorganisation, particularly in brain networks that are
involved in the processing of emotions.
Limitations: This study is limited by a relatively small number of subjects. The contribution of the scanner's noise with the auditory
stimuli is a further limitation.
Ethics committee - additional information: This study was approved by the University's Local Research Ethics Committee.
Causal associations of genetically determined tinnitus with neuroimaging traits: evidence from a Mendelian
randomisation study (7 min)

Purpose: Potential reverse causality and unmeasured confounding factors are common biases in most neuroimaging studies on
tinnitus and central correlates. The causal association of tinnitus with neuroimaging features also remains unclear. This study aimed
to investigate the causal relationship of tinnitus with neuroplastic alterations using Mendelian randomisation (MR).
Methods or Background: Summary-level data from a genome-wide association study (GWAS) of tinnitus were derived from UK
Biobank (n=117,882). The GWAS summary statistics for four global-brain tissues and 14 sub-brain gray matter volumetric traits were
also obtained (n=up to 33,224). A bidirectional MR analysis was conducted to explore the causal relationship between tinnitus and
neuroanatomical features at global-brain and sub-brain levels.
Results or Findings: Genetic susceptibility to tinnitus was causally associated with increased white matter volume (OR=2.361, 95%
CI 1.033–5.393) and total brain volume (OR=2.391, 95% CI 1.047–5.463) but inversely associated with cerebrospinal fluid volume
(OR=0.362, 95% CI 0.158–0.826). A smaller grey matter volume in the left Heschl’s gyrus and right insular cortex and larger grey
matter volume in the posterior division of the left parahippocampal gyrus may lead to an increased risk for tinnitus (OR=0.978, 95%
CI 0.961–0.996; OR=0.987, 95% CI 0.976–0.998; and OR=1.015, 95% CI 1.001–1.028, respectively).
Conclusion: Genetic susceptibility to tinnitus was causally associated with increased white matter volume and total brain volume.
Volume alteration in several cortical regions may indicate a higher tinnitus risk, and further research is recommended for causality
inference at the level of sub-brain regions. Our findings provide genetic evidence for elucidating the underlying pathophysiological
mechanisms of tinnitus-related neuroanatomical abnormalities.
Limitations: The difference in the genetic background between different ancestries limits the broader application of our findings.
Funding for this study: Funding was received as part of Grant 62171297 (H.L.) from the National Natural Science Foundation of
China, No. ZYLX202101 from Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support, No. 2021-135
from Beijing Municipal Health Commission-Beijing Key Clinical Discipline Funding.
Ethics committee - additional information: This study was based on publicly available summary data, and ethical approval of
each GWAS study was obtained from the relevant institutional review board.
Multimodal quantitative magnetic resonance imaging of the thalamus supports distinct outcomes in tinnitus patients
treated with sound therapy (7 min)
Qian Chen; Beijing / China
304
Author Block: Q. Chen; Beijing/CN
Purpose: The objective of this study was to systematically investigate structural and functional alterations in the thalamus and its
subregions using multimodal magnetic resonance (MR) imaging and examine its clinical relevance in tinnitus patients with different
outcomes after sound therapy (narrowband noise).
Methods or Background: In total, 60 patients with persistent tinnitus and 57 healthy controls (HCs) were recruited. Based on
treatment efficacy, 28 patients were categorised into the effective group and 32 into the ineffective group. Five MR imaging
measurements of the thalamus and its seven subregions, including gray matter volume, fractional anisotropy, fractional amplitude of
low-frequency fluctuation, and functional connectivity (FC), were obtained for each participant and compared between the groups.
Results or Findings: Patients in both groups exhibited widespread functional and diffusion abnormalities in the whole thalamus and
several subregions, with more obvious changes observed in the effective group. All tinnitus patients had abnormal FC compared with
the HCs; FC differences between the two patient groups were only observed in the striatal network, auditory-related cortex, and the
core area of the limbic system. We combined the multimodal quantitative thalamic alterations and used it as an imaging indicator to
evaluate prognosis before sound therapy and achieved a sensitivity of 71.9% and specificity of 85.7%.
Conclusion: Similar patterns of thalamic alterations were identified in tinnitus patients with different outcomes, with more obvious
changes observed in the effective group. Our findings support the tinnitus generation hypothesis of frontostriatal gating system
dysfunction. A combination of multimodal quantitative thalamic properties may be used as indicators to predict tinnitus prognosis
before sound therapy.
Limitations: This is not a longitude study.
Funding for this study: Funding was received as part of Grant No. 82302284 from the National Natural Science Foundation of
China.
Ethics committee - additional information: This study was approved by the Ethics Committee of Beijing Friendship Hospital,
Capital Medical University (No. 2017-P2-134-01).
305

ETC Level: LEVEL II
CME Credits: 0.5
Moderators:
Chantal Van Ongeval; Leuven / Belgium


3. To gain deep knowledge of breast radiology and have the opportunity to grow.


306
HW 8R - Framework for Online Radiographer Clinical Education (FORCE)
Categories: Education, Radiographers

CME Credits: 1
Registered participants will need to have a Gmail account to use during the session. Please make sure to have it ready beforehand.

1. To describe best practices, tips and tricks for using FORCE resources within radiography and radiology education.
2. To demonstrate and apply best practices, tips and tricks for using FORCE resources within education.
The FORCE Project: presentation, live software demonstration and Q&A (60 min)
Kate Matthews; Dublin / Ireland
307
CTiR 8 - Clinical Trials in Radiology 1
Categories: Cardiac, Genitourinary, GI Tract, Imaging Methods

CME Credits: 1.5
Moderators:
Monika Radikė; Liverpool / United Kingdom

Incidence of advanced adenoma and cancer after screening CT colonography versus three faecal immunochemical test
rounds in the SAVE randomised trial (8 min)
Lapo Sali; Florence / Italy
Discussant (4 min)
Ciara O'Brien; Toronto / Canada
Multiparametric magnetic resonance imaging in prostate cancer screening at the age of 45: results from the first
screening round of the PROBASE trial (8 min)
Matthias Boschheidgen; Düsseldorf / Germany
Discussant (4 min)
Vibeke Berg Logager; Herlev / Denmark
A randomised trial to assess the impact of preoperative MRI on oncological outcomes following radical prostatectomy:
a 12-year follow-up (8 min)
Daniyal Noor; Finstadjordet / Norway
Discussant (4 min)
Lorenzo E. Derchi; Genoa / Italy
Systematic review of data sharing models in published single- and multicentre medical imaging trials (8 min)
Maria Bosserdt; Berlin / Germany
Discussant (4 min)
308
Medical imaging trial data for quantitative imaging and artificial intelligence methods: multicentre survey of data
sharers and reusers and overview of existing data sharing platforms (8 min)
Melanie Estrella; Berlin / Germany
Discussant (4 min)
Leonor Cerda Alberich; Valencia / Spain
The DISCHARGE pretest probability calculator for the presence of obstructive coronary artery disease (8 min)
Mahmoud Mohamed; Berlin / Germany
Discussant (4 min)
Robert Manka; Zürich / Switzerland
309
RPS 804 - Quantitative analysis of chest CT findings
Categories: Artificial Intelligence & Machine Learning, Chest, General Radiology

CME Credits: 1.5
Moderator:
Mariaelena Occhipinti; Pisa / Italy
Elexacaftor/ tezacaftor/ ivacaftor influences body tissue composition in adults with cystic fibrosis: a fully automated
CT-based analysis (7 min)
Marcel Opitz; Essen / Germany
Author Block: M. Opitz, S. Zensen, J. Haubold, B. M. Schaarschmidt, M. Forsting, L. Umutlu, R. Hosch, F. Nensa, D. Westhölter;
Essen/DE
Purpose: A poor nutritional status is associated with worse pulmonary function and survival in people with cystic fibrosis (pwCF). CF
transmembrane conductance regulator (CFTR) modulators can improve both pulmonary function and body weight, but more data is
needed to evaluate its effects on body composition. This study aims to investigate the body composition in pwCF receiving triple-
combination elexacaftor/ tezacaftor/ ivacaftor (ETI) therapy.
Methods or Background: A pre-trained, deep-learning network was used to perform a fully automated body composition analysis
(BCA) on chest CTs from adult pwCF before and after receiving ETI therapy. Muscle and adipose tissues were quantified and divided
by bone volume to obtain body size-adjusted ratios. Results from BCA were correlated with pulmonary function parameters.
Results or Findings: At baseline, chest CT-based BCA was conducted in 66 pwCF, with 33 (50%) receiving either mono or dual-
combination CFTR modulator therapy. Mono/ dual-combination CFTR modulator therapy was associated with higher intra- and
intermuscular adipose tissue and epicardial adipose tissue ratios, while other BCA markers remained unchanged. After receiving
triple-combination ETI, marked increases were observed in all adipose tissue ratios among pwCF, including the total adipose tissue
ratio (+46.21%, p<0.001). In contrast, only small, but statistically significant, increases of the muscle ratio were measured in the
overall study population (+1.63%, p=0.008). There were weak associations between the rate of change of the muscle ratio and the
rate of change of percent predicted FEV1 (r=0.360, p=0.004).
Conclusion: Our findings suggest that CFTR modulator therapies primarily affect adipose tissues, not muscle tissue, in adults with
CF. These findings may have implications for the future nutritional management of pwCF.
Limitations: CT-based BCA was performed in chest CT scans. Abdominal CT scans or full-body CT scans were not available and body
composition might differ in other parts of the body.
Funding for this study: Funding was received from the German Research Foundation (DFG)-initiated clinician scientist program FU
356/12–2 (DW, JH).
Ethics committee - additional information: The study was approved by the Institutional Review Board (no. 22-11073-BO).
Association between altered subcutaneous adipose tissue and survival in a high-risk population of heavy smokers
participating in lung cancer screening (7 min)
Fabian Bernhard Pallasch; Freiburg im Breisgau / Germany
310
Author Block: F. B. Pallasch, J. Weiß, F. Bamberg, M. Jung, J. B. Fingerhut, M. Reisert; Freiburg im Breisgau/DE
Purpose: There is increasing evidence that body composition is of prognostic relevance. While most studies focus on musculature,
little is known about adipose tissue. The purpose of this study was to apply a deep-learning model for automatic 3D quantification of
subcutaneous adipose tissue (SAT) on chest CT and investigate its association with mortality in a lung cancer screening population.
Methods or Background: 3D SAT was automatically segmented on chest CT of 26,144 individuals participating in the National Lung
Screening Trial at baseline and after one year follow-up (n=52,228 scans). SAT volume (SATvol) and density (SATHU) were quantified.
The primary outcome was all-cause mortality. Additional outcomes were lung cancer and cardiovascular mortality. Cox regression was
used to assess the association between SATvol and SATHU at baseline, and a decrease in SATvol and SATHU (≥10%) and mortality.
Results or Findings: In 26,144 individuals (age 61.4±5.0 years; 40.9% female) 1839 (7%) deaths occurred over a median follow-up
of 6.5 years. At baseline, only SATHU was associated with all-cause mortality after multivariable adjustment for clinical risk factors
(age, sex, race, smoking status, pack years, prevalent hypertension, diabetes, past stroke and myocardial infarction (HR: 1.07 95%CI
(1.02-1.12); p=0.003). After one year, individuals with a decline in SATvol or SATHU ≥10% had a significantly worse outcome
compared to stable SATvol or SATHU (HR adjusted for the same risk factors SATvol: HR 1.94, 95%CI (1.77-2.13), p<0.001; SATHU: HR
3.1, 95%CI (2.79-3.44), p<0.001). Similar associations were found for lung cancer and cardiovascular mortality.
Conclusion: Deep learning models can quantify SAT on chest CT scans. SATHU at baseline and a decrease in SATvol and SATHU
within one year are associated with mortality beyond clinical risk factors, which may help to improve personalised risk assessment.
Limitations: The study is limited by its retrospective analysis.
Ethics committee - additional information: The study is based on a publicly available dataset.
Assessment of chest wall muscle atrophy by computed tomography and correlation with cardiopulmonary function in
systemic sclerosis (7 min)
Martina Giannetti; Rome / Italy
Author Block: M. Giannetti, S. Paciulli, A. Gigante, A. Colalillo, L. Marchitelli, L. Conia, C. Catalano, N. Galea; Rome/IT
Purpose: Systemic sclerosis (SSc) is a multi-organ disease, which can affect the lung parenchyma resulting in an interstitial lung
disease (ILD). Pulmonary function in those patients is influenced by lung disease and chest wall muscles (CWM) activity. High-
resolution CT (HRCT) is a valuable tool to evaluate ILD progression and CWM atrophy. The purpose of this study was to investigate the
potential contribution of CWMarea (CWMA) to the ventilatory efficiency and exercise capacity in patients with SSc.
Methods or Background: Forty-four SSc patients undergone cardiopulmonary exercise testing (CPET), HRCT and transthoracic
echocardiography (TTE) were retrospectively enrolled. The CWMA was manually traced at the level of the 9th thoracic vertebra on CT
images. CWMA was correlated with SSc duration and CPET parameters.
Results or Findings: The patient median age was 53 (IQR 43.5-58) with a BMI of 22.10 kg/m2 (IQR 24.25-20.55).
The median disease duration was 9.50 years (IQR 5.5-15) and the median CWMA was 43.9 cm^2 (IQR 36.8-56.5). We found a median
%FVC of 100 (IQR 107-88), a %FEV1 of 95 (IQR 104.5-86), a %TLC of 93 (IQR 103.5-81.5), a %DLCO of 77 (IQR 86-68.5), and a
DLCO/Va of 81.5% (IQR 93.5-69) for pulmonary functional parameters. TTE revealed a median EF of 60% (IQR 62.5-60), a RV and LV
transverse diameter of 29mm (IQR 31-26 and 44mm (IQR 46-42) respectively), a TAPSE of 22 (IQR 25-21), and a PAPs of 28 (IQR
30-25). CPET parameters used were Watt-max 80% (IQR 105.5-62), VO2max 1211ml/min (IQR 1451-1026.5), VO2max 20.7ml/min/Kg
(IQR23.82-17.9), VO2@AT 789.0ml/min (IQR 952-679), OUES 1364ml/min/L/min (IQR 1629-1229), VEmax 49.55L/min (IQR 60.5-44.4),
VTmax 1.51L (IQR 1.88-1.21), VO2/HRmax 7.95mL/beat (IQR 9.35-6.90), and SpO2max 98% (IQR 99-97). A significant positive
bivariate correlation (P<0.05) was demonstrated between CWM area, and BMI and CPET parameters.
Conclusion: Patients with SSc see a rediction in their accessory respiratory muscle volume and this correlates with a decline in
cardiopulmonary function.
Limitations: The study is limited by its small sample size and single centre methodology.
Image registration enables quantitative regional correlation of structural and functional abnormalities assessed with
CT and MRI in COPD (7 min)
Yiling Xu; Heidelberg / Germany
311
Author Block: Y. Xu1, O. Weinheimer1, S. Triphan1, J. Grolig1, O. von Stackelberg1, J. Biederer1, C. J. Galban2, H-U. Kauczor1, M.
Wielpütz1; 1Heidelberg/DE, 2Ann Arbor, MI/US
Purpose: In the assessment of COPD, quantitative indices of emphysema, and functional small airways disease (fSAD), as well as
perfusion were developed separately for CT and MRI, respectively. Their regional interdependencies, however, have not been
assessed. The aim of this study was to utilise a machine-learning-based image registration pipeline to align pulmonary abnormalities
regionally and examine voxel-wise correlations of emphysema and fSAD with quantitative perfusion MRI.
Methods or Background: Sixty-three patients from the multi-centre COSYCONET cohort underwent same-day low-dose paired
inspiratory-expiratory CT as well as morphofunctional MRI. Lungs were segmented CT and on T1-weighted VIBE MRI using an in-house
pipeline, separately. The segmentation was forwarded to an iterative deformable registration model, where both CT and functional 4D
perfusion MR are registered to the spatial layout of VIBE MR. The correlation between quantitative CT-based parametric response
maps (PRMemphysma, PRMfSAD) and MRI-based pulmonary blood flow (PBF) and perfusion defects (QDP) was studied on a voxel
level. For comparing interpatient PBF, we weighted each voxel according to its relative proportion to the patient’s lung.
Results or Findings: Lung voxels labeled as PRMEmphysema, PRMfSAD, and PRMNormal in CT were classified as perfusion defects
in 74.59%, 57.63%, and 30.98%, of cases respectively. Dice coefficient equals 0.681 beween PRMNormal and well perfused regions.
The normalised PBF was 45.8, 56.9, and 85.6 ml/100ml/min for PRMEmphysema, PRMfSAD, and PRMNormal voxels respectively.
Conclusion: We proposed a novel registration pipeline to spatially align CT, T1-weighted MRI and 4D perfusion MR. Results suggest
CT-based PRMEmphysema and PRMfSAD labels are strongly correlated to low PBF, and had a higher tendency to be classified as
perfusion defects. Perfusion information is supplementary to PRM, requiring further investigation.
Limitations: The method is yet to be evaluated on a larger subgroup of the cohort.
Ethics committee - additional information: The study was approved by the Ethik-Kommission der Medizinischen Fakultät der
Universität Heidelberg, S-656/2012.
Radiomic features changes of lung tissue in 1024 matrix CT images before the visualisation of pulmonary metastases:
a preliminary study (7 min)
Hua Ren; Shanghai / China
1 2 1 1 2
Author Block: H. Ren , S. Dong , H. Yu ; Shanghai/CN, Beijing/CN
Purpose: The purpose of this study was to investigate whether there are microscopical and invisible changes in lung CT images
before the detection of pulmonary metastases.
Methods or Background: Sixteen patients with pulmonary metastases were followed up for more than three months between
January 2019 and June 2023. Radiomic features were extracted from high-resolution CT images before and after the visualisation of
metastases. Elastic registration was applied to align images after metastasis with those before metastasis. Lung metastases were
delineated using 3D-slicer, and contralateral lung tissue was outlined as contrast. Radiomic analysis was performed on the pre-
metastasis images using features including first-order, grey-level co-occurrence matrix (glcm), grey-level dependence matrix (gldm),
grey-level run length matrix (glrlm), grey-level size zone matrix (glszm), neighbourhood grey-tone difference matrix (ngtdm), high-
order filter features, and wavelet-based features. The signatures of pre- and post-metastasis images were analysed.
Results or Findings: A total of 58 lesions met the criteria (29 metastases, 29 controls), yielding 1209 radiomic features. Wilcoxon
tests identified 362 radiomic features with significant differences between pre- and post-metastasis images (P<0.05). These features
were used to build a K-neighbours Classifier model. The results showed an accuracy of 0.775, sensitivity of 0.65, and specificity of
0.733 in predicting metastasis occurrence before lesion visualisation. Logistic regression analysis indicated an AUC value of 0.8.
Conclusion: Lung tissue may undergo subtle changes before the visualisation of pulmonary metastases. Radiomic features might
reflect these changes earlier, but further validation is required with a larger dataset.
Limitations: Due to this study's retrospective single-centre nature, some selection bias may be unavoidable. Furthermore, all CT
scans used in our study were obtained and provided by only one company.
Funding for this study: The study was funded by the National Natural Science Foundation of China (grant number 82071873), the
Clinical Research Special Project of Shanghai Municipal Health Commission (grant number 20234Y0020), and the China International
Medical Foundation (grant number z-2014-07-2301).
Ethics committee - additional information: The study was approved by the Institutional Ethics Committee of Shanghai Chest
Hospital, Shanghai Jiao Tong University School of Medicine. Individual consent for this retrospective analysis was waived.
Lung, bronchi, and bronchus-artery ratios normative data on chest CT scans: a study from birth into adulthood (7 min)
Qianting Lv; Rotterdam / Netherlands
312
Author Block: Q. Lv, Y. Chen, D. Caudri, E-R. Andrinopoulou, J-P. Charbonnier, M. De Bruijne, H. A. W. M. Tiddens, P. Ciet;
Rotterdam/NL
Purpose: The objectives of this study were to characterise the developmental trends of quantitative parameters obtained from chest
computed tomography (CT) and to provide normative data on dimensions of lung and bronchi, as well as bronchus-artery (BA) ratios
from birth to adulthood.
Methods or Background: We collected 1160 chest CT scans of participants with an age ranging from 0 to 24 years. Using an
automated deep learning-based algorithm, bronchus and artery parameters were computed (bronchial outer diameter (Bout),
bronchial inner diameter (Bin), bronchial adjacent artery diameter (A), bronchial wall thickness (Bwt), bronchial wall area (BWA), and
bronchial outer area (BOA)). From these parameters the following ratios were computed: Bout/A, Bin/A, Bwt/A, and BWA/BOA.
Furthermore, the square root of wall area of a 10mm lumen perimeter (Pi10) was obtained. The effects on CT parameters of age, sex,
and iodine contrast were investigated using mixed-effects or regression model analyses.
Results or Findings: Normal inspiratory chest CT scans of 156 females and 219 males (mean age (SD) 12.7 (5.0) years) were
analysed. Bout and Bin progressively increased with age (all p<0.05), but Bwt, Bout/A, Bin/A, Bwt/A, Bwt/Bout, BWA/BOA, or Pi10 did
not. Bout was the only parameter that differed between males and females, being higher in males than females (p<0.01).
Conclusion: Quantitative CT parameters of bronchi exhibit growth-related changes, but ratios between bronchus and artery
dimensions remain constant. We suggest using age-dependent normative values for bronchial widening and bronchial wall thickening.
Limitations: The study is limited as it is mixed-case.
Funding for this study: Funding was received from the Nederlandse Cystic Fibrosis Stichting (NCFS)- Health Holland (PPS).
Ethics committee - additional information: This is a mixed-case study.
Associations of quantitative HRCT-derived scores of interstitial lung disease (ILD) extent with two-year transplant-free
survival in patients with progressive fibrosing ILD in the ILD-PRO Registry (7 min)
Grace Kim; Porter Ranch / United States
Author Block: G. Kim1, A. Swaminathan2, T. Whelan2, M. Neely2, J. Todd2, S. Palmer2, D. Wojdyla2, C. Conoscent3, J. Goldin1; 1Los
Angeles, CA/US, 2Durham, NC/US, 3Ridgefield, CT/US
Purpose: Associations between quantitative measures of fibrosis on HRCT and survival in patients with progressive fibrosing ILDs are
not well established. The purpose of this study was to evaluate the prognostic value of HRCT-derived scores in the ILD-PRO Registry.
Methods or Background: Patients had an ILD other than IPF, reticular abnormality and traction bronchiectasis, and met criteria for
ILD progression within the prior 24 months. HRCT images taken closest to enrolment were analysed following lobar segmentation. A
machine-learning algorithm derived the following quantitative scores: quantitative lung fibrosis (QLF); quantitative ground glass
(QGG); quantitative honeycomb (QHC); quantitative ILD (QILD: sum of QLF, QGG and QHC). Cox proportional hazards models for time
to death or lung transplant were fit and Kaplan-Meier event rates reported.
Results or Findings: Among 331 patients, median QLF, QGG, QHC and QILD scores were 14.8%, 22.7%, 0.07% and 42.3%,
respectively. Event rates for death or lung transplant at 2 years in the highest vs lowest tertiles of QLF, QGG, QHC and QILD scores
were 31.2% vs 19.9%, 30.4% vs 26.3%, 22.8% vs 19.4%, and 31.7% vs 17.3%, respectively. There were no significant differences in
the risk of death or lung transplant across tertiles of any of the scores. The HRs (95% CI) for risk of death or lung transplant for the
highest vs lowest tertiles of QLF, QGG, QHC and QILD scores were 2.20 (1.04, 4.66), 1.72 (0.88, 3.37), 0.92 (0.45, 1.90) and 2.38
(1.09, 5.20), respectively.
Conclusion: Patients with progressive fibrosing ILDs in the ILD-PRO Registry who had QLF or QILD scores in the highest tertiles
showed trends towards an increased risk of death or lung transplant over 2 years compared with those with scores in the lowest
tertiles.
Limitations: The study is limited by only evaluating associations between quantitative HRCT scores and short-term disease
progression in patients with progressive pulmonary fibrosis.
Funding for this study: The IPF-PRO/ILD-PRO Registry is supported by Boehringer Ingelheim Pharmaceuticals, Inc (BIPI) and run in
collaboration with the Duke Clinical Research Institute (DCRI) and enrolling centers.
Ethics committee - additional information: The study was approved by the Institutional Review Board of each participating
centre: ClinicalTrials.gov Identifier: NCT01915511.
QCT-PFT image quality quantification standard establishment and result calibration (7 min)
Ti Anhe Ye; Wuhan / China
313
Author Block: T. A. Ye; Wuhan/CN
Purpose: The purpose of this study was to assess the respiratory effort during quantitative CT pulmonary function testing (QCT-PFT)
and the small airway pathology (SAD) diagnostic accuracy at different image qualities.
Methods or Background: Sixty-seven patients who underwent QCT-PFT and pulmonary function tests (PFT) simultaneously were
retrospectively collected. We designed an objective quantitative parameter to determine the image quality: the degree of respiratory
effort (DRE) = difference in lung volume between inspiratory and expiratory scans/ forced vital capacity. Patients were divided into
three equal groups based on percentiles of DRE: low-, medium-, and high-quality. DRE was used as a correction factor to calibrate the
air-trapping index (ATI), which was used for diagnosing SAD on QCT-PFT. The diagnostic efficacy of SAD evaluated by ATI was
compared in three groups before and after calibration.
Results or Findings: In the high-quality group, the area under the curve (AUC) of ATI for the diagnosis of SAD was 0.86 (95% CI:
0.689-1.025). The AUC was 0.71 (95% CI: 0.482-0.946) in the medium-quality group and 0.53 (95% CI: 0.268-0.794) in the low-quality
group. After ATI calibration, the AUC increased to 0.81 (95% CI: 0.615-1.000) and 0.76 (95% CI: 0.535-0.988) in the medium- and low-
quality groups, respectively.
Conclusion: A reasonable objective parameter was designed to evaluate the image quality of QCT-PFT. The necessary calibration for
medium- or low-quality can significantly improve the efficacy of QCT-PFT in diagnosing SAD and provide effective quality control to
promote QCT-PFT applications further.
Limitations: A primary limitation is that PFT, as the gold standard, is only sometimes accurate, thus affecting the accuracy of DRE.
Secondly, because there was no screening for intrapulmonary lesions, all patients with lung lesions encountered were included in this
study, which may lead to inaccuracies in CT quantitative results and affect the diagnostic efficacy of SAD.
Funding for this study: No funding was received for the study.
Ethics committee - additional information: The study was approved by the Ethics Committee of Union Hospital, Tongji Medical
College, Huazhong University of Science and Technology (No. 0271-01).
Quantitative CT and computational fluid dynamics characterisation for subclinical structural-functional monitoring in
humidifier disinfectant-associated lung injury (7 min)
Changhyun Lee; Seoul / Korea, Republic of
Author Block: J. Choi1, K. J. Chae2, H. Ko2, W. Chung2, G. Y. Jin2, C. Lee2; 1Kansas City, KS/US, 2Seoul/KR
Purpose: The purpose of this study was to characterise subclinical lung structure-function alteration in humidifier disinfectant-
associated lung injury (HDLI) using quantitative computed tomography (qCT) and computational fluid dynamics (CFD) analysis.
Methods or Background: For 103 toxic HD-exposed patients, commercial and in-house software computed 186 multiscale lung
structure-function features through quantitative inspiratory-expiratory CT image matching and one-dimensional CFD breathing
simulations. We first characterised HD-associated structure-function abnormality from 72 adults with normal-appearing CTs and
pulmonary function tests (PFTs), and then reviewed two-year changes in 31 patients in the national HDLI monitoring program of
Korea.
Results or Findings: In HD-affected normal-appearing cases, airway narrowing resulted in a greater tracheobronchial air pressure
drop particularly through the left lower lobe segmental airways (r=0.54, p<0.001). Elevated air trapping, upper and middle lobes
motion, and reduced volume expansion reduced transpulmonary pressure (r=-0.69, 0.53, 0.50; p<0.001, all), while elevated high
attenuation area reduced lower lobe motion (r=-0.50, p<0.001), implying no direct impact of basal lung inflammation and fibrosis on
tidal breathing characteristics. Reduced whole lung motion, reduced right lower lobe ventilation, and elevated left upper lobe
ventilation elevated airway resistance (r=-0.61, -0.62, 0.61; p<0.001, all). Two-year changes of these features characterised
recovering, stable, and worsening subgroups of the 31 monitoring patients with no eminent changes in visual CT features and PFTs.
Conclusion: qCT-CFD analysis provided novel imaging-based regional lung structure-function characterisation of subclinical lung
damage, recovery, and worsening in HDLI. This novel approach may help assess underlying mechanisms of complicated and
worsening lung recovery.
Limitations: The relatively small sample size limits the applicability of the study.
Funding for this study: The study received funding from the Mid-Career Bridging Program through Seoul National University.
Ethics committee - additional information: The study was approved with approval code: 2020-06-037, from the Jeonbuk National
University Hospital.
Clinicoradiological predictors of progressive fibrosing interstitial lung disease (7 min)

Kanghwi Lee; Seoul / Korea, Republic of
314
Author Block: K. Lee, J. H. Lee, J. M. Goo; Seoul/KR
Purpose: The purpose of this study was to investigate the clinicoradiological predictors of progressive fibrosing interstitial lung
disease (PF-ILD), including deep learning-based CT quantification.
Methods or Background: This single-centre retrospective study included ILD patients who had CT examinations with at least a two-
year interval and underwent pulmonary function tests within three months between January 2015 and June 2021. Clinical
characteristics (age, sex, body mass index (BMI), initial percentage of predicted forced vital capacity (FVC), and idiopathic pulmonary
fibrosis (IPF) versus non-IPF) and deep learning-based CT quantification results from initial CT images (total lung volume, ground glass
opacity, reticular opacity, honeycombing, fibrosis extent) were evaluated. Using these features, we performed logistic regression
analyses for predictors of the following three definitions of PF-ILD, respectively: progression assessed by thoracic radiologists and (a)
absolute decline in FVC≥5%, (b) absolute decline in FVC≥10%, or (c) absolute decline in DLCO≥0%. Finally, prognostic factors for
overall mortality were assessed using multivariable Cox regression analysis.
Results or Findings: A total of 468 patients (239 men; 64 +/- 9.5 years; 220 IPF) were included, and 148 (31.6%), 86 (18.4%), and
138 (of 427 patients with DLCO; 32.3%) patients met the (a), (b), and (c) definitions, respectively. After adjusting age, sex, initial
percentage of predicted FVC, and diagnosis of IPF, reticular opacity was a significant predictor of PF-ILD (for definition (a), odds ratio
(OR): 1.069, P=0.008; for (b), OR: 1.068, P=0.021; for (c), OR 1.080, P=0.009), but honeycombing was not. For overall mortality,
reticular opacity (hazard ratio (HR): 1.051, P=0.045), honeycombing (HR: 1.066, P=0.001), and fibrosis extent (HR: 1.06, P<0.001)
were significant prognostic factors.
Conclusion: Deep learning-based quantified reticular opacity at initial CT images was a significant predictor of PF-ILD. Reticular
opacity, honeycombing, and fibrosis extent were independent poor prognostic factors for survival.
Ethics committee - additional information: The study was approved by the Institutional Review Board of Seoul National
University Hospital.
Multiparametric evaluation of radiomics features and dual-energy CT iodine maps for discrimination and outcome
prediction of thymic masses (7 min)
Simon Martin; Frankfurt a. Main / Germany
Author Block: S. Martin, S. Mahmoudi, I. Yel, C. Booz, L. D. Grünewald, J-E. Scholtz, K. Eichler, T. Vogl, V. Koch; Frankfurt a. Main/DE
Purpose: The objective of this study was to investigate the diagnostic value of radiomics features and dual-source dual-energy CT
(DECT) based material decomposition in differentiating low-risk thymomas, high-risk thymomas, and thymic carcinomas.
Methods or Background: This retrospective study included 32 patients (16 males, mean age 66±14 years) with pathologically
confirmed thymic masses who underwent contrast-enhanced DECT between October 2014 and January 2023. Two experienced
readers evaluated all patients regarding conventional radiomics features, as well as DECT-based texture features, including
attenuation (HU), iodine density (mg/ml), and fat fraction (%). Data comparisons were performed using analysis of variance (ANOVA)
and chi-square statistic tests. Receiver operating characteristic (ROC) curve analysis and Cox regression tests were used to
discriminate between low-risk/ high-risk thymomas and thymic carcinomas.
Results or Findings: Of the 32 thymic tumors, 12 (38%) were low-risk thymomas, 11 (34%) were high-risk thymomas, and 9 (28%)
were thymic carcinomas. Values differed significantly between low-risk thymoma, high-risk thymoma, and thymic carcinoma
regarding DECT-based texture features (p≤0.023) and 30 radiomics features (p≤0.037). The area under the curve (AUC) to
differentiate between low-risk/ high-risk thymomas and thymic cancer was 0.998 (95% CI, 0.915–1.000; p<0.001) for the combination
of DECT imaging parameters and radiomics features, yielding a sensitivity of 100% and specificity of 96%. During a follow-up of 60
months (IQR, 35-60 months), the multiparametric approach, including radiomics features, DECT parameters, and clinical parameters,
showed an excellent prognostic power to predict all-cause mortality (c-index = 0.978 (95% CI, 0.958–0.998), p=0.003).
Conclusion: A multiparametric approach including conventional radiomics features and DECT-based texture features facilitates
accurate, non-invasive discrimination between low-risk/ high-risk thymomas and thymic carcinomas.
Ethics committee - additional information: This retrospective study was approved by the Ethical Review Board of the publishing
institution, and written informed consent was waived.
CT quantification of pulmonary function and compensation after lobectomy and segmentectomy in patients with lung
cancer (7 min)
Leqing Chen; Wuhan / China
315
Author Block: L. Chen; Wuhan/CN
Purpose: This retrospective study aimed to compare the effects of video-assisted thoracic surgery (VATS) lobectomy and
segmentectomy on postoperative pulmonary function and compensatory changes in patients undergoing lung cancer surgery.
Methods or Background: A total of 120 patients (82 VATS lobectomy, 38 VATS segmentectomy) were assessed for demographic
characteristics, baseline pulmonary function, tumour volume, T stage, and histological grade. Postoperative pulmonary function and
compensatory changes (percentage of the well-aerated lung to total/unilateral lung volume (WAL%) and non-operated lung (NOL))
were measured at multiple time points up to two years. Logistic regression analysis identified factors associated with WAL% decline
after one year.
Results or Findings: Both VATS lobectomy and segmentectomy led to a decrease in pulmonary function, with no significant
difference in the extent of decline between the two groups (all P>0.05). Lobectomy triggered a more pronounced compensatory
response, characterised by increased ipsilateral NOL volume over time. Segmentectomy induced minimal compensatory changes and
had a minimal impact on pulmonary function. Factors associated with decreased pulmonary ventilation after one year differed
between the lobectomy and segmentectomy groups. In the lobectomy group, a higher preoperative WAL% of ipsilateral NOL (OR:
1.073; 95% CI: 1.017-1.133; P=0.010) was associated with a higher risk of decline in pulmonary function, whereas in the
segmentectomy group, the only influencing factor was the preoperative contralateral mean lung density (MLD) (OR: 0.932; 95% CI:
0.884-0.984; P=0.010).
Conclusion: VATS lobectomy and segmentectomy demonstrate comparable decline in postoperative pulmonary function, with
lobectomy inducing a greater compensatory response, while segmentectomy minimally affects pulmonary function.
Limitations: Limitations identified include the lack of tumour size matching between the two groups. Additionally, the sample size
was relatively small.
Ethics committee - additional information: The study was approved with the approval code: 0516.
316
HW 8Ca - Coronary artery disease and myocardial ischaemia: imaging, diagnosis, and reporting
Categories: Cardiac, Imaging Methods, Vascular

CME Credits: 1.5
Moderator:
James Shambrook; Winchester / United Kingdom


Marco Rengo; Roma / Italy
1. To become familiar with typical and atypical imaging findings of myocardial ischaemia.
2. To become familiar with clinical data and invasive coronary angiography.
3. To learn about the limits and technical drawbacks of cardiac CT and MRI.
4. To learn how to report cardiac CT and MRI using specific templates.
317
PC 8 - EU-JUST CT – European co-ordinated action on improving justification of computed tomography,

an ESR lead project on behalf of the European Commission: project outcomes and implications
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Professional Issues

CME Credits: 1.5
Moderator:

Methodology for the EU-JUST-CT project (16 min)

Alexandra Karoussou-Schreiner; Luxembourg / Luxembourg
1. To learn about the common methodology and tools for carrying out the coordinated national/regional audits of justification of CT
examinations.
2. To appreciate the legal background of the project.
3. To understand the role of the stakeholders involved in the audits.
Practical issues in carrying out co-ordinated pilot audits of CT examinations (16 min)
1. To learn about the practical issues related to pilot audits.

2. To learn about methodological dilemmas in the analysis process.
3. To describe solutions for the analysis of data from different countries.
Results of the audit of CT referrals in seven EU member states (16 min)

1. To learn about the results of the audit of CT referrals in seven studied regions and countries.
2. To appreciate differences in the practice of justification of CT examinations among countries.
3. To understand the consequences of inappropriate referrals on radiology practice and health care in general.
Guidance to assist radiology departments in improving justification (16 min)

Steve Ebdon-Jackson; Reading / United Kingdom
1. To learn how the justification of medical exposures is addressed in European Directives and national legislation.
2. To appreciate the roles and responsibilities of those involved in the justification process and the challenges they face.
3. To understand how to improve the justification of individual CT examinations within imaging departments.
Panel discussion: Outcomes and implications of EU-JUST CT project – can the results influence the practice of the
justification of CT examinations in Europe? (20 min)
318
RPS 802 - Applications of contrast-enhanced mammography (CEM)
Categories: Breast, Contrast Media, Imaging Methods

CME Credits: 1.5
Moderator:
Eva Maria M. Fallenberg; München / Germany
A large-scale single-centre evaluation of enhanced mammography (CEM) in the pre-operative staging of breast cancer
(7 min)
Francesca Pugliese; Florence / Italy
Author Block: F. Pugliese, G. Migliaro, G. Bicchierai, C. Bellini, D. De Benedetto, F. Di Naro, L. A. Incardona, V. Miele, J. Nori;
Florence/IT
Purpose: The purpose of this study was to evaluate the role and efficacy of contrast enhanced mammography in the pre-operative
management of patients with biopsy-proven breast cancer.
Methods or Background: We selected 524 patients with breast cancer who underwent CEM as preoperative staging and had breast
cancer-related surgery at our institution, between 2017 and 2020. We analysed those cases in which CEM led to additional imaging or
biopsy and those in which it changed the type of surgery. CEM sensitivity in identifying the index lesion and sensitivity, specificity,
positive (PPV) and negative (NPV) predictive values, and accuracy in the correct preoperative staging of breast cancer of the whole
population and in various subgroups were calculated.
Results or Findings: CEM changed the surgical plan in 22.48% of cases (118/525 patients) when the extension of the index lesion
was different from what was previously detected by conventional imaging or when an additional lesion was detected. The sensitivity,
specificity, PPV and NPV, and accuracy of CEM were 94.74%, 95.86%, 86.40%, 98.5%, diagnostic accuracy was 95.62% and AUC of
ROC curve was 0.953. Conclusively, after using a chi-square test, the statistical analysis showed a higher diagnostic performance in
patients with palpable lesions vs patients with non-palpable ones (p=0.0022).
Conclusion: CEM changed the surgical plan in 22.48% of our series, proving to be an accurate exam in the presurgical staging of
breast cancer. CEM diagnostic performances support its reliability in the presurgical staging setting, for a lesion-tailored management,
leading to the best surgical plan, avoiding over- or undertreatment.
Ethics committee - additional information: This study is retrospective.
Outcomes of surveillance using contrast-enhanced mammography in women with a personal history of breast cancer (7
min)
Bruce Mann; Parkville / Australia
319
Author Block: J. Matheson1, A. Rose1, C. Nickson2, K. Elder1, B. Mann1; 1Parkville, VIC/AU, 2Sydney, NSW/AU
Purpose: Annual mammography with selective ultrasound is standard for women with Personal History of Breast Cancer (PHBC).
Rates of interval cancers of 3.6 per 1000 screens have been reported. Earlier diagnosis of recurrences and new cancers may allow
more effective treatment. This study aimed to clarify the utility of contrast-enhanced mammography (CEM), a more sensitive scan
than conventional mammography and comparable to MRI, for surveillance.
Methods or Background: This was a retrospective study of 1,190 women with PHBC undergoing CEM surveillance between June
2016 and December 2022. Outcomes of initial CEM and subsequent imaging were collected, including incident surveillance-detected
and interval cancers, recalls for assessment, and pathology and treatment details. Descriptive statistics and hazards modelling was
used.
Results or Findings: There were 3,784 surveillance episodes: 1,190 first CEMs and 2,594 subsequent surveillance episodes. 79% of
women had at least three annual rounds of surveillance. 186/3784 surveillance episodes were recalled for assessment (recall rate
4.9%). 72 (39%) recalled cases were true positive- 50 invasive and 22 DCIS. The cancer detection rate was 19/1000. 51% of recalls
were due to contrast and 35% of these were true positive. Invasive cancers were predominantly stage 1 (64%) or stage 2 (32%) and
most were grade 2 (44%) or grade 3 (47%). The median invasive cancer size was 16mm (IQR 9-25mm). The rate of symptomatic
interval cancers was 0.8 per 1000 screens (program sensitivity 96.0%). Surveillance-detected cancer rates differed significantly by
index cancer subtype (χ2=11.9, p=0.0026), with the highest rates for women with TNBC.
Conclusion: Routine CEM in surveillance of women with PHBC led to higher cancer detection and lower interval cancer rates than
published series. CEM appears to increase the sensitivity of surveillance programs for women with PHBC. This may allow more
effective treatment.
Limitations: This was a single institution study.
Funding for this study: Funding was received by an unrestricted grant from Hologic.
Ethics committee - additional information: The study was approved by the Melbourne Health Research and Ethics Committee,
QA2019129.
Contrast-enhanced mammography in local staging of screen-detected breast cancer (7 min)

Allison Rose; Fitzroy / Australia
Author Block: C. MacCallum1, B. Mann1, C. Nickson2, A. Rose1; 1Parkville, VIC/AU, 2Carlton, VIC/AU
Purpose: BreastScreen Australia provides assessment of suspected breast cancer to the point of diagnosis using conventional
imaging and percutaneous biopsy. Comprehensive local staging is not performed. The purpose of this study was to introduce contrast-
enhanced mammography (CEM) for local staging of screen-detected cancer at our academic hospital. We report findings for otherwise
occult disease and impact on treatment.
Methods or Background: Women with screen-detected cancer undergoing staging CEM between November 2018 and April 2022
were identified retrospectively. Additional contrast-detected abnormalities were investigated with preoperative percutaneous biopsy,
surgical biopsy, or problem-solving MRI. Findings were recorded. Invasive cancer or DCIS were considered as true positives (TP) and
other findings as false positives (FP). Impact on surgical decisions was assessed.
Results or Findings: 204 patients underwent CEM, of whom 62/204 (30%) had 76 additional abnormalities. 36/76 (47%) were TP
and 40/76 (53%) FP. CEM identified malignant occult lesions (mOLs) in 30/204 patients (15%). 75% moLs were invasive and 25%
DCIS. 83% (30/36) of mOLs were ipsilateral. Most invasive mOLs were Grade 2 (20/27, 74.1%). Patients with higher background
parenchymal enhancement (BPE) had a higher rate of mOLs (20% for moderate/marked vs 4% for minimal/mild, p=0.0023). No
statistically significant differences were found by mammographic density or age. Surgical management was changed in 45/204 (22%)
patients, 30 with mOLs and 15 with benign and atypical proliferative lesions. Altered surgery included wider resection (24/45),
conversion to mastectomy (8/45), contralateral breast surgery (6/45), additional ipsilateral excision (5/45), and bracketing (2/45).
Conclusion: CEM for local staging of screen-detected cancers identified additional malignant lesions in 15% of patients (particularly
those with higher BPE), with many clinically significant lesions. Findings led to surgical management change in 22% of patients.
Larger studies with oncological outcomes are needed.
Limitations: This was a single arm study. Larger numbers and oncological outcomes are needed to overcome the limitations of this
study.
Funding for this study: Funding was received by a research grant supported by Hologic.
Ethics committee - additional information: This study was approved by a Quality Assurance project by the Melbourne Health
Research and Ethics Committee.
Comparison of the diagnostic accuracy of contrast-enhanced spectral mammography with delayed image and digital
mammography in the diagnosis of breast cancer in BI-RADS 4 mammographic findings (7 min)
Akmaral Ainakulova; Almaty / Kazakhstan
320
Author Block: A. Ainakulova, Z. J. Zholdybay, D. Kaidarova, Z. M. Amankulov, M. Gabdulina; Almaty/KZ
Purpose: The purpose of this study is to compare the diagnostic accuracy of contrast-enhanced spectral mammography (CESM) with
delayed image and digital mammography (DM) in breast cancer detection in BI-RADS 4 mammographic findings among mass lesions,
suspicious microcalcifications, asymmetries, and architectural distortions.
Methods or Background: Between September 2018 and May 2021, mammographic examinations of 1968 women were carried out
according to clinical indicators. According to the results of mammographic examination, BI-RADS 4 mammographic findings were
revealed in 374 (19%) patients, requiring differential and clarifying diagnostics. All 374 patients underwent CESM with delayed image.
All lesions were histologically verified. Lesions were categorised based on ACR BI-RADS Fifth Edition. On CESM with delayed image,
lesions were categorised by dynamic patterns of contrast enhancement.
Results or Findings: Of the 398 lesions identified, 208 (52.26%) were malignant and 190 (47.73%) were benign. Mass lesions were
detected in 240 (64.2%) women, suspicious microcalcifications in 60 (16%) women, architectural distortions in 31 (8.3%) women, and
asymmetry in 43 (11.5%) women. For DM and CESM with delayed image among mass lesions sensitivity was 82.1% and 95.5%;
specificity was 51.4% and 89.2% respectively. For DM and CESM with delayed image among suspicious microcalcifications,
asymmetries, and architectural distortions, sensitivity was 100%, whereas specificity was 82% and 97.4%; 65% and 96.3%; 57.1%
and 94.1% respectively. The ROC curve of all BI-RADS 4 mammographic findings had an area under the curve of 0.965 for CESM with
delayed image, and of 0.749 for DM (p=0.000).
Conclusion: CESM with delayed image showed high sensitivity and specificity in the diagnosis of breast cancer among mass lesions.
CESM with delayed image significantly increased specificity in the diagnosis of breast cancer among microcalcifications, asymmetries,
and architectural distortions, and may be helpful to avoid unnecessary biopsies in BI-RADS 4 mammographic findings.
Limitations: The study used a small number of patients with suspicious microcalcifications, asymmetries, and architectural
distortions.
Diagnostic performance of lesion conspicuity and contrast-to-noise ratio (CNR) values on contrast-enhanced-
mammogram (CEM) recombined images in predicting malignancy (7 min)
Iris Allajbeu; Cambridge / United Kingdom
Author Block: I. Allajbeu, R. Manavaki, M. Nanaa, N. Healy, V. Papalouka, N. Payne, F. Gilbert; Cambridge/UK
Purpose: The purpose of this study was to compare the diagnostic performance of lesion conspicuity and contrast-to-noise ratio
(CNR) on CEM recombined images in differentiating benign and malignant lesions.
Methods or Background: CEM examinations with BI-RADS scores of 3-5 and histopathologically confirmed findings were eligible for
inclusion. Two breast radiologists assessed lesion conspicuity as low, moderate, or high, based on the BIRADS CEM lexicon. Two
regions-of-interest (ROI) representing lesion and background, respectively, were delineated on the early (CC) and late (MLO) views to
calculate CNR1 and CNR2, respectively, using CNR=(Sa-Sb)/Sb (Sa: maximum value in the lesion ROI, Sb: mean value in the
background ROI). Associations between nominal, or continuous and nominal variables were assessed using chi-squared test or rank-
biserial correlation coefficient (rrb) respectively. Receiver operating characteristic (ROC) curves and area-under-the curve (AUC) were
used to assess diagnostic performance. p values < 0.05 were deemed statistically significant.
Results or Findings: Of 143 lesions, 85 were normal/ benign and 58 malignant confirmed by histopathology. All high conspicuity
lesions were invasive cancers (9/9, 100%) with 82% (49/60) low conspicuity lesions being normal/ benign, and 78% (45/58) malignant
lesions showing either moderate or high conspicuity (χ2=60.1, p<.001). Malignancy was associated with increased CNR1 (rrb=0.65,
p<.001) and CNR2 (rrb=0.41, p<.001) values. CNR1 performed better in differentiating benign from malignant lesions (AUC 0.75,
accuracy: 71%, sensitivity: 80%, specificity 63%) than CNR2 (AUC 0.71, accuracy: 68%, sensitivity: 71%, specificity 65%) or lesion
conspicuity (AUC: 0.75%, accuracy: 71%, sensitivity: 80%, specificity: 64%).
Conclusion: Both conspicuity and CNR quantitative assessment of enhancing lesions on CEM can improve specificity in
differentiating benign from malignant breast disease. Quantitative measurement of lesion enhancement performs better than
conspicuity in predicting malignancy, with CNR calculated on the early view offering the best diagnostic performance.
Limitations: This was a single centre retrospective study with a limited dataset.
Funding for this study: Funding was received as part of the BRAID trial.
Ethics committee - additional information: The study was approved by the BRAID trial (IRAS project ID: 25137).
Feasibility and performance of contrast-enhanced tomosynthesis compared to contrast-enhanced MRI (7 min)

321
Author Block: P. Clauser1, N. Pötsch1, P. Kapetas1, M. Hörnig2, M. Weber1, R-I. Milos1, P. A. Baltzer1, T. H. Helbich1; 1Vienna/AT,
2
Forchheim/DE
Purpose: The performance of contrast-enhanced mammography (CEM) is limited by the 2D image acquisition. We hypothesise that
contrast-enhanced tomosynthesis (CE-DBT) could allow for improved lesion characterisation, truly comparable to contrast-enhanced
magnetic resonance imaging (CE-MRI). The aim of our analysis was to compare the diagnostic performance of a CE-DBT prototype
with CE-MRI.
Methods or Background: Women presenting with suspicious findings on mammography, DBT, or ultrasound were invited to
participate in the study. Participants underwent CE-DBT of the breast with suspicious findings using a dedicated prototype. The
suspicious findings were biopsied and only histologically verified lesions were included in the analysis. Four readers (R1 and R2 non-
experienced; R3 and R4 experienced) evaluated the images, blinded to patients’ history, previous imaging, and histology. The readers
evaluated CE-MRI and CE-DBT (including DBT and synthetic mammography) in separate sessions and gave a BI-RADS score for each
finding. Sensitivity and specificity were calculated and compared.
Results or Findings: We included 84 patients (mean age 56 years, range 39-70) with 91 histologically verified breast lesions (27
benign, 64 malignant). Sensitivity was comparable between CE-DBT and CE-MRI for R3 (both 100%), while sensitivity was higher with
CE-MRI than with CE-DBT for the other readers (CE-DBT raging from 89 to 94%, CE-MRI from 97 to 100%). Specificity was overall
higher for experienced than for non-experienced readers. Specificity improved with CE-MRI for non-experienced readers, while no
significant difference was detected between CE-DBT and CE-MRI for experienced readers.
Conclusion: Our study showed that CE-DBT has a good diagnostic performance for the characterisation of breast lesions, almost
comparable to that of CE-MRI of the breast.
Limitations: This was a single centre, cancer enriched dataset.
Funding for this study: Funding for this study was received from Siemens Healthcare GmBH.
Ethics committee - additional information: The study was approved by the Ethics Committee and patients gave written informed
consent.
Preoperative contrast mammography vs MRI: which one is better? (7 min)

Carmen Mbongo; Pamplona / Spain
Author Block: C. Mbongo, M. B. Barrio Piqueras, P. Malmierca Ordoqui, M. Jiménez Vázquez, A. Elizalde, L. J. Pina Insausti;
Pamplona/ES
Purpose: The aim of this study was to compare the sensitivity and accuracy of Contrast Mammography (CEDEM) vs MRI as
preoperative imaging techniques.
Methods or Background: This is a retrospective lesion-based study. From January 2018 to December 2019, 77 patients underwent
CEDEM, MRI and percutaneous biopsy at our centre as a part of the breast lesion preoperative protocol. CEDEM and MRI images were
then reviewed, and BI-RADS categorisation for each lesion was subsequently collected. The final histopathologic report was also
assessed and considered the gold standard. BI-RADS categories 1 to 3 were considered negative, while 4 and 5 were classified as
positive. A McNemar test was used to compare the sensitivity of both imaging techniques, and ROC curves were performed to assess
the areas under the curve, both using SPSS statistical software.
Results or Findings: The 77 female patients (37-84, mean 55.5 y/o) showed 98 lesions with pathologic correlation (16 benign; 82
malignant: 77 invasive ca, of which 43 were LumA, 22 LumB, 6 pure Her2, 6 Triple negative, and finally 5 DCIS). The sensitivity of
CEDEM vs MRI was 83.3% vs 97.6% (p=0.002). The areas under the curve were 0.87 vs 0.86 (p=0.9), respectively.
Conclusion: Although MRI is significantly more sensitive than CEDEM, the accuracy was similar for both modalities.
Limitations: We had a low number of cases; it was a monocentric study, and no intra/interobserver variability was studied.
Tumour involvement of the nipple at preoperative contrast-enhanced mammography (CEM): preliminary results of a
multicentre diagnostic accuracy study (7 min)
Thiemo van Nijnatten; Maastricht / Netherlands
322
Author Block: T. van Nijnatten1, A. Cozzi2, R. Alcantara3, G. Della Pepa4, J. James5, C. Depretto4, G. P. Scaperrotta4, S. Schiaffino2;
1
Maastricht/NL, 2Lugano/CH, 3Barcelona/ES, 4Milan/IT, 5Nottingham/UK
Purpose: The aim of this study was to assess the diagnostic accuracy of contrast-enhanced mammography (CEM) in the prediction of
pathologic nipple involvement, CEM diagnostic accuracy was also compared with contrast-enhanced breast MRI scans.
Methods or Background: This retrospective multicentre study included patients with biopsy-proven breast cancer who underwent
preoperative CEM in three referral centres (Italy, Spain, the Netherlands), between April 2013 and June 2023. Patients were included if
they did not undergo neoadjuvant therapy and if they underwent either mastectomy or central lumpectomy. After local image review
by board-certified breast radiologists (5-to-10 years of experience), the diagnostic performance of CEM in predicting nipple
involvement was calculated, taking surgical pathology as the reference standard and compared with MRI (McNemar’s test) in a subset
of patients.
Results or Findings: A total 70 patients (median age 64 years, interquartile range 51–73 years) with breast cancer (50/70 invasive
ductal carcinoma, 16/70 invasive lobular carcinoma, 4/70 pure DCIS) were included in the analysis, 59/70 also having undergone MRI.
Nipple involvement at surgical pathology was found in 25/70 cases (35.7%). Considering all 70 patients, CEM had a 72.0% sensitivity
(95% CI 50.6–87.9%) and an 88.9% specificity (95% CI 76.0–96.3%) for the prediction of pathological nipple involvement. All seven
false negative cases presented as non-mass enhancement lesions without peri-areolar skin thickening. Among the 59 patients with
both CEM and MRI, CEM had a 65.0% sensitivity (95% CI 44.1–85.9%), significantly lower (p=0.014) than the 95.0% sensitivity of MRI
(95% CI 85.4–100%), whereas no significant differences (p=0.179) were found between CEM specificity (97.4%, 95% CI 92.5–100%)
and MRI specificity (89.7%, 95% CI 80.2–99.3%).
Conclusion: Preoperative CEM showed moderate sensitivity and high specificity in assessing nipple involvement. Compared to MRI,
CEM had significantly lower sensitivity and non-significantly higher specificity.
Limitations: The study is limited by its reliance on preliminary data, as well as the exclusion of NAT patients.
Ethics committee - additional information: The study is a pooled analysis of data from other studies, which received approval
from Ethics Committees in three different centres (Maastricht University Medical Centre +, the Netherlands; Hospital Del Mar,
Barcelona, Spain; and the Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy).
Nipple sparing mastectomy: can CEM predict the nipple–areola complex (NAC) involvement? (7 min)
Giuliano Migliaro; Florence / Italy
Author Block: G. Migliaro, C. Bellini, G. Bicchierai, F. Di Naro, D. De Benedetto, F. Pugliese, L. A. Incardona, V. Miele, J. Nori;
Florence/IT
Purpose: The purpose of this study was to assess the role of Contrast-Enhanced Mammography (CEM) as a predictor of invasion of
the NAC in patients undergoing nipple-sparing mastectomy (NSM).
Methods or Background: In our retrospective study, three radiologists reviewed 1479 presurgical CEMs of patients who underwent
NSM for biopsy proven malignant lesions, conducted at our centre between 2018 and 2019, in consensus. We included 64 CEMs of 64
patients; we evaluated the TND (tumour-nipple distance) dividing patients according to the shortest TND seen either in CC or in MLO.
The correct extension of the disease, the enhancement type, and the enhancing conspicuity were described for every lesion.
Sensitivity, specificity, negative, and positive predictive values of CEM for nipple invasion (PPV and NPV) were calculated. Our gold
standard was the final histology of the surgical specimen.
Results or Findings: The patients were beween 38 and 84 years old (mean age 56.9, median age 52). The shortest TND between
the two projections varied from 0mm to 75mm. Among the lesions with TND≤10mm (25/64), 19/25 (76%) resulted as invading the
NAC while 6/25 did not; three had a low enhancement conspicuity, and 22 had a moderate or high conspicuity. Amid the lesions with
TND>10mm (39/64), only 7/39 (18%) invaded the NAC. Conversely, 32/39 did not; five had a low enhancement conspicuity and 34
had a moderate or high conspicuity. Comparing the two groups of lesions, satisfying NPV (82.1%) and specificity (84.2%) emerged
and acceptable VPP and sensitivity (76% and 73%) were reported. No statistical significance emerged.
Conclusion: CEM is a useful tool that can help predict the NAC invasion in pre-surgical assessments, excluding when it is not
infiltrated due to the high NPV and specificity of this series.
Limitations: The study is limited by the small number of patients included.
Ethics committee - additional information: This is a retrospective study.
Is late-phase contrast-enhanced mammography after neoadjuvant therapy useful? (7 min)

Dilşah Oral; Izmir / Turkey
323
Author Block: D. Oral, Ç. R. Açar, A. B. Kale, O. Acar, I. Ş. S. Örgüç; Manisa/TR
Purpose: This study aims to evaluate the need for late-phase contrast-enhanced mammography (CEM) in assessing post-treatment
residual tumor in breast cancer patients undergoing neoadjuvant therapy (NAT).
Methods or Background: In this prospective study, 61 patients were included. CEM examinations, comprising both early and late-
phase mediolateral oblique images, were conducted before and after NAT. During the evaluation of CEM images, we separately noted
the level of enhancement and the size of index malignant lesions for both phases. Findings were compared with post-surgical
pathology reports.
Results or Findings: Residual lesions were detected in 40 (65.57%) patients in pathology. The sensitivity of both early and late-
phase images in determining complete response was 90.47%. While the specificity was 55% in the early phase, it increased to 65% in
the late phase. Area under curve was 0.727 and 0.777 for early and late-phase images, respectively in ROC analysis. In the correlation
analysis for residual lesion size, a significant correlation (p<0.01) was observed between pathological results and both images.
Notably, late-phase images exhibited a high degree of correlation, with an R-value of 0.887 compared to 0.867 for the early-phase
images. Among 23 patients, when comparing late-phase images to early-phase images, higher contrast enhancement levels were
observed in four (17%) lesions, an increase in size in twelve (52%) lesions, and both phenomena in seven (30%) lesions. Among the
19 lesions that exhibited larger residual lesion sizes in late-phase images, 18 of them were more closely aligned with pathological
results and one was in complete response in pathological reports.
Conclusion: In our study, late-phase images provided better results in determining the residual tumor compared to the early phase
in 29.50% of the patients.
Ethics committee - additional information: This study was approved by the Clinical Research Ethics Committee of Manisa Celal
Bayar University (Decision Number : E-85252386-050.04.04-200753).
Inter-reader agreement between breast imagers using the first version of the BI-RADS CEM lexicon (7 min)
Calogero Zarcaro; Palermo / Italy
Author Block: C. Zarcaro1, A. Santonocito2, L. Zeitouni3, F. Ferrara4, P. Kapetas5, R-I. Milos5, T. H. Helbich5, P. A. Baltzer5, P. Clauser5;
1
Palermo/IT, 2Turin/IT, 3Riyadh/SA, 4Rome/IT, 5Vienna/AT
Purpose: The aim of this study was to evaluate the inter-reader agreement of the Contrast Enhanced Mammography (CEM) lexicon, a
new addition to the 5th edition of the BI-RADS Mammography lexicon.
Methods or Background: Three breast imaging fellows reviewed 462 lesions in 421 routine clinical breast CEM in a IRB-approved,
single-center, retrospective study using the first version of the CEM BI-RADS lexicon and the fifth edition of the BI-RADS
Mammography lexicon. Blinded to patient outcomes, the readers assessed breast and lesion features on low-energy (LE) images (ACR
breast density, type of lesion, associated architectural distortion), lesion features in recombined (RC) images (type of enhancement,
lesion conspicuity, mass shape, mass margin, mass internal pattern of enhancement, non-mass distribution, non-mass internal
pattern of enhancement, enhancing asymmetry), and provided a final BI-RADS assessment. Inter-reader agreement was calculated
for each feature using Fleiss’ and Cohen’s kappa coefficient. Sensitivity and specificity were calculated.
Results or Findings: On both LE and RC images, there was moderate to substantial inter-reader agreement for breast density and
many lesion features, particularly for type of lesion (k=0.654) and type of enhancement (k=0.664). Agreement was moderate to
substantial also on CEM mass enhancement descriptors, while it was fair to moderate for non-mass enhancement and enhancing
asymmetry descriptors. Inter-reader agreement for final LE and CEM BI-RADS assessment was moderate (ĸ=0.421) and fair
(ĸ=0.364). Sensitivity and specificity were high, although specificity was more variable.
Conclusion: On both LE and RC images, there was moderate to substantial inter-reader agreement for the assessment of breast
density and many imaging findings. Agreement was moderate to substantial on CEM mass enhancement, while it was fair to
moderate for non-mass enhancement and enhancing asymmetry descriptors.
Ethics committee - additional information: This study was approved by the relevant Institutional Review Board (IRB).
Preoperative assessment of nipple-areolar complex infiltration: is CEM as reliable as breast MRI? (7 min)
Silvia Sanità; Udine / Italy
324
Author Block: S. Sanità, M. Lorenzon, M. Cirillo, L. Cereser, R. Girometti, C. Zuiani; Udine/IT
Purpose: The purpose of this study was to compare sensitivity and specificity of Contrast Enhanced Mammography (CEM) and breast
Magnetic Resonance Imaging (MRI) in the assessment of the infiltration of nipple-areolar complex (NAC) in pre-operative assessment.
Methods or Background: We retrospectively included all patients over the age of 18 with newly diagnosed, biopsy-proven breast
cancer, who underwent either CEM or 1.5T MRI between January 2022 and December 2022 for preoperative assessment in our centre.
Exclusion criteria were neoadjuvant chemotherapy, and when a pathological analysis of the surgical specimen was not available.
Enhancing NAC were considered infiltrated with both techniques. We calculated and compared CEM and MRI sensitivity, specificity,
PPV, and NPV for the infiltration of NAC, considering the surgical specimen as the standard of reference. The statistical significance
level used was P<0.05.
Results or Findings: After assessment of 168 MRI and 146 CEM, we included 109 MRI and 108 CEM. 23 MRI-patients and 13 CEM-
patients had NAC infiltration at pathological analysis. Contrast enhancement of NAC was detected in 28/109 MRI and 15/108 CEM. The
sensitivity, specificity, PPV, and NPV of MRI versus CEM were 92% (C.I. 74-99%), 94.05% (C.I. 87-98%), 82.14% (C.I. 66-92%), 97.53%
(C.I. 91-99%) versus 92.86% (C.I. 66-99%), 97.87% (C.I. 93-99%), 86.67% (C.I. 62-96%), and 98.92% (C.I. 93-99%) respectively.
Differences among MRI and CEM were not statistically significant with sensitivity Δ=0.860%, p=0.812; specificity Δ=3.820%;
p=0.154; PPV Δ=4.530%; p=0.359, and NPV Δ=1.390%; p=0.439.
Conclusion: CEM seems to be as reliable as MRI in the assessment of infiltration of NAC.
Limitations: The study was monocentric and retrospective, with a limited number of patients.
325
CUBE 8 - Liver/kidney/lung: tips and tricks in difficult cases

in the audience.
Moderator:

Liver/kidney/lung: tips and tricks in difficult cases (28 min)

Osman Öcal; Munich / Germany
1. To focus on microwave and radiofrequency ablation.

2. To learn about factors making a lesion critical in liver / lung / kidney.
3. To learn how to overcome these difficulties.
326
EFRS 2 - Impact of simulation in education

ETC Level: LEVEL I
This session aims to consider what simulation is currently used in education programmes across Europe and to explore what
simulation is used in education and clinical practice.
Moderator:
Alexandra Partner; Leicester / United Kingdom

New options in simulation in diagnostic imaging in Europe (15 min)

Elaine Wilkinson; Bradford / United Kingdom
New options in simulation in radiotherapy in Europe (15 min)

Pete Bridge; Liverpool / United Kingdom
Discussion (20 min)
Closing (5 min)
327
AI-SC 3 - An overview of key milestones in AI regulation

Moderator:

An overview of key milestones in AI regulation (57 min)

Hugh Harvey; Banstead / United Kingdom
1. To learn about the ethics guidelines for trustworthy AI.

2. To learn about the Assessment List for justworthy AI.
3. To learn about the proposed Artificial Intelligence Act.
328
MD 2 - Pancreatic cancer: new developments in imaging and treatment - recommendations for clinical
practice and directions for the future
Categories: Abdominal Viscera, GI Tract, Multidisciplinary, Oncologic Imaging, Research

CME Credits: 1
Moderator:
Riccardo Manfredi; Rome / Italy

1. To highlight new developments in the treatment of pancreatic cancer.

2. To critically review recent developments in imaging and put this in perspective of the treatment developments.
3. To provide recommendations for clinical practice and directions for future research.
The surgeon's perspective (8 min)

Roberto Salvia; Verona / Italy
The oncologist's perspective (8 min)

Michele Milella; Verona / Italy
The radiologist's perspective (8 min)

329
PL 1 - Radiology, AI, and the Analog-Digital Frontier
Categories: Artificial Intelligence & Machine Learning, Professional Issues

CME Credits: 0.5
Moderator:
Radiology, AI, and the Analog-Digital Frontier (28 min)

Rick Abramson; Nashville / United States
330
CUBE 9 - Cryoablation – reloaded

in the audience.
Moderator:

Cryoablation – reloaded (28 min)

Matthias Fuerstner; Klagenfurt / Austria
1. To learn about indications and contraindications for cryoablations.

2. To learn about the advantages and disadvantages for cryoablation compared to heat-based systems.
3. To learn about special indications e.g. palliative treatment of soft tissue tumors.
331
RPS 910 - Musculoskeletal ultrasound and ultrasound-guided intervention
Categories: Imaging Methods, Interventional Radiology, Musculoskeletal

CME Credits: 1.5
Moderator:
High-resolution ultrasound of the postoperative flexor digitorum tendon of the fingers after zone-II reconstructive
surgery (7 min)
Michelle Pansecchi; Genova / Italy
Author Block: M. Pansecchi, F. Zaottini, R. Picasso, F. Pistoia, M. Maccio, S. Rinaldi, D. Bianco, G. Marcenaro, C. Martinoli; Genoa/IT
Purpose: Despite advancements in tendon repair techniques, there are still several complications that can occur after tendon
surgery. These include adhesion formation, tendon repair rupture, and stiffness of the joints. The aim of this study was to describe the
potential role of high-resolution ultrasound (US) to detect postoperative flexor tendon complications after zone-II surgery.
Methods or Background: Twenty-five consecutive patients were submitted to imaging examination for impaired finger
flexion/extension after zone-II reconstructive surgery for a previous tear of the flexor digitorum profundus and/or superficialis tendons.
Dynamic US examination was performed by means of long- and short axis planes using 18-5 and 22-8MHz linear array probes.
Results or Findings: High-resolution US was able to identify suture material and distinguish stitches of variable thickness and
appearance. US imaging over the tendon repair was able to identify focal swelling and prominent knots obstructing tendon gliding.
After FT repair, interphalangeal joint contraction and stiffness was found in eight cases without any tendon gliding abnormality.
Obstructing gliding and adhesions were observed in twelve cases. In 4/12 cases, adhesions were subtle to recognize and could only
be identified on dynamic scans based on traction movements of peritendinous tissues. Tendon repair rupture was observed in n=3
cases as a new tendon discontinuity at the repair site. In one case, bone shortening after P1 fracture and fixation with metallic plate
was diagnosed by US. In another case, the FDP tendon was screwed.
Conclusion: Although difficult to perform and operator-dependent, high-resolution US should be regarded as the modality of choice
to evaluate the postoperative flexor digitorum tendons in the fingers due to its dynamic capabilities and excellent detail resolution.
Limitations: Further studies based on larger series are needed to define the ultimate value of ultrasound in this field.
Ethics committee - additional information: The study was conducted in accordance with the Declaration of Helsinki.
Poland syndrome: the role of ultrasound (7 min)

Michelle Pansecchi; Genova / Italy
332
Author Block: M. Pansecchi, F. Zaottini, R. Picasso, F. Pistoia, M. Maccio, S. Rinaldi, G. Marcenaro, L. Tovt, C. Martinoli; Genoa/IT
Purpose: The study aimed to describe the ultrasound scanning technique and ultrasound scanning in patients affected by Poland
syndrome. Poland syndrome is a rare congenital disease characterized by malformations of the pectoralis major muscle and pectoralis
minor muscle associated with the thoracic wall, hand, and costal ribs malformations.
Methods or Background: We examined the pectoralis region of n=106 consecutive patients affected by Poland syndrome (range
2-53 years old, mean 22 years old). Patients were examined using a 18-4Mhz and 24-5Mhz matrix linear array probes. For each
patient several parameters were evaluated: the anatomical integrity of the three heads of the pectoralis major muscle (clavear,
sternocostal and abdominal), the diameter of the pectoralis major tendon, the anatomical integrity of the pectoralis minor muscle, the
diameter and flowmetry of the subclavian, axillary, and internal mammary artery ipsilateral to the affected side, and the anatomical
integrity of the latissimus dorsi muscle.
Results or Findings: The pectoralis major agenesis was observed in n=21 patients, classified as a type I anomaly. In n=49 patients
the presence of the clavicular head and the absence of the sternocostal and abdominal heads was observed, classified as type II
anomaly. In n=36 patients the presence of the clavicular head and a short sternocostal head was observed and classified as a type III
anomaly. The pectoralis minor was only detected in four patients. No differences were noted in the examinated vessels regarding the
diameter and flow in comparison to the contralateral side.
Conclusion: Ultrasound examination can complement the clinical examination to assess any pectoralis region abnormalities in
patients affected by Poland syndrome.
Ethics committee - additional information: The study was conducted in accordance with the Declaration of Helsinki.
Ultrasound-guided thread-release of the first extensor compartment of the hand: initial experience in the anatomical
model (7 min)
Lisa Lechner; Vienna / Austria
Author Block: L. Lechner, S. A. Jengojan, P. Sorgo, G. Bodner, V. Moser, L. Hirtler; Vienna/AT

Purpose: The objective of this study was to develop and evaluate a minimally-invasive ultrasound-guided thread-release technique
for safely and effectively decompressing the first extensor compartment in anatomical specimens.
Methods or Background: De Quervain’s tenosynovitis is a common disabling condition of the hand, arising due to the compression
of the tendons in the first extensor compartment. If conservative treatments fail, surgical intervention is required. In recent years,
minimally-invasive ultrasound-guided procedures, especially for carpal tunnel syndrome and trigger finger, are on the rise. However,
thus far the thread technique was not studied in the release of the first extensor compartment for possible future treatment of De
Quervain´s tenosynovitis in an anatomical or clinical setting.
For this study, we developed a protocol for the minimally-invasive ultrasound-guided thread-release and performed it prospectively on
ten fresh cadaveric hands. Subsequently, the specimens were dissected and assessed by an experienced anatomist regarding
efficacy and safety of the procedure. Ultrasound visibility, degree of transection, and unnecessary damage to surrounding
neurovascular structures were documented.
Results or Findings: Ultrasound visibility was sufficient in all specimens. Out of ten interventions, a complete release was achieved
in nine cases. Slight, clinically irrelevant, lesions on the underlying extensor tendons were observed in one case. Macroscopically
visible neurovascular structures were not harmed in any case.
Conclusion: Real-time ultrasound-guided thread-release is a reliable and safe minimally-invasive method for releasing the first
extensor compartment in anatomical specimens while protecting the surrounding structures.
Limitations: This study was conducted on anatomical specimens. Thus, results may differ in the clinical setting. Due to cadaver
availability, the initial sample size was limited. Therefore, further anatomical and clinical trials are warranted.
Ethics committee - additional information: This study was approved by the Ethics Committee at the Medical University of Vienna.
Ultrasound-guided thread-transection of the arcade of frohse is an effective and minimally-invasive method for
posterior interosseus nerve release in the anatomical model (7 min)
Lisa Lechner; Vienna / Austria
333
Author Block: L. Lechner, L. Hirtler, P. Sorgo, G. Bodner, V. Moser, S. A. Jengojan; Vienna/AT
Purpose: The purpose of this study was to expand the area of application for effective and safe minimally-invasive ultrasound-guided
thread-interventions to the transection of the Arcade of Frohse.
Methods or Background: Transection of the Arcade of Frohse becomes clinically relevant in the treatment of radial tunnel
syndrome, a rare and painful compressive neuropathy in the proximal forearm. Ultrasound-guided thread-release is a promising new
technique for the decompression of peripheral nerves. This study aims to gain initial experience in performing thread-release on the
posterior interosseus nerve by transecting the Arcade of Frohse in the anatomical model. Ten ultrasound-guided thread-interventions
were performed in fresh cadaveric hands. Subsequent anatomical dissections evaluated outcome of the transection and possible
damage to adjacent structures. Ultrasound visibility was also documented for every intervention.
Results or Findings: A complete transection of the Arcade of Frohse was achieved in eight out of ten specimens. Ultrasound
visibility was sufficient in all cases. There was no macroscopically visible damage to surrounding neurovascular structures. However,
small lesions to the supinator muscle and the superficial extensor muscle were unavoidable in three cases.
Conclusion: Ultrasound-guided thread-transection is a safe and effective minimally-invasive procedure for transecting the Arcade of
Frohse in an anatomical model.
Limitations: This study was conducted on fresh anatomical specimens. Therefore, patient results may differ and further cadaveric
and clinical studies are warranted. Furthermore, the study sample size is limited due to cadaver availability.
Ethics committee - additional information: This study was approved by the Ethics Committee at the Medical University of Vienna.
Evaluation of tibial and median nerves in patients with diabetic peripheral neuropathy using shear wave elastography
(7 min)
Amandeep Singh; Amritsar / India
Author Block: A. Singh; Amritsar/IN

Purpose: Diabetic peripheral neuropathy (DPN) is a major complication of diabetes mellitus. Clinical examination and nerve
conduction study (NCS) of both tibial and median nerves are done to diagnose DPN. The purpose of this study was to perform high-
resolution ultrasound with shear wave elastography (SWE) to assess tibial and median nerves mean stiffness as an indicator for
development of DPN.
Methods or Background: The study was conducted on 90 subjects which included 30 healthy individuals, 30 diabetic patients
without DPN and 30 diabetic patients with DPN. The tibial and median nerve stiffness was measured with SWE. The differences in
stiffness values among patients with DPN, patients with clinically defined DPN, patients without DPN, and healthy volunteers based on
clinical features and electrodiagnostic tests were evaluated.
Results or Findings: The tibial nerve stiffness based on mean (EMean), minimum (EMin), and maximum
(EMax) shear elasticity indices was significantly higher in patients with DPN and clinically defined DPN than that in patients without
DPN and control subjects (p<0.05). EMin cutoff value had a sensitivity, specificity, PLR, and NLR. The inter- and intraobserver
agreements
were excellent for the SWE measurements.
Conclusion: Tibial nerve stiffness is significantly higher in diabetic patients with DPN and clinically defined DPN. The EMean and EMin
have a good accuracy for identifying DPN. SWE has a potential use for cases with clinically defined DPN.
Limitations: The study was limited by its small sample size.
Ethics committee - additional information: This study was approved by the Institutional Ethical and Research Committee.
Imaging features of hourglass constriction of the radial nerve with surgical correlation (7 min)
Juin Kim; Seoul / Korea, Republic of
334
Author Block: J. Kim, Y. Kim, C-H. Lee, S. Lee; Seoul/KR
Purpose: The objective of this study was to evaluate the imaging characteristics of hourglass constriction (HC) of the radial nerve
using ultrasonography (USG) and MRI correlating with surgical observations.
Methods or Background: From March 2020 to September 2023, the preoperative USG and MRI of six patients (5 men, 1 woman;
mean age 30.5 years old) who were surgically confirmed were retrospectively evaluated. The location and number of HCs identified
by both USG and/or MRI, as well as the signal intensity of the nerve in axial fat-suppressed T2 weighted images (FST2), were assessed
and compared with surgical observations.
Results or Findings: A total of 37 HCs were identified across the six patients. All patients exhibited three or more multiple HCs. The
most common locations for HCs were the intermuscular space between the brachialis and brachioradialis muscles (19/37, 51.4%),
followed by the lateral epicondyle level (7/37, 18.9%) and the arcade of Frohse (4/37, 10.8%). On FST2, the peripheral high signal
intensity and central low signal intensity (referred to as the "bull's eye sign") was most frequently observed proximal to the HC
(20/37, 54.1%). Out of the 37 HC sites, 23 were found to have constriction bands around the perineurium of the fascicle of the deep
radial nerve. This was most often located near the branching site from the radial nerve within the intermuscular space between the
brachioradialis and brachialis muscles.
Conclusion: Hourglass constrictions of the radial nerve typically show multiple involvements in the perineurium of the deep radial
nerve fascicle with bull’s eye sign, especially around the branching site from the radial nerve between the brachialis and
brachioradialis muscles.
Limitations: This study was a retrospective study of a single centre, with a small sample size and a lack of consensus in the accurate
cutoff degree of significant hourglass constriction.
Ethics committee - additional information: This study was approved by the Hanyang University Research Committee.
High-frequency ultrasound of pulleys of the flexor tendons of the toes (7 min)

Federico Pistoia; Genova / Italy
Author Block: F. Pistoia, F. Zaottini, R. Picasso, M. Maccio, D. Bianco, S. Rinaldi, M. Pansecchi, C. Martinoli; Genoa/IT
Purpose: The purpose of this study was to determine the ability of high-frequency ultrasound (HRUS) to assess the pulleys of the
flexor tendons of the toes.
Methods or Background: The initial phase of the study involved examining the pulley system of flexor tendons of the toes in six
cadaveric specimens using high-frequency ultrasonography. Coloured latex was then injected under US guidance at various pulley
levels. In the subsequent part of the research, a 22-8 MHz US probe was used to evaluate the toe pulley system in 40 feet of 20
healthy volunteers. Then, five patients with hammer toes were evaluated using the same US probe.
Results or Findings: The coloured latex was placed correctly in all six specimens. The toe pulleys were consistently identified with
HRUS on all forty healthy volunteers. All five patients demonstrated a mean thickness value of the AIII pulley of the affected toes
superior to the respective toes in the healthy group.
Conclusion: High-frequency US allows excellent depiction of toe pulleys. However, further research is necessary to understand the
potential impact of thickened pulleys on hammer toe deformity.
Limitations: This is a first explorative study. Further research in larger population is needed to understand the clinical value of this
ultrasound application.
Ethics committee - additional information: This study was performed in line with the principles of the Declaration of Helsinki.
Ultrasound-guided percutaneous irrigation of extra-shoulder calcific tendinopathy (7 min)

Francesca Serpi; Milan / Italy
335
Author Block: F. Serpi1, U. Viglino2, C. Messina1, S. Fusco1, S. Gitto1, F. Lacelli3, L. M. Sconfienza1, D. Albano1; 1Milan/IT, 2Genoa/IT,
3
Pietra Ligure/IT
Purpose: This study aimed to investigate the efficacy and safety of ultrasound-guided percutaneous irrigation of calcific
tendinopathy (US-PICT) applied out of the shoulder, comparing its effectiveness to US-PICT of the rotator cuff.
Methods or Background: Patients subjected to US-PICT for extra-shoulder calcific tendinitis (Case Group) were compared to those
subjected to US-PICT of the rotator cuff (Control Group). We had pre-procedure VAS (visual analogue scale) pain score, 1-month and
3-month VAS of patients of the Case Group, pre-procedure, and 3-month VAS of patients of the Control Group.
Results or Findings: The Case Group consisted of 41 patients (27 women; mean age: 45±9years): 26 gluteus medius, 5 patellar
tendon, 3 rectus femoris, 2 gluteus maximus, 2 common extensor tendon, 1 extensor carpi radialis longus, 1 pes anserinus, 1
peroneus longus. The Control Group included 41 patients (27 women; mean age: 47±11 years). The mean pre-procedure VAS of the
Case Group was 8.8±0.7 with a significant (P<0.001) drop at 1 month (4.5±0.6) and 3 months (3.6±0.6). The mean pre-procedure
VAS of the Control Group was 8±1.4 and dropped to 3.1±1.6 after 3 months (P<0.001). Post-treatment VAS at 3 months was not
significantly different between two Groups (p=0.134). Similarly, the decrease of VAS from baseline to three months was not
significantly different between the two Groups (p=0.264).
Conclusion: This study demonstrated the safety and effectiveness of US-PICT as a valuable therapeutic option for extra-shoulder
calcific tendinitis, with similar clinical outcome to the same procedure performed in the rotator cuff.
Limitations: This study had a relatively small sample size with few cases in some localisations. A further limitation is the
retrospective nature of the study, warranting prospective studies to compare the procedure with other conservative approaches and
to understand how to manage these patients postoperatively.
Shear wave of patellar and Achilles tendon on healthy patients: intervendor intraobserver and interobserver
comparison (7 min)
Umberto Viglino; Genoa / Italy
Author Block: U. Viglino1, F. Esposito2, D. Albano3, S. Gitto3, C. Messina3, L. M. Sconfienza3; 1Genoa/IT, 2Napoli/IT, 3Milan/IT
Purpose: Shear wave elastography is a renowned ultrasound technique that quantitatively evaluates the elasticity of tissues. The
aim of this study is to validate the reproducibility of elasticity measurements of the Achilles and patellar tendons in healthy subjects
using ultrasound machines from three different manufacturers and to compare intraobserver and interobserver variations between
two examiners.
Methods or Background: High-resolution linear probes were used on three ultrasound machines from different manufacturers. The
study included 13 healthy volunteers with no prior history of tendon injuries. Data acquisition was performed in both supine and prone
rest positions, with three measurements taken for each tendon on each side of the body, for a total of 26 tendons. Each joint was
repositioned after the measurement. A comparison of intraobserver variations was made with a reevaluation after two weeks (T1).
Results or Findings: Intraobserver reproducibility of the three different acquisitions on each tendon was demonstrated for both
observers on the three ultrasound machines (p<0.001). Intraobserver reproducibility T0-T1 was statistically significant (p<0.05) for
only one observer across the three ultrasound machines for both tendons. Interobserver reproducibility was statistically significant
(p<0.05) only for patellar tendon on two of the ultrasound machines [0.398 (CI 0.02-0.676); 0.421 (CI 0.048-0.691)]. Intervendor
reproducibility was poor, not statistically significant [Reader 1 0.257 (CI -0.138-0.581) for patellar and 0.222 (CI -0.174-0.556) for
Achilles] [Reader 2 0.075 (CI -0.315-0.443 for patellar and -0.118 (CI -0.477-0.275) for Achilles].
Conclusion: This study demonstrated that shear wave ultrasound could be a useful quantitative tool for the study of tendons but its
reproducibility is acceptable only in specific conditions (such as with the same sonographer). Reproducibility has shown to be
suboptimal in other applications, for example in reexamination, with a different sonographer, and poor with different ultrasound
machines.
Limitations: This study was limited by the lack of supersonic shear imaging (SSI).
Establishing baseline enthesis stiffness range in a healthy adult population (7 min)

Mohammad Khairi Jahidi Mahazer; Terengganu / Malaysia
336
Author Block: M. K. J. Mahazer; Terengganu/MY
Purpose: The purpose of this study is to determine the potential of shear wave elastography (SWE) in measuring the median
stiffness value of enthesis in healthy subjects and to assess interobserver reliability.
Methods or Background: Ninety-eight healthy male and female subjects aged 20-50 years underwent SWE of the bilateral enthesis
of quadriceps, suprapatellar, infrapatellar, Achilles, and triceps performed by two experienced sonographers. Each enthesis was
evaluated five consecutive times using the ElastQ software. Median shear wave velocity values were compared for each enthesis by
each examiner. Interobserver agreement was evaluated.
Results or Findings: Kruskal Wallis and ANOVA tests showed significant differences in median shear wave velocity (SWV) of
enthesis for quadriceps, suprapatellar, and left Achilles between the 40-50 age group vs the 20-29 age group and the 30-39 vs 40-49
age groups. However, insignificant difference of median SWV was found in the infrapatellar and right Achilles. The median (IQR)
stiffness of quadriceps for age groups 20-39 and 40-50 were 2.96m/s (0.42) and 2.80m/s (0.36), suprapatellar for age groups 20-39
were 3.00m/s (0.27), and for ages 40-50 were 2.90m/s (0.32), and left Achilles 20-39 years old 33.12 m/s (0.55) and 40-50 years old
(2.91m/s (0.29) respectively. Good interobserver agreement was found between quadriceps, suprapatellar, infrapatellar, and Achilles
(95% CI; 0.84) (p<0.01).
Conclusion: The study suggests that shear wave elastography holds potential as a reliable technique for quantifying enthesis
stiffness. It is noteworthy that age does not appear to have a significant influence on the median shear wave velocity (SWV) of
entheses, except for the quadriceps, suprapatellar, and left Achilles in patients aged 40-50.
Limitations: The study was limited by its small sample size, and the lack of additional imaging such as MRI.
Ethics committee - additional information: This study was approved by the University of Malaya Medical Centre, MREC ID NO:
2021112-10731.
High-resolution ultrasound of the anterior talofibular ligament: correlation between morphological and clinical
features (7 min)
Giovanni Marcenaro; Genova / Italy
Author Block: G. Marcenaro; Genoa/IT

Purpose: The anterior talofibular ligament (ATFL) is usually considered as a single fascicle but it is actually made of two different
bands, the superior and the inferior bands. The main objective of this study is to determine the role that these two fascicles play in
the ankle stability.
Methods or Background: We recruited 86 patients who accessed the emergency department for ankle sprain. An x-ray exam has
been used to rule out patients with fractures. After 14 days patients underwent an ultrasound examination to look for injuries of the
superior and inferior fascicle of the ATFL. Each fascicle has been given a score (0 to 3) to describe the severity of the injury. An
objective measure of the articular instability has been made measuring the minimum (a) and maximum (b) distance between the tip
of the fibula and the body of the talus, positioning the foot in plantar flexion and inversion (a) and in dorsiflexion and eversion (b).
Two months after the exam, patients were asked to fill in the CAIT questionnaire to provide objective data about ankle instability,
which was correlated with the findings of the ultrasound exam.
Results or Findings: In a population of 86 patients, the multivariate analysis shows that there is a significant relation between injury
of the calcaneofibular ligament CFL and ankle instability (p≤0.029) and between the two fascicles of the ATFL an injury of the superior
is more likely to cause instability.
Conclusion: The CFL is the most important ligament for preserving ankle stability. There is a trend showing that the superior band of
the ATFL is more responsible for ankle stability than the inferior, which is to be confirmed with a broader dataset.
Limitations: Recruitment for a follow-up study is in progress, as this study is based on the questionnaire responses of 100 patients.
Ethics committee - additional information: The study has been approved by the ethics committee of San Martino Hospital HRCCS
of Genoa, Italy.
337
EFRS 3 - MSK European Reporting Networking
Categories: Musculoskeletal, Professional Issues, Radiographers, Research

ETC Level: LEVEL I
This session aims to present an overall picture of European Radiographers' perspectives on MSK Reporting, enhancing the importance
of developing standards of practice for this field.
Moderator:
Altino Jorge Conde Da Cunha; Cumieira / Portugal

Introduce Radiographers MSK European Reporting Networking (20 min)

Paul O'Riordan; Leicestershire / United Kingdom
Janni Jensen; Odense / Denmark
Closing (5 min)
338
RPS 917 - Optimising imaging in emergency radiology: current and future trends
Categories: Abdominal Viscera, Emergency Imaging, EuroSafe Imaging/Radiation Protection, General Radiology, Head and Neck,
Management/Leadership
CME Credits: 1
Moderator:
Idil Güneş Tatar; Leuven / Belgium
An evaluation of multiple upper limb x-rays: is adjacent imaging necessary? (7 min)

Dania Abu Awwad; Camperdown / Australia
Author Block: D. Abu Awwad, E. U. Ekpo; Sydney, NSW/AU

Purpose: Medical imaging is a valuable diagnostic tool in emergency departments. However, the reliance on imaging leads to
increased practices that may deviate from evidence-based guidelines. X-ray requests for extremities sometimes also simultaneously
ask for X-ray projections for other adjacent regions. However, the usefulness of these additional projections is unclear. Our aim was to
assess the diagnostic yield of X-rays performed on upper limb regions adjacent to the area of injury.
Methods or Background: Data were collected retrospectively and included all patients who had X-rays of multiple upper limb
regions. Clinical history notes and radiological reports were used to establish the diagnostic yield of X-rays performed on a region
adjacent to the injured limb, and the association between patient symptoms and diagnostic outcomes were assessed using Chi-
Square analysis.
Results or Findings: A total of 1455 upper limb X-ray imaging were performed on 591 patients. A third of these patients had three
or more X-rays. About 95% of the abnormalities were detected on X-rays performed at the site of injury. The diagnostic yield of X-rays
performed on the region adjacent to the injured limb was only 1.5%. Chi-Square analysis showed that X-ray examination of adjacent
limb was less likely to detect new (previously undetected) abnormalities (χ2(2) = 498.981; p = <0.001). New abnormalities were
more common in the wrist or shoulder, and the majority were due to an injury and in women older than 65 years.
Conclusion: Adjacent upper limb images have very low diagnostic yield or impact. A change in the practice is needed to maximise
the benefits of imaging and reduce the cost and risks associated with additional X-ray imaging of asymptomatic adjacent limbs.
Ethics committee - additional information: This study was approved by the Sydney Local Health District Human Research Ethics
Committee (2022/ETH01291).
High-pitch head CT in uncooperative emergency department patients with acute head trauma: evaluation of image
quality and detection of acute intracranial pathologies (7 min)
Jan-Erik Scholtz; Frankfurt a. Main / Germany
339
Author Block: J-E. Scholtz, D. Leitner, L. D. Grünewald, S. Mahmoudi, V. Koch, C. Booz, T. Vogl; Frankfurt a. Main/DE
Purpose: The purpose of this study was to evaluate image quality and diagnostic accuracy of an unenhanced high-pitch head CT
protocol used in uncooperative emergency department (ED) patients after acute trauma.
Methods or Background: We retrospectively compared high-pitch (1.55) head CT protocol (CTDIVOL 34.4 mGy) used in
uncooperative ED patients and 1:1 age-matched ED patients who underwent standard-pitch (0.55) head CT protocol (CTDIVOL 39.5
mGy) after acute trauma between January 2023 and August 2023. Gray matter (GM) signal-to-noise ratio (SNR), white matter (WM)
SNR, GM-WM contrast-to-noise ratio (CNR), and posterior fossa artefact index (PFAI) were compared. Two radiologists performed
qualitative image evaluation of intracranial and bone structures, artefacts, and diagnostic evaluation for acute intracranial pathologies
and fractures.
Results or Findings: Overall, 87 patients [male, n=55(63.2%), 68±22 years] underwent high-pitch CT protocol. Quantitative image
evaluation did not show significant differences between high-pitch and standard protocols: GM SNR, mean±SD, 18.3±3.4 vs.
17.6±2.6, WM SNR, 15.2±2.5 vs. 14.8±2.3, GM-WM CNR 2.3±0.7 vs 2.2±0.6, PFAI, 3.4±0.3, all p≥0.41. Grey-white differentiation
[median (IQR); 5(4,5) vs. 5(4,5)], delineation of cerebrospinal fluid spaces, and bone structures were rated very good for both
protocols without significant differences (p≥0.05). Artefacts in the supratentorial region were low in both protocols (p=0.09), while
beam-hardening artefacts in the infratentorial region were slightly worse in high-pitch scans (p=0.001). Diagnostic confidence to
evaluate for intracranial haemorrhage, brain herniation, and fracture was very good for both protocols without significant differences.
Inter-rater agreement for the detection of intracranial haemorrhages (n=40), herniation (n=17), and fractures (n=14) was excellent
for both protocols (ĸ,1.0).
Conclusion: High-pitch head CT protocol is a safe and robust alternative in uncooperative ED patients to evaluate for acute
intracranial pathologies after trauma.
Limitations: The high-pitch head protocol has not been evaluated for ischaemic stroke imaging.
Ethics committee - additional information: This was a retrospective study with a waiver for informed consent.
A simple clinical algorithm for prediction of relevant CT findings could reduce the number of low-value cranial CT scans
by more than 20% (7 min)
Marta Sánchez Canales; Murcia / Spain
Author Block: J. M. Plasencia Martinez, M. S. Canales, E. Otón González, N. I. Casado Alarcón, E. C. Cotillo Ramos, H. Ortiz Mayoral, B.
Molina-Lozano, J. M. Garcia Santos; Murcia/ES
Purpose: The number of cranial CT scans is exponentially increasing, without a clear benefit. Our aim is to determine the yield of
urgent cranial CT depending on the clinical reason for consultation (RC), and to put forward a diagnostic algorithm.
Methods or Background: This was a retrospective, randomised study in patients who underwent urgent cranial CT for non-
traumatic reasons, between 2017 and 2021. Using binary logistic regression, the discriminatory capacity of the variables was
analysed. The independent variables were RC and personal history, and the dependent variables are relevant radiological findings
(RRF) described in urgent cranial CT: acute ischaemia and haemorrhage, mass, oedema, or unknown hydrocephalus, and relevant
clinical-radiological findings (RCRF): patients with RRF and/or need for hospital admission.
Results or Findings: We recruited 702 patients, median age 62 [interquartile range 47-76] years, 339 (48.3%) males.
Variables that increased the RCRF risk (Odds Radio -OR- >1, P<0.05) were motor deficit (OR_5.086; 95%CI_2.893-8.940), speech
deficit (OR_5.042; 95%CI_2.871-8.854), seizure disorder (OR_4.688; 95%CI_2.106-10.436), cognitive impairment (OR_2.639;
95%CI_1.316-3.851), sensory deficit (OR_2.223, 95%CI_1.300-3.800), oncological history (OR_1.750; 95%CI_1.238- 2.475), and
sudden altered mental status (OR_1.553; 95%CI_1.055-2.286). Variables that decreased the RCRF risk (OR<1, P<0.05) were dizziness
(OR_0.652; 95%CI 0.45-0.943), nausea-vomiting (OR_0.594; 95%CI_0.385-0.917), syncope-presyncope (OR_0.580;
95%CI_0.357-0.941), acute headache (OR_0.528; 95%CI_0.368-0.756), and vertigo (OR_0.267; 95%CI_0.142-0.505). When the above
RCs did not show discriminative ability, the ability of the remaining RCs was re-evaluated. An algorithm depicted is segregated, with
sensitivity and negative predictive value above 92%, which would reduce 23.9% of CTs in the full sample and 35.2% in non-admitted
patients, failing to diagnose 7% of RRF.
Conclusion: A clinical algorithm is proposed for relevant radiological findings in non-traumatic urgent cranial CT scans, potentially
avoiding at least 23% of CT scans.
Limitations: The study is awaiting prospective external validation.
Ethics committee - additional information: This study was approved with code CETI: 02/22.
Evaluation of low-dose whole brain CT perfusion-driven CT angiography utilising a deep learning image reconstruction
algorithm (7 min)
Yuhan Zhou; Zhengzhou / China
340
Author Block: Y. Zhou, L. Lei, Z. Wang, W. Cao, S. W. Yue; Zheng Zhou/CN
Purpose: The purpose of this study is to investigate the image quality and diagnostic accuracy of low-dose whole brain CT perfusion-
driven CT angiography with deep learning image reconstruction (DLIR) in patients with patients with acute ischeamic stroke (AIS)
caused by large vessel occlusion (LVO).
Methods or Background: This prospective study collected 62 patients with AIS caused by LVO, divided into standard-dose (80keV
150Ams) and low-dose (80keV 100Ams) CTP examination. The FBP, ASIR-V40% / 80% were using standard-dose group, and ASIR-
V80% and DLIR (M, H) were using low-dose group. ROIs were drawn in the CTP arterial phase images of siphon portion of the internal
carotid artery (ICA), middle cerebral artery M1 segment (MCA-M1), and temporal lobe for objective evaluation of image noise, SNR,
and CNR. Subjective evaluation included edge sharpness of vascular lumen, display of small blood vessels, and overall image quality.
Results or Findings: The low-dose DLIR-H group showed lower image noise. CNR of low-dose DLIR-H group in the ICA and MCA-M1
were higher than standard-dose FBP, ASIR-V40% groups, and low-dose DLIR-M group (P<0.05). In subjective evaluation, the low-dose
DLIR-H group was superior to the standard-dose FBP, ASIR-V40% groups, and low-dose DLIR-M group (P<0.05). The radiation dose in
the low-dose group was significantly lower than that in the standard dose group (P<0.05).
Conclusion: Reconstruction of low-dose CTP arterial phase images through DLIR may aid in the detection of small responsible
vessels. It provides a valuable low-dose scanning option for clinical evaluation without the need for additional scanning protocols and
extra radiation exposure and shortens the examination time.
Limitations: The number of AIS patients included in our study was limited and more cases are expected for further studies.
Ethics committee - additional information: Our retrospective study has obtained approval from the Institutional Review Board.
Evaluation of an optimised CT protocol for the assessment of suspected cauda equina syndrome and nerve root
compression (7 min)
Philip Jude Dempsey; Dublin / Ireland
Author Block: P. J. Dempsey, G. Lambe, D. T. Ryan, J. W. Power, A. Yates, G. Kenny, P. J. Macmahon; Dublin/IE
Purpose: The evaluation of suspected cauda equina syndrome and nerve root compression relies on urgent MRI. However, access to
MRI can be limited in many healthcare systems, especially in the out-of-hours setting. CT, by comparison, is more readily accessible.
The purpose of this study was to evaluate the diagnostic accuracy of an optimised CT lumbar spine protocol in the assessment of
suspected cord and nerve root compression using MRI as the reference standard.
Methods or Background: Individuals presenting to the emergency department and referred for MRI for suspected nerve root
compression were prospectively enrolled to have a CT spinal stenosis protocol in addition to MRI. Central canal and neural foraminal
stenosis was qualitatively and quantitatively assessed at each lumbar level on CT by an expert radiologist blinded to the clinical
information and outcome. Following a 4 week interval the process was repeated for each MRI.
Results or Findings: 59 individuals were included in the final analysis. In 22 (39%) cases no significant stenosis was identified. In a
further 22 (37%) cases disc pathology was identified that was managed conservatively. 13 (22%) individuals proceeded to urgent
surgical decompression. In one (2%) instance an alternative diagnosis was identified. The sensitivity, specificity, positive and negative
predictive values for CT in detecting acute neural compression were 97% (CI: 82-99%), 97% (CI: 83-99%), 97% (CI: 92-99%) and 97%
(CI 83- 99%) respectively.
Conclusion: CT accurately detected nerve compression and emergent surgical lesions. This can be a useful tool in selecting
appropriate patients for emergent MRI in the on-call setting to allow for better emergency resource allocation.
Limitations: The study was limited by its relatively small sample size, its reliance on external multicentre validation, and from the
lack of a universally agreed stenosis assessment method. This was a single centre study.
Relevance of computed tomography for the detection of septic foci: a systematic review and meta-analysis (7 min)
Ann-Christine Stahl; Berlin / Germany
341
Author Block: A-C. Stahl, K. Rubarth, M-I. Opper Hernando, G. Muench, M. Dewey, J. Pohlan; Berlin/DE
Purpose: This systematic review and meta-analysis was conducted to investigate the diagnostic accuracy of computed tomography
(CT) in patients with suspected or confirmed sepsis.
Methods or Background: We searched Medline by using PubMed and Embase by using Ovid with a systematic search strategy to
identify studies assessing the role of CT in patients with suspected sepsis or sepsis of unknown origin. Reviews, meta-analyses,
editorials, comments, or case reports were not eligible for inclusion. In addition, we searched the references of the included studies to
ensure no relevant studies were missed. PubMed was last searched on 01.01.2023 and Ovid on 22.01.2023. We extracted the
sensitivity and specificity using 2-by-2 tables from each study. The methodological quality was assessed with the help of the
QUADAS-2 tool. Pooled estimates of per-CT sensitivity, specificity, and diagnostic odds ratio were calculated by using random effects
meta-analysis for proportions via logit transformation and presented as forest plots. This review was registered in PROSPERO (ID:
CRD42023412009).
Results or Findings: Five studies which included an overall of 838 patients met our inclusion criteria. The different reference
standards used were the final diagnosis from the discharge note, operation, local drainage, needle aspiration, and autopsy. The
pooled estimate of sensitivity was 83.2% (95% CI: 69.5–91.4%) while the pooled estimate of specificity was 62.4% (95% CI:
49.9-73.5%). The pooled estimate of the diagnostic odds ratio was 9.0 (95% CI: 2.2-36.3). Both sensitivity (I2=67%, p=0.02) and odds
ratio (I2=77%, p<0.01) showed significant variation between the studies.
Conclusion: CT has good sensitivity but limited specificity in patients with suspected or confirmed sepsis and may thus be
considered an appropriate test.
Limitations: Due to the low number of included studies as well as their heterogeneity the generalizability of our results may be
limited.
Ethics committee - additional information: As the presented study is a systematic review and meta-analysis no approval by an
ethics committee is needed.
The problem-solving role of the 'wait and repeat CT' approach in the diagnosis and treatment of acute abdomen (7 min)
Esra Akçiçek; Ankara / Turkey
Author Block: E. Akçiçek, A. G. Erdemir, I. S. Idilman, M. R. Onur, E. Akpinar; Ankara/TR

Purpose: Computed tomography (CT) findings in patients with acute abdominal symptoms are sometimes insufficient to make a
definitive diagnosis. Patients may be offered repeat imaging and, to our knowledge, the results of this approach have not been
studied. We aimed to investigate the impact of the 'wait and repeat CT' approach on the diagnostic and therapeutic processes of
emergency patients with acute abdominal symptoms who could not be diagnosed with the initial CT but were diagnosed with a repeat
CT.
Methods or Background: Approval was obtained from the ethics committee. Patients admitted to the emergency department
between January 2013 and July 2023 (≥18 years) were selected from those who had a repeat CT examination on the same admission.
Trauma patients and initial examinations with artefacts were excluded. According to the findings of repeat CT, patients were divided
into five subgroups: preliminary diagnosis did not change (group-A), suspicious preliminary diagnosis confirmed (group-B),
progression after initial diagnosis (group-C), regression after initial diagnosis (group-D), and completely different diagnosis (group-E).
The diagnoses in each group and the relationships between the groups were analysed statistically.
Results or Findings: 228 examinations were found, of which 72 were excluded due to artefacts and seven due to trauma. Of the
149 patients, 21 were in group-A (14.09%), 60 in group-B (40.26%), 32 in group-C (21.47%), 25 in group-D (16.77%), and 11 in group-
D (7.38%). Overall, the most frequent cases were partial bowel obstruction (n: 41, 27.51%) and the majority of patients whose
findings regressed to the second CT (group-D; n:18, 72.0%). The "wait and repeat" approach is statistically significant in the follow-up
and discharge decision of patients with partial obstruction (p<0.01).
Conclusion: The 'wait and repeat' approach may be helpful in follow-up of acute abdomen, especially if it is a bowel obstruction.
Limitations: The number of patients was limited.
Ethics committee - additional information: This study was approved by the Hacettepe University Ankara, Ethics commitee No:
GO 23/566.
The distribution of computed tomography findings for nonagenarians with acute abdominal conditions: a descriptive
study for more than a decade of observation (7 min)
Ayse Nur Bolukcu; Ankara / Turkey
342
Author Block: A. G. Erdemir, A. N. Bolukcu, E. Akçiçek, M. R. Onur, I. S. Idilman, E. Akpinar; Ankara/TR
Purpose: The objective of the study was to apply computed tomography (CT), which is useful to determine the causes of acute
abdomen in nonagenarian (≥90 years) individuals, since they are less likely to present with the traditional symptoms. We aimed to
describe the CT scan findings of acute abdominal etiologies in nonagenarian patients over a ten-year period, which to our knowledge
has never been studied in this context.
Methods or Background: Nonagenarians with acute abdomen admitted to the emergency department between January 2013 and
June 2023 with CT (n:620) were studied. Of these, trauma patients (n:112) were excluded. For the remaining 508 patients, an
assessment was made based on symptoms, demographics, etiology, surgical interventions, and mortality within the following day,
week, and month.
Results or Findings: Positive findings for acute abdomen were detected in 258 patients (50.78%). There were 131 patients
(50.77%) requiring surgical intervention and 54 of them (20.93%) were able to undergo surgical intervention. According to primary
acute abdominal pathology, they were categorized as follows: bowel obstruction (n:36, 13.95%), acute cholecystitis (n:29, 11.24%),
arterial occlusion (n:28, 10.85%), stercoral colitis (n:21, 8.13%), bowel infection (n:16, 6.20%), kidney stone (n:15, 5.81%), bowel
perforation (n:8, 3.10%), venous occlusion (n:6, 2.32%), acute appendicitis (n:4, 1.55%). We categorised patients based on their
mortality within the following day (22.72%), week (34.09%), and month (43.18%). Malignancy (n:114, 22.48%) was the most frequent
chronic abdominal pathology accompanying them.
Conclusion: The spectrum of acute abdominal etiology in nonagenarians, studied for the first time on such a large scale, differs
considerably from other segments of the population. Most importantly, admission to CT for acute abdomen in nonagenarians can be
an indicator of the likelihood that more than a third (or even nearly half) may die within a month.
Limitations: It was a single-centered and retrospective study.
Ethics committee - additional information: This study was approved by Hacettepe University Noninterventional Clinical Research
Ethics Committee.
343
OF 9E - An infinite number of ways to design good studies

CME Credits: 1
In this session, organised by the ESR Journal Family, you will learn about key design elements of your research: sample size, bias, p-
value, data management and more. It will help you overcome obstacles and produce relevant and methodologically accurate studies
that will move your research career forward.
Moderator:
Francesco Sardanelli; Genova / Italy

How to care about sample size and biases: when is enough really enough? (15 min)
Caterina Beatrice Monti; Milan / Italy
Why avoid p-hacking and other significant errors (15 min)

Nathaniel Mercaldo; Boston / United States
Relevant issues on splitting, randomising and merging data 101 (15 min)
344
EDiR 9 - The role of artificial intelligence (AI) in exam marking
Categories: Education, Management/Leadership, Professional Issues, Students

CME Credits: 1
AI is increasingly having a major global impact on everyday life. There is a growing recognition of the importance of AI in education
and academic research. The European Board of Radiology (EBR) has always embraced the potential of the latest technologies to
enhance the conduction of the European Diploma in Radiology (EDiR). The EBR did not want to miss this opportunity to see if and to
what extent AI could be applied in the EDiR. The EDiR has multiple free-text questions that require experts in the field to review the
answers manually. This is a time-consuming task and error-prone. The EBR wanted to test whether natural language processing (NLP)
can automatically analyse these free text answers to support the review process. Exams are a very important and indispensable part
of education. AI for automatic evaluation of descriptive answers can improve the efficiency and accuracy of grading exams, reducing
the time and effort required. Join us on this exciting journey, and let's unravel the ins and outs of Natural Language Processing (NLP)
and exam marking with some of the brightest minds in the field.
Moderator:

Can AI help in the correction of medical examinations? (15 min)

What is natural language processing (NLP), and why is it so important nowadays? (15 min)
Benedikt Kämpgen; Würzburg / Germany
Is it possible to implement NLP in the EDiR? (10 min)

Discussion (15 min)
345
RT 9 - Legal aspects of medical AI
Categories: Artificial Intelligence & Machine Learning, EuroSafe Imaging/Radiation Protection

CME Credits: 1
The session is an introduction to the legal framework surrounding the implementation of AI in clinical practice and radiology
departments and what challenges are perceived by leaders in healthcare. The discussion will help us bridge the gaps and highlight
pitfalls so that implementation can be as smooth as possible.
Moderator:

The imminent EU regulation, and what it means in practical solutions (10 min)
Mathias K. Hauglid; Oslo / Norway
How do we convert in-house AI research into a medical device? (10 min)

Filippo Pesapane; Milan / Italy
Challenges and possibilities when implementing AI in healthcare (10 min)

Lena Petersson; Halmstad / Sweden
Discussion (25 min)
346
RC 903 - Pros and Cons: photon-counting CT (PCCT): a game changer for cardiovascular imaging
Categories: Cardiac, Contrast Media, Imaging Methods, Vascular

CME Credits: 1
Moderator:

Pro (20 min)

Tilman Stephan Emrich; Mainz / Germany
This house believes that PCCT will revolutionise cardiovascular imaging with material decomposition.
Con (20 min)

Pál Maurovich-Horvat; Budapest / Hungary
This house believes that PCCT is a step-change for cardiovascular imaging with improved resolution.
Panel discussion: Is PCCT a revolution or a step-change for cardiovascular imaging? (15 min)
347
ESR/ESTRO - Which liver ablation technique is best?
Categories: Abdominal Viscera, Evidence-Based Imaging, Interventional Oncologic Radiology, Multidisciplinary

CME Credits: 1
The session is focused on evaluating various liver ablation techniques to determine the best approach within the multidisciplinary
framework of interventional oncologic radiology and radiation therapy. Experts from interventional radiology and radiation oncology
share their insights and review the available evidence on the most commonly utilised liver ablation techniques, and discuss the
benefits and limitations of these techniques, considering factors such as tumour size, location, proximity to critical structures, but also
efficacy, safety, treatment outcomes, and patient-specific considerations. The advancement in imaging-guided radiological ablation
techniques as percutaneous thermal and transarterial chemo- and radio-ablative techniques, stereotactic body radiation therapy
(SBRT), and image-guided radiation therapy (IGRT) are highlighted, also addressing the potential of combination therapies as an
additional treatment arsenal for clinicians. The audience will recognise that integrating interventional oncologic radiology, radiation
therapy, and evidence-based imaging into a comprehensive framework will determine an optimisation of liver ablation techniques,
tailor treatment strategies to individual patients, and consolidate evidence to refine treatment guidelines.
Moderators:
Cihan Gani; Tübingen / Germany
Thomas Karl Helmberger; Munich / Germany

Multidisciplinary tumour board (50 min)
348
E³ 22B - Shoulder injuries: beyond the cuff

CME Credits: 1
Moderator:

Intraarticular lesions of the glenohumeral joint (25 min)

Christian W.A. Pfirrmann; Forch / Switzerland
1. To differentiate labral variants from tears.

2. To learn about typical injury patterns in shoulder instability.
3. To become familiar with the postoperative complications after joint stabilisation.
Acromioclavicular (AC) joint and extraarticular lesions (25 min)

Frank W Roemer; Erlangen / Germany
1. To understand the anatomy and traumatic injury of the AC joint and ist stabilisers.
2. To describe the main traumatic pathologies of non-cuff-related periarticular tendon and muscle injury.
3. To describe post-traumatic inflammatory conditions and nerve entrapment syndromes.
349
TC 927 - Dementia: towards an aetiologic diagnosis

CME Credits: 1
This educational session will discuss how in current clinical practice imaging modalities such as MRI and molecular imaging
techniques can be used to support an aetiologic diagnosis of dementia. This session will give general radiologists and those interested
in neuroradiology and nuclear medicine/molecular imaging the knowledge needed for appropriate selection and structured
assessment of imaging exams in diagnostic work-up of dementia. The session format consists of in part lectures, followed by an
interactive case-based discussion.
Moderators:
Alexander Drzezga; Cologne / Germany

MR imaging in dementia aetiology: patterns of atrophy and vascular lesions (19 min)
Molecular imaging in dementia aetiology: current concepts (19 min)

Dementia aetiology: interactive case discussion (19 min)

350
OF 9R - Opportunities for change in nuclear medicine and PET imaging
Categories: EuroSafe Imaging/Radiation Protection, Hybrid Imaging, Nuclear Medicine, Radiographers

CME Credits: 1
This session is focused on the evolving landscape of nuclear medicine and Positron Emission Tomography (PET) imaging. Comprising
three distinct talks, this session explores strategies to enhance safety, optimise practice, and leverage cutting-edge technology and
data analysis in these critical areas of medical imaging. These talks offer invaluable knowledge, strategies, and insights to empower
radiographers and healthcare professionals to navigate opportunities for positive and effective changes when utilising nuclear
medicine and PET imaging, ensuring the highest quality patient care while enhancing safety and diagnostic precision.
Moderator:
Mélanie Champendal; Lausanne / Switzerland

Strategies to reduce staff doses in PET/CT (16 min)

Shauna Murphy; Dublin / Ireland
How can radiographers optimise practice whenusing a CZT camera? (16 min)
Iain Tabone; Msida / Malta
Radiomics in PET/CT (16 min)

Pedro Silva Costa; Porto / Portugal
351
RC 914 - Justification of radiographic examinations: getting it right
Categories: EuroSafe Imaging/Radiation Protection, Imaging Methods, Professional Issues, Radiographers

CME Credits: 1
Moderators:
Graciano Paulo; Coimbra / Portugal
Aleksandar Gjoreski; Skopje / Macedonia

Aleksandar Gjoreski; Skopje / Macedonia
How can referral guidelines and clinical decision support tools support the justification processes? (15 min)
Maryann Hardy; Bradford / United Kingdom
1. To describe the current legislative requirements for justification across Europe.

2. To review existing evidence-based guidelines and clinical decision support systems.
3. To discuss how referral guidelines and clinical decision support tools can enhance the justification process.
Interventions to reduce the quantity of low-value radiographic examinations (15 min)

Catherine Chilute Chilanga; Drammen / Norway
1. To identify the stakeholders within the healthcare system who have an influence on reducing low-value radiographic examinations.
2. To identify effective interventions to reduce low-value radiographic examinations from the aspect of each stakeholder.
3. To describe each stakeholder's roles in reducing low-value radiographic examinations.
Can AI help to reduce inappropriate referrals? (15 min)

Jaka Potočnik; Dublin / Ireland
1. To review the role of AI in the justification of imaging examinations.

2. To describe how AI tools may impact medical imaging referrals in the future.
3. To discuss the opportunities and challenges of implementing AI within existing imaging referral pathways.
Panel discussion: How can radiographers successfully work within a multidisciplinary team to support effective
justification? (10 min)
352
RC 904 - Incidental findings on chest CT: to ignore or to act?
Categories: Breast, Cardiac, Chest, Musculoskeletal

CME Credits: 1
Moderator:
Chiara Pozzessere; Lausanne / Switzerland

Chiara Pozzessere; Lausanne / Switzerland
Incidental findings in the breasts (15 min)

Thomas Frauenfelder; Zurich / Switzerland
1. To identify the different types of incidental findings commonly seen in the breast during chest CT scans.
2. To describe the findings in relation to the technique used.
3. To manage the incidental findings in the breast, including decision-making processes about when to ignore and when to act.
Incidental findings in the heart (15 min)

Rodrigo Salgado; Antwerpen / Belgium
1. To learn about the characteristics of the heart on non-gated chest CT.

2. To understand the different categories of incidental findings which can be encountered.
3. To learn about the imaging findings of common and less common but important cardiac incidental findings.
4. To understand the limitations of a non-gated CT examination and how to subsequently formulate the findings in a radiology report.
Incidental findings in the chest wall (15 min)

Violeta Vasilevska Nikodinovska; Skopje / Macedonia
1. To identify the different types of incidental bone and soft tissue lesions of the chest wall.
2. To learn about the imaging findings of common and rare incidental findings of the chest wall.
3. To learn about further diagnostic and imaging approaches to different chest wall lesions.
Panel discussion: Structured reporting and good practices for reporting incidental findings on chest CT (10 min)
353
RPS 913 - Spectral imaging: quantification and performance evaluation
Categories: Imaging Methods, Physics in Medical Imaging, Research

CME Credits: 1
Moderator:
Osvaldo Rampado; Turin / Italy
Comparison of image quality of abdominal CT examinations and virtual noncontrast images between photon-counting
and energy-integrating detector CT (7 min)
Ludovica Lofino; Milan / Italy
Author Block: L. Lofino, F. Schwartz, F. Ria, M. Z. Zarei, E. Samei, A. Abadia, D. Marin; Durham, NC/US
Purpose: The purpose of this study was to compare the image quality of portal venous phase (PVP) abdominal CT examinations and
virtual non-contrast images (VNC) between photon-counting CT (PCCT) and energy-integrating Detector CT (EID).
Methods or Background: Multi-phase CT scans from one PCCT and two EID CTs were retrieved. 45 BMI-matched patients were
included: 15 for PCCT and 30 for EID. In vivo image quality parameters were measured and compared for PVP and VNC. CTDIvol
values were also recorded for all examinations. Because scanner tube current modulation adapts to patient size, the radiation dose
was compared among scanners accounting for BMI using a figure of merit: FOM=1/(BMI*lnCTDIvol). A five-point scale was used to
assess the reader's perception of image quality.
Results or Findings: Compared to the two EID, PCCT yielded significantly improved resolution and noise magnitude for both PVP
(MTFf10 = 0.55 ± 0.08 for PCCT vs. 0.50 ± 0.04 and 0.49 ± 0.03 for Flash and Force; noise = 9.76 ± 3.10 vs. 15.35 ± 4.14 and 10.70
± 1.34) and VNC (MTFf10 = 0.56 ± 0.01 for PCCT vs. 0.51 ± 0.05 and 0.51 ± 0.03 for Flash and Force; noise = 9.59 ± 2.77 vs. 13.90
± 3.57 and 10.83 ± 2.83), P<0.02. A similar trend was confirmed in a subset of overweight patients. Our FOM analysis suggests that,
for equal radiation exposure levels and comparable patient size, PCCT yields 20% noise reduction compared to the two EID, with 18%
reduction in overweight patients. Reader’s perceived image noise was lower and overall image quality was higher for PCCT compared
to EID.
Conclusion: PCCT yields a significantly lower radiation dose, with improved image quality in both the PVP and VNC of abdominal CT
examinations.
Limitations: This was a single center retrospective study with a limited number of cases.
Ethics committee - additional information: An ethics committee approved the study.
Assessment of artefacts, material identification, and quantification in the presence of metal objects using photon-
counting spectral CT (7 min)
Aamir Younis Younis Raja; Abu dhabi / United Arab Emirates
354
Author Block: B. Tariq1, N. Werghi1, E. I. Memisoglu1, N. Maalej1, O. Sikander2, F. Naseer2, A. Y. Y. Raja1; 1Abu Dhabi/AE, 2Islamabad/PK
Purpose: Metal artefacts in computed tomography (CT) obscure the visualisation and assessment of anatomical structures and lead
to misinterpretation of patient diagnosis and treatment. This study aims to use photon-counting spectral CT and evaluate the
identification and quantification of clinically relevant materials such as iodine (as a contrast agent) and hydroxyapatite (a mineral in
bones and teeth) in the presence of metal objects.
Methods or Background: A multi-material phantom was used with inserts of varied quantities of iodine (4.83, 9.66, and 14.56
mg/cm3) and hydroxyapatite (201.6 and 406.9 mg/cm^3). Three sets of scans were acquired: one without a metal insert, one with
steel, and one with aluminium. Image acquisition used a Mars spectral scanner (Microlab 5x120) operated at 118 kVp, 80 μA, and 160
ms with 981 circular projections. Images were reconstructed in five energy bins: 7-40, 40-50, 50-60, 60-79, and 79-118 keV. Energy
and material-density images were assessed by linear regression, sensitivity, specificity, area under the curve (AUC), and root-mean-
square-error (RMSE). For demonstrative purposes, a biological sample (a sheep heart) with a steel insert was also scanned and
evaluated.
Results or Findings: Results indicate reduced metal artefacts and an enhanced signal-to-noise ratio (up to 25%) in the higher
energy bins. All energy bins revealed strong linearity (R^2>0.97) across the concentrations of material. Material identification and
quantification were measured for iodine (without metal- Sensitivity 80%; Specificity 90%; AUC 0.80; RMSE 22%; with metal- Sensitivity
>77%; Specificity >90%; AUC >0.74; RMSE 26%) and hydroxyapatite (without metal- Sensitivty 83%; Specificity 93%; AUC 0.8;
RMSE,14%; with metal- Sensitivity >78%; Specificity >93%; AUC >0.77; RMSE >22%). Images of the biological sample showed
comparable results to the multimaterial phantom.
Conclusion: SPCCT can accurately identify and quantify clinically relevant materials, such as iodine and hydroxyapatite, in the
presence of metal objects.
Funding for this study: Funding was received through the Research and Innovation grant from the Khalifa University, UAE
(Project#: 8474000563).
Iodine quantification accuracy kVp-switching dual-energy CT (7 min)

Jan Heemskerk; Leiden / Netherlands
Author Block: S. Blazis1, T. Ruytenberg2, N. De Jong2, J. Heemskerk2; 1Goes/NL, 2Leiden/NL

Purpose: The purpose of this study was to determine (1) whether quantification of iodinated contrast medium of a kVp-switching
dual-energy CT is accurate, and (2) what the limit of this quantification is.
Methods or Background: Dual-energy CT has been proposed as a tool for semi-quantitative perfusion imaging, using iodine
concentration maps. We have investigated the accuracy and limitations of iodine quantification of a fast kVp-switching DECT. Two
phantoms were scanned: a Gammex multi-energy phantom, and a 180mm diameter cylindrical water-filled (‘linearity’) phantom
containing twelve 16mm diameter tubes with iodinated contrast medium with concentrations ranging from 0 to 30 mgI/ml.
Results or Findings: In line with what has been documented elsewhere, measurements with the ‘linearity’ phantom indicate that
lower iodine concentrations (<2mgI/ml) are underestimated with 60-20%. For higher concentrations, the accuracy of iodine
quantification is in the order of a few percent. Nevertheless, depicted iodine concentrations increase monotonously with actual
concentrations, indicating potential for scoring purposes or thresholding for clinical applications. Gammex phantom images show that
distinction between iodine and calcium is challenging for the kVp-switching DECT. This is also apparent from VNC images, where
iodine concentrations >3mgI/ml are no longer fully suppressed, and HU values for calcium inserts are reduced.
Conclusion: Dual-energy CT is increasingly proving itself as a useful expansion of CT imaging. Discrimination and quantification of
iodinated contrast medium is a promising additional functionality for, for example, detection of intra-cranial hemorrhage or
myocardial defect. Fast kVp-switching DECT switching facilitates this iodine quantification, but within limits, and allows creation of
virtual non-contrast maps at no extra dose cost. Quantitative iodine imaging with CT remains challenging for clinically relevant
concentrations <3mgI/ml, and further investigation of accuracy of iodine quantification and VNC-maps is warranted.
Limitations: This project included phantoms scans only.
Ethics committee - additional information: This project only included phantom scans.
Material decomposition using iodine quantification on spectral CT for characterising nodules in the cirrhotic liver (7
min)
Subhash Chand Kheruka; Lucknow / India
355
Author Block: S. C. Kheruka, R. Balaji, R. F. Al Sukaiti, N. A. Almaymani, N. Al Makhmari, A. Al Balushi, H. Said Mohammed, K. Al
Riyami, S. Al Rashdi; Muscat/OM
Purpose: The purpose of this research is to determine whether material breakdown using iodine measurement on spectral CT can
characterise cirrhotic liver nodules. This approach separates iodine from other liver components to offer quantitative and functional
data to improve diagnosis and therapy.
Methods or Background: This study utilised spectral CT imaging techniques for material decomposition and iodine quantification.
Dual-energy acquisition and image reconstruction algorithms were employed to separate iodine from other materials in the liver. The
study included patients with liver cirrhosis and nodules, and both conventional and spectral CT scans were performed. Quantitative
analysis of iodine content was conducted using specific regions of interest within the liver nodules.
Results or Findings: The results of this study demonstrated the feasibility and potential clinical utility of material decomposition
using iodine quantification on spectral CT for characterising nodules in the cirrhotic liver. The technique allowed for the direct
measurement of contrast enhancement within nodules, aiding in the assessment of perfusion characteristics and angiogenesis. The
differentiation between iodine-rich lesions, such as hepatocellular carcinoma (HCC), and non-enhancing lesions, such as cysts or
hemangiomas, was achieved, facilitating accurate diagnosis and treatment planning. Additionally, the technique enabled the
detection of subtle iodine enhancements within indistinct or small nodules that may be missed on conventional CT scans.
Conclusion: Materials decomposed using iodine measurement on spectral CT may improve cirrhotic liver nodule characterisation.
Quantitative metrics, including iodine concentration, wash-in and wash-out rates, and iodine maps can distinguish benign from
malignant lesions. Diagnostic accuracy, therapeutic choices, and personalised patient care may improve using this method.
Limitations: Further study and clinical validation are needed to prove the therapeutic value, optimise imaging techniques, and
assess patient outcomes of this method.
Performance of iodine quantification and monochromatic attenuation through high-pitch dual-source photon-counting
CT: a phantom study (7 min)
Peng Liu; Shanghai / China
Author Block: P. Liu, S. Zhou, Z. Xu, H. Dong, J. Li, S. Lin, W. Yang, F. Yan, L. Qin; Shanghai/CN
Purpose: The objective of this study was to investigate the feasibility and accuracy of iodine quantification and monoenergetic
attenuation using PCD-CT in standard pitch and high-pitch scanning at different scan parameters in a phantom model.
Methods or Background: Four inserts with known iodine concentrations (2, 5, 10, and 15 mg/mL) were placed in the removable CT
phantom and scanned using high-pitch (3.2) and standard pitch (0.8) on PCD-CT. Two tube voltages (120 and 140 kVp) and four
radiation doses (1, 3, 5, and 10 mGy) were alternated. Each scan setting was repeated three times. Mean iodine concentration and
monoenergetic attenuation were recorded. Percentage absolute bias (PAB) was assessed for iodine quantification. Image noise and
monoenergetic attenuation were evaluated at 40, 70, 100, and 140 keV.
Results or Findings: 96 acquisitions were conducted. In small phantom, the PAB was 2.96% (1.75%, 4.56%) and 1.67% (1.00%,
3.42%) for high-pitch and standard pitch, respectively. In large phantom, these numbers were 3.72% (1.75%, 5.97%) and 2.94%
(1.75%, 4.70%) respectively. Linear regression analysis revealed that only phantom size significantly influenced (P < 0.001) the
accuracy of iodine quantification. Background noise increased with a decrease in keV level and radiation doses. Attenuation errors at
70, 100, and 140 keV remained below 10 HU, with 37.5% cases surpassing 10 HU at 40 keV. Linear regression analysis revealed
comparable accuracy of monoenergetic attenuation between high-pitch and standard pitch (P=0.332).
Conclusion: High-pitch scanning in PCD-CT can be used to quantify iodine density and monoenergetic CT values with high accuracy,
thereby potentially benefiting multienergy-based tissue differentiation and material decomposition in clinical settings.
Limitations: The results of our study were focused on a phantom and have not been verified in patients, which must be conducted in
the future.
Ethics committee - additional information: The study is a phantom experiment.
Multi-contrast ultra-high-resolution K-edge imaging with edge-on deep-silicon photon-counting detectors: phantom
evaluations of a clinical prototype scanner (7 min)
356
Author Block: T. W. Holmes1, N. Nezami2, A. Jobayer3, R. Bujila4, J. Maltz3, Z. Yin5, A. Pourmorteza1; 1Atlanta, GA/US, 2Baltimore,
MD/US, 3Waukesha, WI/US, 4Stockholm/SE, 5Niskayuna, NY/US
Purpose: Photon-counting detectors (PCD) have helped improve diagnostic value of CT by providing spectral information, increasing
the spatial resolution, and reducing the radiation dose. The combination of these advantages could enable dose-efficient ultra-high-
resolution (UHR) multi-contrast imaging by using the K-edges of materials. Properly timed injections of two contrast agents can
generate co-registered maps of the anatomy in different perfusion phases and eliminate multi-phasic examinations. Moreover, use of
solid k-edge materials such as gold and tungsten as fiducial markers in surgical tools could improve CT-guided interventions. Si-based
PCDs with eight adjustable thresholds are a new development that promise better spatial and spectral resolution compared to the
existing CdTe/CZT technologies. We investigated the utility of Si-PCD in multi-contrast UHR tasks.
Methods or Background: We prepared a series of anthropomorphic phantoms mimicking coronary arteries with atherosclerotic
plaques and stents, tumours embolised with radiopaque microspheres, and human head including various concentrations and
mixtures of iodine- and gadolinium-based contrast agents. Another series were made including different components of a 1-mm
micro-robot designed for neurosurgery containing solid neodymium and tungsten components. The phantoms were scanned on a
whole-body prototype Si-PCD scanner and material-specific UHR images were generated. We compared iodine and gadolinium
quantification, image resolution, and material separation of the system to a state-of-the-art dual-energy energy-integrating detector
(EID) CT scanner.
Results or Findings: While the dual-energy EID system failed at separating the two materials, 8-energy Si-PCD could distinguish
them. Iodine quantification accuracy was not significantly different between the two systems in the absence of gadolinium.
Gadolinium accuracy was [-0.3 0.7] mgGd/mL for Si-PCD. Spatial resolution was significantly higher in Si-PCD as measured by in-stent
and free lumen diameters.
Conclusion: Si-PCD has the potential to improve spatial resolution and provide multi-material quantification simultaneously.
Limitations: This study utilised a prototype scanner.
Funding for this study: This study was part of a sponsored research agreement with GE Healthcare.
LLCD: how does low contrast detectability assessment in contrast-enhanced spectral mammography become liquid? (7
min)
Raffaele Villa; Monza / Italy
Author Block: C. Ingraito1, R. Villa2, N. Paruccini2, E. De Ponti2; 1Milan/IT, 2Monza/IT

Purpose: Contrast-enhanced spectral mammography (CESM) is an advanced imaging technique used in breast cancer diagnosis,
involving the injection of a iodinated contrast agent and the acquisition of two energy images. Despite its increasing widespread use
in clinical practice, it remains a non-standardised technique that lacks standardised protocols; moreover the specific designed
phantoms do not provide quantitative but qualitative estimation of image quality. The purpose of this work is to analyse the feasibility
of a new phantom, accompanied by a statistical method, to provide quantitative assessments of Low Contrast Detectability (LCD).
Methods or Background: A homemade low-contrast phantom, made of a central uniform region together with a step-wedge, was
3D-printed and filled with different areal concentrations of iodinated contrast agent (0.25÷1 mg/cm2). The phantom was placed over
slabs of different equivalent breast tissue thicknesses. Images from three mammographic equipment of two vendors were acquired in
CESM modality. The homogeneous region was used to calculate contrast thresholds for details of varying size with a statistical
approach. The iodine step-wedge permitted to express these LCD thresholds also as an absolute in terms of areal concentrations of
iodinated contrast agent (LLCD).
Results or Findings: The proposed statistical method is found to be reproducible. The differences in LCD threshold results coming
from the same vendors equipment are comprised in the found uncertainty of 10%. The findings show the expected behaviour of LCD
thresholds as the dose rate increases and as the slab thickness decreases.
Conclusion: This phantom, along with the statistical method, enables spectral characterisation and quantitative LCD assessments
and can be thought of as a useful tool for quality assurance constancy testing.
Limitations: The current phantom structure can be improved to provide a more accurate and practical filling-and-sealing system.
Dual-energy tomosynthesis of the chest using a triple-layer x-ray detector (7 min)

Steven Tilley; Waterloo / Canada
357
Author Block: S. Tilley, J. Potipcoe, K. S. Karim; Waterloo, ON/CA
Purpose: The purpose of this study was to investigate the feasibility of dual energy (DE) tomosynthesis using a triple-layer x-ray
detector. Of particular interest is whether a set of DE images, acquired with a tomosynthesis acquisition protocol and processed using
a traditional DE algorithm, can be reconstructed into material specific three-dimensional volumes. Adding the material separation
capabilities of DE with the depth localisation of tomosynthesis may provide enough patient information to negate the costly use of
computed tomography (CT) for many patients. Furthermore, using a multi-layer detector for DE as opposed to multiple exposures
may permit portable, spectral tomosynthesis.
Methods or Background: We acquired 61 projection images of a chest phantom using the Reveal 35C x-ray detector. The x-ray
source was linearly translated parallel to the phantom’s longitudinal axis to produce an angular sweep of 30 degrees. A a standard
digital radiograph (DR) and two DE images, bone and soft-tissue, were calculated from each triple-layer projection image. A variant of
logarithmic subtraction was used for the DE images, using the same parameters for each projection. We reconstructed these three
datasets using the same model-based conjugate gradient algorithm with quadratic regularisation and a 5mm slice thickness.
Results or Findings: Three sets of tomographic slice data were reconstructed: DR, bone, and soft-tissue. Structures in these images
were successfully separated by slice (localisation) and DE image (material identification). The DE data used the same reconstruction
algorithm as the DR data, indicating that tomosynthesis can be readily applied to DE images.
Conclusion: This proof-of-concept study demonstrates the feasibility of DE tomosynthesis using a triple-layer detector. Future work
will explore dose requirements, sensitivity to various pathologies, improved reconstruction techniques, and application in a portable
system.
358
RC 907 - Bladder imaging
Categories: Education, Genitourinary, Imaging Methods, Oncologic Imaging, Research

CME Credits: 1
Moderator:

Conventional imaging strategy (US-CTU) (15 min)

Alberto Hebert Vargas; New York / United States
1. To learn about the clinical indications for conventional imaging.

2. To learn what to look for with conventional imaging.
3. To understand the limitations of conventional imaging.
How to stage bladder cancer with VI-RADS? (15 min)

Valdair Francisco Muglia; Ribeirao Preto / Brazil
1. To learn about the clinical indications for the use of VI-RADS scoring in different clinical settings.
2. To learn how to appropriately acquire and report MRI images using VI-RADS for staging.
3. To understand the clinical implications.
Assessment of response to therapy using MRI (15 min)

1. To learn about the clinical indications for the use of MRI.

2. To learn how to appropriately acquire and report MRI images according to new scoring.
3. To understand the clinical implications of imaging.
Panel discussion: What is the role of imaging in the haematuria pathway? (10 min)
359
RPS 901 - Inflammatory bowel disease (IBD): an ongoing challenge in imaging
Categories: GI Tract, Imaging Methods, Research

CME Credits: 1
Moderator:
Isabelle De Kock; Ghent / Belgium
Multiparametric MRI in the assessment of CD transmural activity: a cluster analysis (7 min)

Francesca Maccioni; Rome / Italy
Author Block: F. Maccioni, L. Busato, A. Valenti, S. Cardaccio, A. Longhi, M. Barletta, S. Oliva, M. Aloi, C. Catalano; Rome/IT
Purpose: Several MRI parameters have been validated as expressions of Crohn's disease activity, using colonoscopy as the gold
standard (GS). However, the only reliable GS for Crohn's disease activity is the pathological specimen, which is rarely available. This
study was conducted to select the MRI parameters most indicative of disease activity, compared with several clinical reference GSs,
but also independently of a GS, through a cluster analysis.
Methods or Background: We retrospectively analysed 60 paediatric patients with Crohn's disease at onset and at follow up, who
underwent MREnterography , ileo-colonoscopy and clinical-laboratory examinations in a 30-day time interval. The sixteen different
MRI parameters most indicative of intestinal inflammation were intracorrelated and correlated with the Simple Endoscopic Score for
Crohn's Disease, with the pCDActivity score, foecal calprotectin and C-reactive protein. Statistical analysis performed included cluster
analysis and Cramer's analysis.
Results or Findings: Significant results were obtained using cluster analysis, which identified two clusters with severe (Group 1) and
mild-to-moderate (Group 2) disease. Eight out of sixteen MRI parameters varied significantly within the clusters: the length, DWI, T2
fat suppressed wall signal, T2 fat suppressed fat signal, arterial-phase contrast enhancement, venous-phase contrast enhancement,
late-phase contrast enhancement, stratified wall enhancement, and comb sign (p<0.001). No significant association was found
between MRI, clinical, laboratory or endoscopic parameters, except a mild association between foecal calprotectin and arterial,
venous
and late-phase contrast enhancement.
Conclusion: The cluster analysis identified eight MRI parameters that significantly increased in mild to severe Crohn’s disease. By
grading these nine parameters, an effective non-invasive activity score for disease monitoring can be developed and validated in
paediatric patients.
Limitations: This is a single-centre retrospective analysis. A deeper analysis of the missed correlation between MRI, clinical, and
endoscopic parameters is required.
Evaluation of body composition and bone density in human adults with inflammatory bowel diseases (7 min)
Jessica Ritter; Munich / Germany
360
Author Block: J. Ritter1, M. Nickel1, M. El Husseini1, P. H. Gouder1, F. Nensa2, J. S. Kirschke1, R. Braren1; 1Munich/DE, 2Essen/DE
Purpose: Chronic inflammatory bowel diseases such as ulcerative colitis (UC) and Crohn's disease (CD) are associated with
malabsorption disorders. These carry the potential to alter body parameters such as bone density and fat-muscle composition. The
aim of this study was to detect and quantify potential changes in the age- and gender-specific parameters of bone density and body
composition.
Methods or Background: A total of 433 patients (n=153 UC, n=280 CD) with chronic inflammatory bowel disease were investigated
retrospectively. The patient groups studied were age-correlated against each other and a reference cohort (n=811). All patients
received a CT scan within the clinical setting. Bone marrow density (BMD) values were extracted using an custom-made segmentation
and classification algorithm. Body tissue composition was assessed in a subcohort of 116 patients (35 UC, 81 CD) using a deep
learning–based volumetric tissue classification system.
Results or Findings: Significantly reduced BMD compared to a reference cohort was seen in almost all age and gender groups in
both UC and CD (p-value range: 1.35E-08 to 0.03). Within the subcohort, the body composition showed significantly lower muscle
mass in male UC patients and gender-independent in patients with CD. The proportion of adipose tissue, on the other hand, did not
significantly change. In the subgroup of CD patients, significantly lower muscle mass was found with short-term cortisone use, and
among continuously-medicating female patients. However, earlier cortisone use did not lead to significant changes in muscle mass.
Conclusion: Chronic inflammatory bowel diseases are associated with loss of BMD and muscle mass. The latter are mainly observed
in patients with Crohn's disease on recent cortisone therapy.
Limitations: The limitation of the study is the limited availability of therapy data leading to a relatively small subcohort.
Ethics committee - additional information: The study was approved by the Ethics Committee of the University Hospital rechts der
Isar, Technical University of Munich (Number 87/18S).
CT-based radiomics signature of visceral adipose tissue and bowel lesions for identifying patients with Crohn’s disease
resistant to infliximab (7 min)
Yangdi Wang; Gangzhou / China
Author Block: Y. Wang1, Z. Luo2, Z. Zhou1, X. Shen1, X. Wang1, B. Huang2, S-T. Feng1, X. Li1; 1Guangzhou/CN, 2Shenzhen/CN
Purpose: The purpose of this study was to develop a radiomics model combining VAT and bowel features to improve the predictive
efficacy of infliximab (IFX) therapy on the basis of bowel model.
Methods or Background: This retrospective study included 231 CD patients (training cohort, n=112; internal validation cohort,
n=48; external validation cohort, n=71) recruited from two tertiary centres. A machine-learning VAT model and bowel model were
developed separately to identify CD patients with primary nonresponse to IFX. A comprehensive model consisting of VAT and bowel
radiomics features was further established to verify whether features extracted from VAT would improve the predictive efficacy. Area
under the curve (AUC) and decision curve analysis were used to compare the prediction performance. Clinical utility was compared
using integrated differentiation improvement (IDI).
Results or Findings: VAT model and bowel model exhibited comparable performance for identifying patients with primary
nonresponse in internal [AUC of VAT model vs bowel model, 0.737 (95% CI, 0.590-0.854) vs 0.832 (95% CI, 0.750-0.896)] and external
validation cohort [AUC of VAT model vs bowel model, 0.714 (95% CI, 0.595-0.815) vs 0.799 (95% CI, 0.687-0.885)], with relatively
good net benefit. Comprehensive models adding VAT into bowel model yielded a satisfactory predictive efficacy in both internal [AUC,
0.840 (95% CI, 0.706-0.930)] and external validation cohort [AUC,0.833 (95% CI, 0.726-0.911)], with significantly improved predictive
efficacy (IDI=4.2% and 3.7%; both P<0.05).
Conclusion: VAT has an effect on IFX treatment response and improves the performance for identification of CD patients at high risk
of primary nonresponse to IFX therapy.
Limitations: An initial limitation was the use of CTE rather than MRE to develop radiomics models. In future studies, CT-based
radiomics framework may facilitate artificial development in the field of MR through transfer learning. Besides, the radiomics
signatures extracted from single-phase images underutilised all of the information. Lastly, sample size is limited.
Funding for this study: Funding was received from the National Natural Science Foundation of China [grant numbers 82070680,
82072002, 82270693, 82271958].
Ethics committee - additional information: This study was approved by the Institutional Review Board of The First Affiliated
Hospital of Sun Yat‐sen University (No. [2020]351), and the need for informed patient consent was waived.
Integrating gut microbiota and metabolites to decode variations in magnetic resonance enterography features of
bowel damage in Crohn's disease: development of a multi-scale diagnostic model (7 min)
Lili Huang; Guangzhou / China
361
Author Block: L. Huang, R. Zhang, X. Shen, Y. Wang, X. Wang, Z. Li, Z. Li, X. Li; Guangzhou/CN
Purpose: Bowel damage (BD) significantly impacts Crohn's disease (CD) patient's prognosis, showing varied magnetic resonance
enterography (MRE) features. This study seeks to investigate the potential mechanisms underlying MRE features of BD using multi-
scale data and develop an optimal diagnostic model for BD.
Methods or Background: 230 CD patients were prospectively recruited and categorized into BD (n=103) and non-BD (n=127) using
the Lémann index. All patients underwent MRE. Faecal 16S rRNA gene sequencing, and faecal and blood metabolomics were
conducted. Relationships between MRE findings, gut microbiota, and faecal/blood metabolites were analysed using causal mediation
analysis. Diagnostic models for BD were constructed using gut microbiota, faecal metabolites, blood metabolites, and MRE findings
alone or in combination, with their performance assessed using ROC analysis.
Results or Findings: Seven MRE features, including penetration, bowel thickness, and perienteric T2WI signal, differed significantly
between BD and non-BD patients. The components of gut microbiota and faecal/blood metabolites were distinct between the two
groups. Causal mediation analysis revealed that Blautia may promote intestinal penetration accompanied by perienteric inflammation
(i.e., perienteric T2WI signal) through faecal arachidonic acid and blood ceramide (d18:2/23:1). Individual component models,
including intestinal microbiota, faecal metabolites, and blood metabolites, yielded AUCs of 0.588 (95% CI: 0.453 - 0.713), 0.671 (95%
CI: 0.538 - 0.787), 0.640 (95% CI: 0.533 - 0.737), respectively. Incorporating MRE into these factors to construct a multi-scale model
increased the AUC by 0.708 (95% CI: 0.582 - 0.823).
Conclusion: BD-induced MRE findings are partially attributable to microbial and metabolite factors, providing novel insights into the
possible mechanisms driving these MRE alterations. The inclusion of MRE enhances diagnostic performance of the multi-scale model
for BD, rendering it a promising tool for BD diagnosis.
Limitations: The diagnostic model lacks validation in an external dataset.
Ethics committee - additional information: This study was approved by the Ethics Committee of the First Affiliated Hospital, Sun
Yat-sen University (No. IIT-2021[215], Guangzhou, China).
Comparison of findings on transabdominal ultrasound and magnetic resonance enterography with macroscopic
changes on colonoscopy in patients with Crohn's disease (7 min)
Vlastimil Válek; Brno / Czechia
Author Block: V. Válek, T. Rohan, L. Prokopova, D. Bartusek, M. Smela, S. Konecny, J. Husty, V. Zboril, P. Kavrikova; Brno/CZ
Purpose: This study sought to compare the sensitivity and specificity of ultrasound and magnetic resonance (MRI) enterography in
the evaluation of inflammatory changes in Crohn's disease (CD) in patients with different macroscopic findings on colonoscopy.
Methods or Background: This retrospective study analysed 47 consecutive patients with CD who underwent contrast enhanced MRI
enterography, expert bowel ultrasound and colonoscopy within 1 month at the University Hospital between 2018 and 2022. Based on
the endoscopic findings, patients were divided into four groups: 1 (ulcers), 2 (ulcers and aphthae below 5mm), 3 (aphthae below
5mm, oedema, erythema), and 4 (aphthae below 5mm). In all patients, four segments of the colon and terminal ileum were evaluated
(together 215 segments).
The sensitivity and specificity of ultrasound and MRI in the evaluation of inflammatory changes were compared in all analysed
locations (Fisher's exaxt test). Macroscopic colonoscopic findings were the gold standard.
Results or Findings: Group 1 achieved significantly better sensitivity and specificity in the assessment of inflammatory activity in
CD than the other groups (p<0.001), with sensitivity and specificity on MRI 90% and 91% and 88% and 87% on ultrasound. The
combination of ultrasound and MRI findings increased sensitivity and specificity to 94% and 93% (p=0.02; 0.003). In the independent
colon evaluation, ultrasound achieved higher sensitivity (89%, 81%, 58%, 60%) compared with MRI (83%, 63%, 33%, 20%) in all
groups. The difference was significant in group 1 and 2 (p<0.001) and nonsignificant in groups 3 and 4 (p=0.622; 0.576).
Conclusion: A combination of MRI and bowel ultrasound is useful in the evaluation of bowel activity in patients with CD. Ultrasound
can contribute to the assessment of disease activity, especially in the colon, compared to MRI.
Limitations: The study was limited by being retrospective, and by its small patient cohort.
Ethics committee - additional information: This study was approved by the Institutional Review Board of the University Hospital,
Brno.
Time-dependent diffusion MRI for quantitative microstructural mapping of intestinal strictures in Crohn’s disease (7
min)
Xinyue Wang; Guangzhou / China
362
Author Block: X. Wang, L. Huang, X. Li; Guangzhou/CN
Purpose: Microstructural characteristics of intestinal strictures are crucial determinants impacting the selection of therapeutic
strategies and prognosis of patients with Crohn’s disease (CD). We aimed to evaluate the feasibility and efficacy of time-dependent
diffusion MRI (TD-dMRI) based microstructural mapping for noninvasive characterisation of intestinal inflammation in intestinal
strictures in CD.
Methods or Background: 46 patients with CD were prospectively enrolled between May 2023 and September 2023. TD-dMRI were
acquired with pulsed and oscillating gradient diffusion MRI sequences at an equivalent diffusion time of 4.2-12.0 msec on a 3.0-T
scanner. TD-dMRI-based microstructural parameters, including cell diameter, extracellular diffusivity, intracellular volume fraction,
cellularity, and diffusivities with different effective diffusion time, were estimated with a two-compartment model. Additionally,
conventional apparent diffusion coefficient (ADC) was also calculated for comparison. Intestinal inflammation was assessed
pathologically. The accuracy of these microstructural imaging parameters were confirmed through their correlation with
histopathologic measurements.
Results or Findings: For evaluation of intestinal inflammation, intracellular volume fraction was positively correlated with
pathological inflammation scores (r=0.52, p＜0.001). Among all parameters, intracellular volume fraction demonstrated the highest
diagnostic performance with an area under the receiver operating characteristic curve of 0.83 (95% CI: 0.71, 0.96) in discriminating
moderate-to-severe inflammation from non-to-mild inflammation, followed by cell diameter (AUC=0.59, p=0.17), cellularity
(AUC=0.51, p=0.88), extracellular diffusivity (AUC=0.48, p=0.74), diffusivity at 25 Hz (AUC=0.45, p=0.38) and 50Hz (AUC=0.40,
p=0.34), and ADC (AUC=0.24, p＜0.001). Microstructural mapping was supported by positive correlations between TD-dMRI-based and
pathologic examination-based intracellular volume fraction (r=0.80, p＜0.001).
Conclusion: TD-dMRI allows for the characterization of microstructural features of intestinal strictures in CD patients. Intracellular
volume fraction shows great potential as a promising biomarker for noninvasive detection of varying degrees of bowel inflammation.
Limitations: The study was limited by its relatively small sample size.
A rare course of fistula and location of abscess in patients with perianal fistula: a sign for Crohn’s disease? (7 min)
Sezer Nil Yılmazer Zorlu; Ankara / Turkey
Author Block: S. N. Yılmazer Zorlu, D. Kuru Öz, G. A. Erden; Ankara/TR

Purpose: The objective of this study is to compare the presence of intramural abscesses formed within the internal anal sphincter,
which is the extension of the circular muscle layer of the rectum and a distinct craniocaudal course of the perianal fistula within
external the anal sphincter (EAS) in patients with and without Crohn's disease (CD).
Methods or Background: Magnetic resonance images (MRI) with possible presence of perianal fistulas obtained between 2011 and
2023 were retrospectively rewieved. Patients with malignancy, ulcerative colitis, haematological disease, and anorectal surgery were
excluded. We divided our cohort into two groups according to medical records; those with CD and those without any underlying
aetiology. Images were evaluated for both number and type of fistula according to Park’s classification, localisation of internal orifice
,presence of course within EAS, and presence of intramural abscesses.
Results or Findings: Of the 578 patients with perianal fistula, 109 patients had CD while 469 patients did not have any known
disease and have been considered idiopathic. There were 19 patients with concomitant intramural abscess (CD:14, idiopathic:5).
There was a statistically significant difference between the groups in terms of the presence of intramural abscess(p<0.001). In CD
patients, it was statistically observed that the internal orifice was more frequently located in the distal rectum and at the 3 o'clock
position relative to the anal clock, compared to the idiopathic group(p<0.001; p=0.04 respectively). There were 60 fistulas
(idiopathic:55, CD:5) that extended along the course of the external sphincter. 66.7% of extrasphincteric fistulas were found in CD
patients, although other fistula types were primarily seen as idiopathic. Regarding the fistula localization and presence of course
within EAS, there was significant difference between the groups (p<0.001).
Conclusion: Despite the rarity, intramural abscess may be indicative of CD.
Limitations: The findings of this paper must be confirmed with studies involving larger patient numbers. Furthermore, comparisons
cannot be made with the intraoperative findings.
Ethics committee - additional information: This study was approved by the Ankara University Clinical Research Ethics
Committee.
Changes in 3D volumetry of perianal fistulas can predict the clinical outcome in perianal fistulising Crohn's disease (7
min)
Jordi Rimola Gibert; Barcelona / Spain
363
Author Block: J. Rimola Gibert1, B. Caballol1, C. Saavedra2, S. Rodriguez Gomez1, A. Fernàndez-Clotet1, M. C. Masamunt1, E. Ricart1, I.
Ordás1; 1Barcelona/ES, 2Ibague/CO
Purpose: Changes in the 3D volumetry of perianal fistulas measured by MRI could be an imaging biomarker of interest. The objective
of the study is to determine the value of volumetric changes in perianal fistulas in patients with Crohn's disease (CD) after medical
treatment to predict subsequent outcome.
Methods or Background: This was a retrospective single-center pilot study. We have included CD patients with perianal fistulas
who started biological therapy between 2012 and 2020 and have (1) both pre- and post-treatment (3-18 months) pelvic MRI and (2)
follow-up after post-treatment MRI of at least 2 to 5 years. According to the last visit, patients were categorised as clinically active or
in remission (absence of suppuration). Using specific software we calculated the 3D volumetry of the fistulas, the active component
(hypersignal on T2) and the fibrotic component (hyposignal on T2). We compared MRI volumetric changes between both MRIs and
calculated the value of MRI volumetric changes in predicting remission using a multivariate stepwise logistic regression analysis.
Results or Findings: Among the 24 patients included, 13 were in perianal remission in the last follow-up. We determined that the %
of change of % of active component as predictor of clinical remission (OR 1.06 [1.02-1.11] P=0.008). The logistic regression analyses
indicated that the AUC of the ROC curve was 85,3% (p<0.0001). Our results show that a reduction of ≥16% of the % active
component of the fistula has a sensitivity of 84.6 and specificity of 81% to predict clinical remission.
Conclusion: Changes in the 3D volumetry of perianal fistulas on MRI have value in predicting clinical evolution. The decrease in the
active component could be a therapeutic target.
Limitations: The low number of patients and the follow-up range are the main limitations of this study.
Ethics committee - additional information: This study was approved with the local code HCB/2021/0380.
364
E³ 24B - Hybrid imaging in oncology (part 2): a universe beyond FDG!
Categories: Hybrid Imaging, Nuclear Medicine, Oncologic Imaging

ETC Level: LEVEL II
CME Credits: 1
Moderator:

PSMA in prostate cancer management (12 min)

Alexander Walter Sauter; Baden / Switzerland
1. To know the appropriate indications for PSMA PET.

2. To recognise common pitfalls in Ga68 & F18 PSMA PET.
3. To learn how PSMA-PET can complement prostate MRI.
4. To understand how PSMA-PET is a gatekeeper for therapy planning.
Neuroendocrine tumours diagnostics and theranostics (12 min)

Wolfgang Fendler; Essen / Germany
1. To understanding the rationale for using FDG in addition to somatostatin receptor (SSR) PET in selected cases of NET.
2. To be aware of novel approaches to neuroendocrine tumours that lack SSR including tracers for CCK-2 and GLP-1.
3. To be able to discuss molecular imaging approaches for phaeochromocytoma and paraganglioma (PCPGL) that complement MRI
and CT and when to use them.
4. To understand the importance of the universality and depth of target expression in determining suitability for radionuclide therapy
and how to measure this.
FAPI PET/CT: simply the new FDG? (12 min)

Frederik L Giesel; Heidelberg / Germany
1. To understand the pathophysiology behind FAPI PET.

2. To know about the major current indications of FAPI PET.
3. To discuss the potential of future FAPI applications in oncological imaging.
Patient's perspective on prostate cancer screening (9 min)

1. To learn that PSMA PET is to be included in metastatic prostate cancer diagnosis to better inform patients on their treatment
options and prognosis.
2. To understand that patients are through their organisations and the internet well aware of new technological progress and that
they want to benefit from these advances.
3. To understand that in rare prostate cancer types, like NET-PC, PSMA-PET can reveal the presence of the variant if compared with
"traditional" imaging.
Panel discussion: PSMA, DOTA, FAPI: theranostic rockstars on the rise? (10 min)
365
RW 9 - How to improve PET-CT reports
Categories: Nuclear Medicine, Professional Issues

CME Credits: 1
Moderator:

Tips on improving your PET-CT reports (15 min)

Thomas Wagner; London / United Kingdom
1. To write concise and clear reports answering the clinical question.

2. To tailor the report to the referring clinician and the clinical situation.
3. To summarise complex findings and create a well-structured interpretation with clear recommendations.
Short case review, interactive discussion and critiquing of reports (40 min)
Thomas Wagner; London / United Kingdom
1. To critique reports and suggest ways of improving them.

2. To discuss various ways to adapt reports to the clinical context, the findings, and the referring team.
366
CUBE 10 - Treatment of recurrent malignant stenosis of the bile duct bifurcation: role of the
radiographer

Treatment of recurrent malignant stenosis of the bile duct bifurcation: role of the radiographer (30 min)
Lena Manzo; Torino / Italy
1. To outline the current role of IR in oncology.

2. To explain the key principles for caring for a patient undergoing interventional oncological procedures.
3. To discuss the role of the radiographer in the management of recurrent malignant stenosis of the bile duct.
367
MD 3 - Current controversies in imaging and treatment of cerebral ischaemia: recommendations for

Categories: Imaging Methods, Interventional Radiology, Multidisciplinary, Neuro, Research, Vascular

CME Credits: 1
Moderator:
Wim van Zwam; Maastricht / Netherlands

1. To highlight current controversies in the treatment of cerebral ischaemia.

3. To provide recommendations for clinical practice and directions for the future.
The neurologist's perspective (8 min)

Diana Aguiar De Sousa; Lisbon / Portugal
The interventional neurologist's perspective (8 min)

Marc Ribo; Barcelona / Spain
The interventional radiologist's perspective (8 min)

368
HW 10Cb - Exploring cardiomyopathies: imaging insights and reporting techniques
Categories: Cardiac, Imaging Methods, Multidisciplinary

CME Credits: 1.5
Moderator:


Guido Ligabue; Modena / Italy
1. To become familiar with typical and atypical imaging findings of most common cardiomyopathies.
2. To become familiar with clinical data and other supporting diagnostic modalities.
3. To discuss the limits and technical drawbacks of cardiac MRI.
4. To learn how to report cardiac MRI using specific templates.
369
RPS 1009 - New insights in percutaneous tumour ablation
Categories: Abdominal Viscera, Chest, Imaging Methods, Interventional Oncologic Radiology, Interventional Radiology,
Musculoskeletal
CME Credits: 1.5
Moderator:
Afshin Gangi; Strasbourg / France
Correlation between thermal dose in real-time thermometry and postprocedural ablation zone in MRI-guided
microwave ablation of liver tumours (7 min)
Author Block: O. Öcal1, S. Lentini1, O. Dietrich1, P. Bour2, T. Faller2, J. Ricke1, M. Seidensticker1; 1Munich/DE, 2Pessac/FR
Purpose: This study aims to evaluate the correlation of thermal dose volume with the ablation zone in postprocedural first-day
images. Microwave ablation is an established treatment alternative to surgical resection with similar efficacy and better tolerability in
patients with primary or secondary liver tumours with limited disease burden. Nevertheless, it is associated with around 10% local
disease recurrence, possibly due to incomplete tumour ablation. MRI guidance offers better lesion targeting with higher soft-tissue
contrast during the procedure, as well as the possibility of real-time thermometry. However, no in-vivo comparison of real-time
thermometry with the ablation zone in follow-up images has been reported.
Methods or Background: All procedures were performed in general anesthesia on a 1.5T MRI scanner. The real-time thermometry
sequence was acquired using gradient-echo EPI sequences, and thermal dose (CEM43 of 240 minutes as a threshold) maps were
created using dedicated software (Certis Solution, Certis Therapeutics, Pessac, France). Thermal dose volumes were compared with
the ablation zone in postprocedural first-day images using Pearson correlation test.
Results or Findings: 24 patients with 27 lesions were included in this study. The volume of thermal dose and ablation zone showed
a strong correlation (R=0.89, p<0.001).
Conclusion: The real-time thermal dose mapping shows a very good correlation with the ablation zone volume obtained one day
after the procedure. Real-time visualisation of inadequate ablation margins could reduce the local recurrence rates with the possibility
of reablating lesions within the same procedure.
Limitations: - Retrospective nature of the study
- Relatively small sample size
Ethics committee - additional information: This study was approved by the Ludwig Maximilian University of Munich.
Comparison of thermal ablation combined with synchronous TACE and TACE in liver metastasis of neuroendocrine
tumours of different pathologic grades and different primary sites (7 min)
Huiyi Sun; Shanghai / China
370
Author Block: H. Sun; Shanghai/CN
Purpose: The objective of this study was to compare the thermal ablation combined with synchronous TACE and TACE in patients
with liver metastasis of NETs of different pathologic grades and primary sites.
Methods or Background: A retrospective study was performed on patients with liver metastases of NETs in the interventional
therapy Department of Zhongshan Hospital of Fudan University from November 1, 2006 to July 31, 2022, who were divided into
synchronous ablation group and TACE group and subgroups according to pathological grades and primary sites. The patients were
followed up until July 31, 2023. The endpoint was progression-free survival (PFS) and overall survival (OS).
Results or Findings: A total of 86 patients were collected, including 34 patients in simultaneous ablation group and 52 patients in
TACE group, 45 patients with G2 stage, 51 patients of pNETs. The median PFS was 18.0 months in the TACE group (95% CI, 6.0-30.0
months) and 29.0 months in the synchronous ablation group (95% CI, 10.0-48.0 months), with no statistical difference (P=0.22). In
the 45 patients with G2 stage, the median OS and PFS are 47.0 and 12.0 months in the TACE group and 59.0 and 32.0 months in the
synchronous ablation group (P=0.45 0.032),with the difference of PFS was statistically significant. Of the 51 pNETs patients, the
median OS and PFS are 47.0 and 18.0 months in the TACE group and 59.0 and 34.0 months in the synchronous ablation group
(P=0.22 0.005),with the difference of PFS was statistically significant.
Conclusion: Simultaneous ablation can delay disease progression of liver metastasis of neuroendocrine tumours, and has a good
safety, especially for patients with liver metastases of neuroendocrine tumours of intermediate or low grade or pancreatic origin.
Limitations: Due to the small sample size and confounding factors, some differences are not reflected.
Funding for this study: Funding was received from the National Health Commission Capacity Building And Continuing Education
Center (GWJJ2022100303), and Fudan University Integrated Medical Engineering Program (yg2022-6).
Ethics committee - additional information: This study was approved by Zhongshan Hospital of Fudan University.
3D fusion is superior to 2D point-to-point contrast-enhanced US to evaluate the ablative margin after RFA for
hepatocellular carcinoma (7 min)
Haiyi Long; Guangzhou / China
Author Block: H. Long1, X. Zhou1, X. Zhang1, J. Ye2, T. Huang1, L. Cong3, G. Huang1, X. Xie1; 1Guangzhou/CN, 2Foshan/CN,
3
Shenzhen/CN
Purpose: The purpose of the study was to compare the 3D and 2D contrast-enhanced ultrasound (CEUS)-derived techniques in
evaluating the ablative margin (AM) after radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC).
Methods or Background: In total, 98 patients with 98 HCCs were enrolled. The 2D CEUS point-to-point imaging (2DCEUS-PI) and the
3D CEUS fusion imaging (3DCEUS-FI) were compared to distinguish an adequate AM ≥ 5 mm. Risk factors for local tumour progression
(LTP) after RFA were analysed by the Kaplan-Meier method with the log-rank test.
Results or Findings: The mean registration time of 3DCEUS-FI and 2DCEUS-PI was 5.0 and 9.3 min, respectively (p < 0.0001). The
kappa coefficient was 0.680 for agreement between 2DCEUS-PI and 3DCEUS-FI in the evaluation of AM (p < 0.0001). Tumours with AM
< 5 mm by 2D CEUS-PI were all identified as AM < 5 mm by 3D CEUS-FI. Nonetheless, 16 (26%) tumours identified as AM ≥ 5 mm by
2DCEUS-PI were re-classified as AM < 5 mm by 3DCEUS-FI. During a median follow-up time of 31.2 months (range, 3.2-66.0 months),
LTP was identified in 8 tumours. The estimated 1-/2-/3-year cumulative incidence of LTP was 4.4%, 8.1%, and 10.3%, respectively.
Higher estimated cumulative incidence of LTP was identified in tumours with AM < 5 mm by 2DCEUS-PI (at 3-year, 27.2% vs 0%; p <
0.001), and by 3DCEUS-FI (at 3-year, 20.7% vs 0%; p = 0.004).
Conclusion: 3DCEUS-FI excelled in the evaluation of AM when compared with 2DCEUS-PI. With equivalent efficacy in the prediction of
LTP, 3DCEUS-FI was superior to 2DCEUS-PI for its automatic and time-saving procedure.
Limitations: Further studies of these techniques in guiding intraoperative management for HCC are warranted to validate their value
in achieving adequate AM and lowering the LTP rate.
Funding for this study: This research was supported by the National Natural Science Foundation of China (Grants No. 92059201,
No. 81530055, and No
815014937.
Ethics committee - additional information: This prospective study was approved by the institutional review board of the First
Affiliated Hospital of Sun Yat-sen University and
informed consent was obtained from all patients.
Gradual incremental high-power radiofrequency ablation with multi-electrodes for small hepatocellular carcinoma: a
prospective study (7 min)
Sungjun Hwang; Seoul / Korea, Republic of
371
Author Block: S. Hwang, J. M. Lee; Seoul/KR
Purpose: This study prospectively assesses the local tumor progression (LTP) and intrahepatic remote recurrence (IRR) rates of
gradual, stepwise, high power RFA in treating HCCs (≤4 cm).
Methods or Background: Patients with single HCCs (≤4 cm) scheduled for treatment with gradual, stepwise, high energy RFA,
utilising a separable clustered electrode and a two-channel generator, ranging from 60 W to 200 W for each generator, between
January 2020 and July 2022, were prospectively enrolled. The ablation procedure targeted the index tumor, guided by real-time US-
CT/MR fusion imaging, and alternately delivered monopolar energy to twp of the three clustered electrodes. Primary, secondary, and
tertiary endpoints encompassed the 3-year LTP rate, IRR rate, and recurrence free survival (RFS) rate, respectively. Technical success,
complications, and cumulative incidences of LTP and IRR, along with RFS, were assessed and estimated using the Kaplan–Meier
method.
Results or Findings: Among 110 participants (83 men and 27 women, mean age: 66.4±7.6 years), 116 HCCs (mean size: 1.65±0.59
cm) were treated with no major complications. LTP and IRR were observed in four and 29 patients, respectively. At a median follow-up
of 41.0 months (range: 35.4-46.6 months), the estimated 1-year, 2-year, and 3-year cumulative incidences were as follows: LTP
(0.9%, 3.6%, 7.0%) and IRR (13.9%, 20.5%, 31.4%). The corresponding recurrence-free survival rates were: LTP (99.1%, 96.4%,
93.0%) and IRR (86.1%, 79.5%, 68.6%).
Conclusion: Gradual increment of dual switching monopolar RF applicators may help improve LTP and IRR compared to conventional
methods.
Limitations: The study was conducted in a single centre with a single-arm, which might introduce some degree of bias in the results.
Multi-center studies with a larger sample size and control groups are essential for further validating the results and making more
generalised conclusions.
Funding for this study: This study received technical support and was supported by a research grant from STARmed Co.
Ethics committee - additional information: This study was conducted ethically in accordance with the World Medical Association
Declaration of Helsinki and approved by the Institutional Review Board of Seoul National University Hopsital. All enrolled participants
provided written informed consent.
Microwave ablation for colorectal liver metastases with ultrasound fusion imaging assistance: a stratified analysis
study based on tumour size and location (7 min)
Erjiao Xu; Shenzhen / China
Author Block: E. Xu, J. Lin, H. Liu, S. Liang, R. Yan; Shenzhen/CN

Purpose: This study aimed to investigate the efficacy of microwave ablation (MWA) for colorectal liver metastases (CRLM) with
ultrasound fusion imaging assistance, and conduct stratified analysis based on tumour size and location.
Methods or Background: Patients with CRLM who underwent MWA with the assistance of ultrasound fusion imaging from February
2020 to February 2023 in our hospital were enrolled into this retrospective study. Ultrasound fusion imaging was used for detection,
guidance, monitoring and immediate evaluation throughout the MWA procedures. The technical success, technical effectiveness, local
tumour progression (LTP), intrahepatic progression and overall survival (OS) were recorded and analysed. The subgroup analysis of
efficacy of MWA for CRLM was performed according to tumour size and location.
Results or Findings: A total of 51 patients with 122 nodules were enrolled. Both technical success and effectiveness were acquired
in all nodules (122/122, 100%). In a median follow-up period of 19 months (range, 5-39 months), LTP was observed in 2.5% of the
nodules (3/122). The cumulative intrahepatic progression rates were 38.7% and 52.1% at 1 year and 2 years, respectively. The
cumulative OS rates were 90.8% and 77.5% at 1 year and 2 years, respectively. Patients were divided into subgroups according to
tumour size (≥30 mm, n=13; <30 mm, n=38) and tumour location (perivascular, n=20; non-perivascular, n=31 and subcapsular,
n=36; non-subcapsular, n=15). The cumulative intrahepatic progression rates were similar between the subgroups regarding tumour
size and perivascular location, while significantly higher in the subcapsular group than in the non-subcapsular group (p=0.021).
Conclusion: Ultrasound fusion imaging assisted-MWA exhibited satisfactory local efficacy for CRLM, especially for non-subcapsular
tumors.
Limitations: First, it was a retrospective single-arm study performed in a single centre. Second, this study merely focused on short-
term local tumors and intrahepatic effectiveness.
Funding for this study: Funding was received from the National Natural Science Foundation of China (No. 82272011) and Natural
Science Foundation of Guangdong Province (No. 2022A1515012155).
Ethics committee - additional information: This study was approved by the Eighth Affiliated Hospital of Sun Yat-Sen University.
Microwave ablation in malignant liver lesions: extracellular vesicle surface expression as prognostic parameter (7 min)
372
Author Block: T. J. Vogl1, J. Trebicka2, E. Oppermann1, L. Ysermann1, W. Gu1, A. Pascher2, B. Strücker2, W. Bechstein1, M. Juratli2;
1
Frankfurt a. Main/DE, 2Münster/DE
Purpose: The study aimed to characterise extracellular vesicles (EV) by liquid biopsy in preinterventional patients with primary and
secondary hepatic malignancies treated with microwave ablation.
Methods or Background: Blood samples of 38 HCC patients and 15 patients with hepatic metastases were collected immediately
pre- and post-interventional. Subsequently, the characterisation of 37 surface epitopes of EVs by magnetic bead-based particle
sorting and fluorescence-associated cell scanning (FACS) was conducted.
Results or Findings: The HCC-group before intervention revealed an activation of T-cell-associated EV protein expressions with
significantly increased CD40, CD86 and CD8, vs. the non-HCC group. Additionally, the pre- and post-interventional HCC group showed
an increment of the B-cell associated marker CD20 vs. the non-HCC group, respectively. The tumour cell associated surface epitopes,
CD44, CD133 and CD24 were significantly higher expressed in the pre-interventional HCC vs. non-HCC group (p=0.029, p=0.009,
p=0.004). Regarding the clinical parameters, the HCC-group showed a significantly higher expression of CD9 in patients with
recurrent HCC, nonalcoholic steatohepatitis (NASH)-related HCC and cirrhotic HCC. Furthermore, the subgroup analysis of HCC-
patients showed a significantly lower expression of CD44 in NASH-related HCC and cirrhotic HCC. Pre-interventional cytokine levels of
TH1 and Treg cells (IL2+IL17) correlated strongly with elevated CD44 levels in HCC patients.
Additionally, pre-interventional IL6 levels correlated strongly with a high expression of CD19 and CD86. Regarding the post-
interventional changes increased CD19, CD20 and CD44 levels correlated strongly with an elevated CD4/CD8 ratio and negatively
correlated with CD4/CD8 ratio changes.
Conclusion: EV surface expressions correlated with cytokine levels in pre-interventional HCC-patients showed a CD4+ TH1 response,
associated with CD44 expression.
Limitations: Small number of participants.
Ethics committee - additional information: This study was approved by the ethic committee of the Johann Wolfgang Goethe
university, Frankfurt (19-443, 10.09.2020).
Interventional treatment of unresectable pancreatic cancer: single-centre evaluation over 12 years using microwave
ablation (MWA) and transarterial chemoembolisation (TACE) (7 min)
Author Block: T. J. Vogl, R. Cojocaru, H. Adwan; Frankfurt a. Main/DE

Purpose: The aim of th e study was to retrospectively evaluate the safety and efficacy of microwave ablation (MWA) and transarterial
chemoembolisation (TACE) in the treatment of unresectable pancreatic cancer (UPC) regarding local tumour progression and
complication rates.
Methods or Background: From 2010 to 2022, a total of 150 patients with UPC underwent either combined TACE and MWA
treatment (n=67, 40 males/27 females) or TACE as monotherapy (n=83, 40 males/43 females). In the first group, 23/67 (34.3%)
patients had metastatic pancreatic cancer and 44/67 (65.7%) locally advanced pancreatic cancer. In total, 222 TACEs and 71 MWAs
were performed (average: 3.3 TACEs/patient prior to MWA. 4 patients received 2 MWAs, 63 patients 1 MWA). 18/67 (26.9%) patients
were treated with curative intent, 49/67 (73.1%) with palliative or symptomatic intent. In the second group, 45/83 (54.2%) patients
had metastatic pancreatic cancer and 38/83 (45.8%) patients locally advanced pancreatic cancer. In total, 250 TACEs were performed
(average three TACEs/patient). 3/83 (3.6%) patients were treated with curative intent, 80/83 (96.4%) with palliative/symptomatic
intent.
Results or Findings: No major complications occurred during the procedure. In 4/67 (6%) patients, mild hemorrhage was observed
after MWA without further consequences. In the first group, a follow-up was available for 47/67 patients two to six months post MWA
with PET-CT (n=1), abdominal CT (n=8) and abdominal MRI (n=38). Local tumour progression was observed in 5/47 patients (10.6%).
In the second group, 1-month MRI follow-up was available for all 83 patients. Local tumour progression was observed in 9/83 (10.8%)
patients.
Conclusion: TACE and MWA are both safe methods in the local treatment of unresectable pancreatic cancer. Combined TACE and
MWA could be a promising treatment option for patients with UPC.
Limitations: Retrospective study design.
Ethics committee - additional information: This study was approved by the ethics committee of the Johann Wolfgang Goethe
University Frankfurt.
CT-based deep-learning radiomics nomogram for the prediction of immediate response in colorectal cancer lung
metastases treated by radiofrequency ablation (7 min)
Haozhe Huang; Shanghai / China
373
Author Block: H. Huang, W. Li; Shanghai/CN
Purpose: The aim of this study was to construct deep learning radiomics nomogram to assess the instant response in lung
metastases of colorectal cancer (CRC) after radiofrequency ablation (RFA).
Methods or Background: We retrospectively included 515 lung metastases in 233 CRC patients who received RFA (412 in the
training group and 103 in the test group). Multivariable analysis was performed to identify independent risk factors for developing the
clinical model. Tumor and ablation regions of interest (ROI) were split into three spatial habitats through K-means clustering and
dilated with 5 mm and 10 mm thicknesses. Deep learning (DL) features and radiomics features including intratumour, peritumour,
and habitat were extracted from intraoperative CT data. Predictive models using common machine learning classifiers were
constructed, and a nomogram was developed by combining DL and radiomics signature with clinical factors. The performance was
primarily evaluated using the area under the receiver operating characteristics curve (AUC) via the DeLong test, calibration curves
through the Hosmer-Lemeshow test, and decision curve analysis.
Results or Findings: A total of 412 out of 515 metastases (80%) achieved complete response. Four clinical variables (cancer
antigen 19-9, simultaneous systemic treatment, site of lung metastases, and electrode type) were utilised to construct the clinical
model. The Habitat signature combined with the Peri-5 signature achieved the best radiomics signature, with AUC values of 0.994 and
0.870 in the training cohort and test cohort, respectively. The nomogram performed well on both data cohorts (0.997 and 0.909,
respectively) and outperformed the clinical model and DL signature.
Conclusion: The proposed CT-based DL radiomics nomogram can offer precise predictions and valuable assistance to physicians in
developing personalised treatment strategies.
Limitations: It was a single-centre retrospective study with a limited sample size.
Funding for this study: This work was supported by the National Natural Science Foundation of China (No. 82272095).
Ethics committee - additional information: This study was performed in line with the principles of the Declaration of Helsinki.
Approval was granted by the Ethics Committee of Fudan University Shanghai Cancer Center (Approval Number: 1612167-18).
High-power (150W) microwave ablation (MWA) is as safe and effective as 100W MWA for the treatment of lung lesions
(7 min)
Lucilla Violetta Sciacqua; Milan / Italy
Author Block: L. V. Sciacqua, A. M. Ierardi, S. Carriero, C. Lanza, G. V. D. Amato, G. Carrafiello; Milan/IT

Purpose: The objective of this study was to compare the safety and efficacy of high-power (150W) vs. standard (100W) microwave
ablation (MWA) of lung lesions.
Methods or Background: We retrospectively evaluated 28 consecutive patients (age 43-84) treated with standard (100W, 16/28
patients) or high-power (150W, 12/28 patients) MWA at our Institution from January 2021 to July 2023.
All histologic specimens were centrally revised at our institution, with 18/28 lesions being classified as primary lung malignancies and
10/28 as metastases (seven from colorectal adenocarcinoma, one from laryngeal cancer and two from hepatocellular carcinoma).
Six patients received MWA after failure of first-line treatments, including surgery and radiotherapy.
Tumour response was evaluated three months after treatment with computed tomography (CT).
Results or Findings: Mean lesions’ size was 19 mm (range 7-40 mm) with an average ablation time of 3’42’’ (range 1’30’’-6’30’’).
Technical success was defined as the adequate positioning of the treatment antenna under Cone Beam CT (CBCT) guidance and was
achieved in all patients.
Complete ablation of the target lesion was confirmed in all patients at follow-up CT evaluation.
Peri-procedural complications (i.e., pneumothorax flap) were reported in five patients, all belonging to the 100W arm, and were all
resolved by day 6 post-treatment. Four patients, equally distributed into the two arms, demonstrated modest pleural effusion at
follow-up imaging.
Conclusion: 150W MWA demonstrated overlapping safety and efficacy when compared to 100W MWA, but its shorter ablation time
can allow for sparing more healthy tissue.
Our results, if confirmed on larger cohorts and in prospective trials, might further support the role of high-power 150W MWA for the
treatment of lung lesions.
Limitations: The main limitation of our study is represented by the small size of the treatment cohort and by its retrospective
nature.
Electrochemotherapy for the treatment of metastatic bone disease (7 min)

Nicolas Papalexis; Bologna / Italy
374
Author Block: N. Papalexis, G. Peta, M. Carta, S. Quarchioni, E. Costantino, L. Campanacci, M. Miceli, G. Facchini; Bologna/IT
Purpose: The primary objective of this study was to assess the effectiveness and safety of electrochemotherapy (ECT) using
bleomycin for the treatment of bone metastases, with a focus on evaluating radiological responses, pain reduction, and improvements
in the patients' quality of life.
Methods or Background: Between 2009 and 2022, we enrolled 106 cancer patients (mean age of 61 years) who had bone
metastases. Among these patients, 64 had lesions in the upper limbs, 18 in the lower limbs, 29 in the pelvis, and 10 in the thorax and
vertebrae. The treatment involved ECT, which utilises electric pulses to enhance the local delivery of bleomycin into the tumour cells.
A bolus injection of bleomycin (15 mg/m2) was administered intravenously, followed by the application of 8 electric pulses at
1000V/cm between each pair of electrodes (performed eight minutes after the drug injection) using the Cliniporator VITAE (Igea
S.p.A., Carpi, Italy) and specialised electrodes. The procedures were conducted with the guidance of CT or fluoroscopy.
Results or Findings: Of the patients, 94 received a single course of ECT, 11 underwent two courses, three received three, and two
had four courses. All eligible patients reported a reduction in pain ranging from 30% to 100%, with an average pain relief of 57%
following local ECT treatment. Furthermore, 74% of assessable patients experienced a pain reduction of over 50% after undergoing
ECT treatment. No general complications related to the treatment were observed, although two cases of skin necrosis were
documented as local complications.
Conclusion: Our data strongly support the feasibility and safety of employing ECT in patients with bone metastatic lesions. Patients
exhibited positive radiological responses and reported enhancements in their daily activities and overall health.
Limitations: Retrospective study, lack of control group.
Ethics committee - additional information: This study was approved by the ethics committee and data protection authority by
the Institutional Review Board for retrospective studies.
Initial experience with Bleomycin Electroporation in locally advanced and metastatic tumour spread (7 min)
Irem Bayram; Halle a.d. Saale / Germany
Author Block: I. Bayram, R. Brill, E. Elolf, M. Guntau, M. Damm, W. A. A. Wohlgemuth, S. Schob, C. D. Loberg; Halle a.d. Saale/DE
Purpose: Malignant primary tumours or metastases in superficial localisation often present with a variety of local complications, for
example pain, ulceration, necrosis with infections and bleeding. The management of a malignant wound is always complex and at
some point may become futile. Bleomycin Electroporation (BEST) is a proven treatment approach for skin melanoma and superficial
vascular malformations. Based on the efficacy in context of the aforementioned, we evaluated if local tumour complications, most
importantly haemorrhage, can be controlled by BEST.
Methods or Background: 5 patients (3 female, 2 male) with locally aggressive tumours were treated with BEST. Tumour entities
included breast cancer, giant cell sarcoma, pleomorphic dermal sarcoma and a vertebral metastasis of a thoracic vertebral body from
gastric cancer. All tumours showed significant bleeding accompanied by ulcerations, superinfection and necrosis. BEST was performed
by systemic application of 25mg Bleomycin i.v. together with current applied via hexagonal electrodes, needle length 1.5 – 2.5cm in
14 – 58 applications.
Results or Findings: Bleeding was immediately stopped in all cases. Pre- and postinterventional thermography revealed significant
downcooling after the procedure caused by the vascular lock effect.
Conclusion: Bleomycin electroporation is safe and effective in treatment of local tumour complications.
Evone® flow-controlled ventilation during percutaneous interventional radiology procedures: a clinical feasibility and
safety assessment (7 min)
Genti Xhepa; Varese / Italy
375
Author Block: G. Xhepa, A. Silvani, S. Cappio, G. Raia, A. Leoncini, F. Pedersoli, A. Saportio, S. Rizzo, F. Del Grande; Lugano/CH
Purpose: Evone® is a flow-controlled ventilation (FCV) device that actively removes air from the lungs using a small-bore cuffed tube
(Tritube). This constant flow ventilation enables protective patient ventilation with minimal diaphragm movement and smooth motion
of abdominal organs.
The objective of the study is to evaluate feasibility and safety of Evone® FCV in percutaneous interventional radiology procedures.
Methods or Background: Patients who underwent percutaneous procedures between 01/01/2022 and 30/04/2023. Exclusion
criteria were age below 18 as well as contraindications for percutaneous procedures. MDMT consensus and written consent was
obtained before the procedure. The primary endpoints were safety, feasibility, and technical success, defined as the biopsy/ablation of
the target lesion. Secondary endpoints were procedure time, patient dose and hospitalisation. Adverse Events (AEs) were classified
according to the validated 2023 SIR Classification.
Results or Findings: During the period under review 40 percutaneous procedures were performed in 20 patients (13 males) under
general anaesthesia with Evone® system in 22 treatment sessions. Median age was 66.2 years (range 47–83 years). Of the lesions
found, 14 were in the kidney (9 left, 5 right), 7 were hepatic and 3 pulmonary. The median number of procedures per patient was 2
and there were 1.7 procedures per lesion (24 lesions). The procedures performed comprised 20 biopsies, 16 Microwave Ablations
(MWA), 2 Radiofrequency Ablations (RFA) and 2 Cryoablations. The technical success rate was 100%. The median time expended per
session was 83 minutes. The median DLP per session was 2294.1 mGy/cm. The AEs rate per procedure was 15%. The rate of mild AEs
was 7.5% (1 local pain, 2 PNX not treated). The rate of moderate AEs was 7.5% (2 urinary tract infection, 1 acute urinary retention,
both successfully treated). Three patients had day hospital procedures. The median duration of hospitalisation per session was 64.4
hours, with 1 patient being hospitalised for 360 hours.
Conclusion: The Evone® FCV system improves conditions in IR procedures while providing safe ventilation.
Limitations: The small number of patients as well as the heterogeneity of treated lesions and image guiding procedures were
identified as limitations.
Ethics committee - additional information: Informed consent froms were signed by all patients.
376
RPS 1014 - Optimisation of practice, care and safety: tips and tricks
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Professional Issues, Radiographers

CME Credits: 1.5
Moderators:
Tomislav Stankovic; Zagreb / Croatia
Aart J. Van Der Molen; Leiden / Netherlands
A phantom image quality study comparing ultra-low and standard dose computed tomography protocols for the
investigation of non-accidental injury (7 min)
Niamh Moore; Cork / Ireland
Author Block: L. Kingston, M. F. F. McEntee, N. Moore, R. Young, C. S. Lee, A. England; Cork/IE

Purpose: The purpose of this study was to determine if there is potential for the development and utilisation of a low dose computed
tomography (CT) protocol for the diagnosis of suspected non-accidental injury (NAI) and to establish if the suggested ultra-low dose
(ULD) CT protocol could deliver image quality (IQ) comparable to that of a standard dose (STD) CT protocol.
Methods or Background: CT images were acquired on a Revolution Apex™ scanner, using a new-born whole body phantom, and
reconstructed using deep learning software. Two protocols were compared in terms of IQ, an ULD protocol, with a dose-length product
(DLP) of 1.49mGycm2 and a STD protocol, with a DLP of 22.92mGycm2. Participants graded IQ using a modified 4-point Likert scale.
Mann-Whitney and chi-squared tests were carried out to determine if there was a significant difference in IQ between the protocols. A
t-test was performed to determine if there was a significant difference in participants level confidence in the IQ between the
protocols.
Results or Findings: 27 participants met the inclusion criteria and scored both image sets. Both the Mann-Whitney and chi-squared
test demonstrated significant difference (P ≤ 0.05) in the IQ between the protocols, for both the evaluation of cortical and trabecular
bone. T-test results demonstrated a significant difference (P ≤ 0.05) in the level of confidence amongst participants between the
protocols.
Conclusion: This phantom-based study suggests that the IQ of the STD protocol was significantly better when compared to the ULD
protocol, for the investigation of suspected NAI. Further work is required to fully optimise an ULD-CT protocol for the investigation of
NAI.
Limitations: This was a phantom-based study using images without the presence of pathology.
Ethics committee - additional information: This study was approved by the Medical School SREC, University College Cork.
Evaluation of anatomical structures by pulmonary angio-CT in patients with PTE and COVID-19 (7 min)
Ana Filipa Colucas; Olhão / Portugal
377
Author Block: A. F. Colucas, S. I. Rodrigues, L. P. V. Ribeiro, A. F. C. L. Abrantes, O. Lesyuk; Faro/PT
Purpose: The aim of this study was to identify anatomical changes in the pulmonary angio-CT associated with the clinical worsening
of COVID-19 and pulmonary thromboembolism. Results were compared between non-COVID-19 patients and COVID-19 patients, in
order to understand whether the presence of this infection interfered with the workflow at the hospital level and whether it had an
influence on changes in anatomical structures at the cardiac level.
Methods or Background: This quantitative descriptive-correlational study with a sample of 134 angio-CT exams was conducted in a
public hospital, on patients with pulmonary Thromboembolism (PTE), separated into two groups; 81 non-Covid-19 patients and 53
COVID-19 patients. Measurements were taken of the Right Ventricle (RV), Left Ventricle (LV) and Pulmonary Artery Trunk (PAT) to
assess whether there was RV dilatation (when RV/LV >1).
Results or Findings: Pulmonary angio-CT scans have increased 88% since the COVID-19 pandemic and consequently, an increase in
positive diagnoses for PTE. Regarding the results, 85.2% and 82.7% of the non-COVID-19 patients showed dilatation of the RV and
dilatation of the PAT respectively. Considering the COVID-19 group, 92.5% and 96,2% of the patients showed dilatation of the RV and
PAT, respectively. Considering the Spearman test, the correlation between "COVID-19" and the "RV dilatation" is verified.
Conclusion: Comparing the results for both groups, it was found that the group with COVID-19 presented a higher incidence of
anatomical changes at the cardiac level, and male gender has been a predictive factor for the presence of both pathologies.
Limitations: It is difficult to identify if the non-covid patients had previous infections and a bigger sample is necessary to evaluate
the changes in anatomical structures.
Ethics committee - additional information: This study was approved by the ethics committee; decision number 164/2022.
Ultra-low-dose CT protocol for left ventricular ejection fraction (LVEF) and combination with chest-abdomen-pelvis CT
(7 min)
Martin Weber Kusk; Esbjerg / Denmark
Author Block: M. W. Kusk1, S. Hess2, O. Gerke2, C. Stolzenburg Oxlund1, L. Deibjerg Kristensen1, T. E. J. Ormstrup1, J. M. Christiansen1,
S. J. Foley3; 1Esbjerg/DK, 2Odense/DK, 3Dublin/IE
Purpose: The aim of this study was to test the accuracy of ultra-low-dose CT for left ventricular ejection fraction (LVEF-CT). A
secondary objective was to examine the feasibility of combination with routine Chest-Abdomen-Pelvis (CAP) CT in oncology patients
as potential one-stop replacement for nuclear multigated (MUGA) scans to monitor chemotherapy cardiotoxicity.
Methods or Background: Patients underwent LVEF-CT with cardiac MRI as reference. In oncology patients, MUGA scans were also
performed and LVEF-CT was combined with CAP CT, with modified injection protocol. For each modality, two readers measured LVEF.
We assessed bias using a Bland-Altman analysis and correlation via Pearson correlation. Interreader agreement was measured using
ICC. ROC analysis was performed using MRI 50% LVEF cut-off for reduced LVEF and sensitivity/specificity calculated at maximum
Youden Index. We compared CAP-scans to previous scans, using ROI mesasurements, visual grading characteristics and diagnostic
acceptability scores.
Results or Findings: 77 datasets of 82 patients were suitable. Mean doses were 1.4 mSv for LVEF-CT, 5.7 mSv for MUGA and 7.4
mSv for CAP. CT-derived LVEF bias to MRI varied from 2-10 %, dependent on measurement modes. ROC AUCs exceeded 0.95 and
sensitivity/specificity of 100% were achieved with cut-offs from 53-63% LVEF. CT-derived LVEF correlated better with MRI (R=0.87 vs
0.71) and had higher ICC than MUGA (0.98 vs 0.62). CAP image quality showed decreased renal medulla/cortex discrimination, but the
number of diagnostic scans was not significantly different from previous scans. Mean examination times were 16, 28, and 32 minutes
for CT, MRI and MUGA, respectively.
Conclusion: CT LVEF measurement is feasible at a quarter of MUGA dose and can be combined with chest-abdomen-pelvis CT, for a
one-stop examination.
Limitations: Temporal changes in LVEF were not examined. MUGA subgroup had limited patients with reduced LVEF to measure
MUGA classification accuracy.
Funding for this study: The study was funded by the Esbjerg Fund (IC Møllers Fond), the research fund of the Danish Radiographers
Association (Radiograf Rådet) and the Karola Jørgensen Fund for health research at Esbjerg University Hospital.
Ethics committee - additional information: This study was approved by the regional committees on health research ethics for
Southern Denmark (approvals. no. S-20210094 and S-20220031).
Radiomic features of kidney stones with use of virtual non-contrast on photon counting CT: a phantom study (7 min)
Simon Lysdahlgaard; Esbjerg / Denmark
378
Author Block: S. Lysdahlgaard1, L. N. Hansen1, I. Riebenholt Nielsen1, B. R. Mussmann2, J. Jensen2, H. U. Jung3, M. W. Kusk1;
1
Esbjerg/DK, 2Odense/DK, 3Vejle/DK
Purpose: This study aimed to assess the accuracy of kidney stone radiomics and Hounsfield unit (HU) measurements using the
Virtual Non-contrast (VNC) technique compared to the traditional True Non-contrast (TNC).
Methods or Background: A Gammex RMI-461a head/body phantom incorporating 18 brushite and 18 uric acid kidney stones. All
stones were scanned using low-dose non-contrast computed tomography (NCCT) and standard-dose abdominal protocols. Scans were
conducted both without and with stones embedded in contrast media. Images were reconstructed using the VNC technique. Each
stone was segmented using 3D slicer and 1.150 radiomics features were extracted. T-tests were used to test for statistically
significant differences in mean feature values between stone classifications, with Bonferroni correction applied. A p-value of > 0.05
was considered significant.
Results or Findings: Radiomic features from brushite and uric acid for abdomen and NCCT protocols with contrast media yielded
286 and 285 significant values (p < 0.05), respectively. Following Bonferroni correction, these numbers were 35 and 31 values (p <
0.000044). Radiomic features for both stone types using abdomen and NCCT protocols without contrast media were 446 and 466
significant values (p < 0.05), with Bonferroni correction yielding 238 and 224 values (p < 0.000044), respectively. Comparative
analysis showed that 17 radiomic features were significant in scans in TNC, while 196 features were significant in scans in VNC.
Conclusion: This study found fewer statistically significant radiomic features between brushite and uric acid stones in VNC compared
to TNC, for both NCCT and abdomen protocols scanned on PCCT. Radiomic features extracted from different stone types further
emphasise the potential of this method in enhancing diagnostic precision.
Limitations: This study was a phantom study, with only two types of kidney stones, and an operator-dependent segmentation tool.
Ethics committee - additional information: Consent was waived by the ethics committee.
An investigation into the impact of radiation dose information on the subjective evaluation of CT image quality (7 min)
Author Block: C. S. Lee, L. Kingston, M. F. F. McEntee, R. Young, N. Moore, M. Maher, A. England; Cork/IE
Purpose: The use of medical imaging radiation exposure has dramatically increased in recent years and is continuing to do so. Ultra-
low dose examinations may combat this, but switching to ultra-low dose protocols in CT is difficult due to the potential loss of image
quality. This has been assessed within numerous studies which have found high interobserver variability. One possible reason for this
high variability is the presence of contextual bias. This study aims to investigate the presence and magnitude of bias created through
the presence of radiation dose information on images.
Methods or Background: Two CT datasets of images were created using a paediatric phantom. The phantom was scanned using a
'standard' dose protocol and an 'ultra-low' dose protocol. Participant 'radiographers' were asked to score the image quality of each
image using the 'PGMI' scoring system. Participants were requested to focus on scoring image quality for cortical and trabecular bone,
across the full skeleton. Participants were given the images either with or without the dose information present. Data was analysed
statistically.
Results or Findings: 53 participants were included. 25 completed the dataset with the dose information present, and 38 completed
the dataset without the dose information. Overall, it was found that the dataset with dose information provided received a 23% and
22% higher score for both cortical and trabecular bone image quality, respectively for the standard dose images and 7% and 4% for
those acquired using the ultra-low dose protocol.
Conclusion: The addition of radiation dose information created contextual bias in both datasets and it had a larger impact on the
standard dose CT images.
Limitations: This study was based on CT images acquired using a paediatric phantom.
Implementation and standardisation of CT optimisation tools across Europe: results from the iViolin project survey (7
min)
379
Author Block: S. J. Foley, J. McNulty, N. J. Abualroos; Dublin/IE
Purpose: To identify the current availability and implementation of CT optimisation tools during lung, colorectal, and stomach cancer
routine scanning protocols across different centres in Europe.
Methods or Background: An online survey consisting of 18 questions was disseminated to six different clinical sites across Europe
via iViolin management team in January 2023. The survey was addressed to key imaging representatives (namely a radiologist,
radiographer and medical physicist) from each of the six contributing university hospitals. The survey analysed the availability and
frequency of use of the available optimisation tools during lung, colorectal, and stomach cancer routine scanning protocols.
Results or Findings: Responses were received from all six centres in five European countries with data reported for 22 CT scanners
(Siemens: 64%, Philips: 18%, GE: 14%, Canon: 5%). Significant variations in the availability and utilisation of CT optimisation tools in
different centres were identified, with automatic exposure control and iterative reconstruction algorithms widely available and in
frequent use. The least available tools were additional filtration, high-pitch scanning, and deep-learning reconstruction. Respondents
stated that barriers to implementation include subjective local preferences.
Conclusion: Results demonstrate varying availability and use of CT optimisation tools. While surveyed centres have access to many
optimisation tools, some challenges remain in their full implementation and usage.
Limitations: Primary limitation is the limited sample size consisting of just six hospitals from five countries but convenience sampling
was used to expedite data collection.
Funding for this study: This project is co-funded under the EU4Health Programme 2021–2027 under grant agreement no.
101056832.
Ethics committee - additional information: As a low risk study which collected data on CT protocols in current use and
information was not sensitive application for full ethical review was not necessary
Evaluation of sexual dimorphism of paranasal sinus and sella turcica by computed tomography: a retrospective study
of a Portuguese population (7 min)
Rosa Ramos Gaspar; Coimbra / Portugal
Author Block: R. R. Gaspar1, M. Lopes2, T. Ribeiro2, B. M. Magalhães1, P. M. Martins2; 1Coimbra/PT, 2Aveiro/PT

Purpose: A potential for sexual dimorphism has been pointed out regarding the study of paranasal sinuses (PS) and sella turcica (ST)
in a forensic context, but results are conflicting in the literature. The aim of this study is to ascertain if PS: maxillary sinus (MS)
sphenoidal sinus (SS) and ST measurements present sexual dimorphism in a Portuguese population.
Methods or Background: This was a retrospective, observational study including 53 CT scans of the PS from Portuguese subjects
(31 males/22 females. Exclusion criteria were age below 18 years, trauma, tumours, previous surgery, and sinus dysplasia.
Measurements of length (AP), width (WI), height (HE), distance between MS (BIL), and volume of MS were obtained bilaterally; for the
SS AP, WI, and HE were also obtained. ST measures included AP diameter, depth, and length. All linear measurements were done with
RadiAnt DICOM Viewer. The volume of the MS was estimated from linear measurements. Data were analysed using SPSS-24.
Normality and homogeneity were assessed. T-student and Mann-Whitney tests were applied. Statistical significance was set at .05.
Results or Findings: Males present mean values higher than females except for measures of the sella turcica, in all instances.
Statistically significant differences between sexes were found for five of the measurements: left WI (p=.042), left HE (p=.004), left
volume (p=.033) of the MS, HE of the SS (p=.011), and ST AP Diameter (p=.016). A discriminant equation involving the measures
with normal distribution was done. The height of the left MS shows an accuracy of 66% in predicting sex.
Conclusion: The present study reveals that, at least, the height of the maxillary sinus presents sexual dimorphism. The methods
hold promise in the field of forensic sciences to characterise the Portuguese population.
Limitations: A larger sample size should be considered.
Ethics committee - additional information: In this retrospective study, the data under analysis were subject to irreversible
anonymisation and stored by the investigator in an encrypted digital format during this research.
The application of low-dose chest CT for the diagnosis and monitoring of pulmonary infections in neutropenic patients
(7 min)
380
Author Block: E. M. Agadakos, A. Zormpala, N. Zaios, C. Kapsiocha, M. Voulgarelis, N. Sypsas, L. A. Moulopoulos, V. Koutoulidis;
Athens/GR
Purpose: The aim of our study was to investigate the image quality and the diagnostic performance of low-dose Chest CT (LDCCT)
for the diagnosis and monitoring of pulmonary infections in neutropenic patients.
Methods or Background: Data from 164 consecutive neutropenic patients of a total of 256 CT examinations [149 LDCCT and 107
Standard Dose Chest CT (SDCCT)] were obtained between May 2015 and June 2019 . Examiners analysed the image quality and
evaluated the radiologic findings associated with pulmonary infections to determine differences in diagnostic performance between
the two imaging protocols.
Results or Findings: The LDCCT protocol reduced radiation dose to patients by 47%. LDCCT images compared well in terms of
image noise and image quality to SDCCT images. Detection of consolidation and ground glass opacity was significantly lower with the
LDCCT protocol (LDCCT: 27.5% and 64.4%, respectively, SDCCT: 45.8% and 82.2%, respectively) with all the respective p-values,
unadjusted and adjusted for sex, age and BMI analyses as well as the corresponding odds ratios. Similarly, LDCCT diagnostic
sensitivity rated lower for nodules ≥3mm and ground glass halo in nodule(s) but was not affected by sex, age and BMI. There were no
statistically significant differences in diagnostic performance between the two protocols for the detection of cavitation in nodules,
diffuse interlobular septal thickening, pleural effusion, pericardial effusion and lymphadenopathy.
Conclusion: LDCCT achieved a remarkable dose reduction, image quality and noise levels rated well yet underestimated four
important radiologic findings associated with pulmonary infections in neutropenic patients.
Limitations: The limitations encountered during the current study were the lack of follow-up of patient progress and the effect on the
prognosis and survival rate. Another limitation was that both data sets were obtained using only a single IR algorithm strength
setting.
Funding for this study: This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-
profit sectors.
Ethics committee - additional information: The study was compliant with the General Data Protection Regulation (679/2016 EE)
and was approved by the hospital's scientific council and ethics committee.
What radiographers need to know to identify optimised protocols for pulmonary embolism studies using computed
tomography (7 min)
Lucas Kozma; La Croix sur Lutry / Switzerland
Author Block: L. Kozma, L. Pittet, M. Champendal, L. Flaction, C. S. D. Reis; Lausanne/CH

Purpose: CTPA is the preferred method for assessing pulmonary embolism, but protocol variations result in huge radiation doses
variations. This study synthesised optimised CTPA protocols to improve radiation protection, contrast media use, and image quality
identified in the literature.
Methods or Background: Optimised CTPA protocols for adult patients were used, including special groups (pregnant women; high-
BMI individuals), with a focus on improving radiation protection; use of iodinated contrast agents; image quality, exam success rate
were identified following JBI methods. "Embase" and "PubMed" databases were searched. Two independent reviewers performed a
blinded assessment. Articles published after 2018, with over 100 participants, meeting the above criteria were included. Data was
extracted and analysed, with categorised subsets and descriptive statistics being applied.
Results or Findings: Twenty-eight articles, involving 100 to 3998 participants, revealed various optimisation techniques falling into
five categories: acquisition, injection, breathing, reconstruction parameters, adaptations for specific populations. Reducing scan
length with a 37-48% dose reduction, high-pitch scans, dose modulation (auto-mA and ODOM), lower kV (low-kV), auto-kVp, and dual-
energy methods were proposed to lower dose and enhance patient outcomes. For pregnant women, 100 kV beams were
recommended, with free breathing and an increased injection rate.
Conclusion: Optimising CTPA protocols is a complex task due to the amount of parameters that is necessary to consider. It requires
focused studies on specific protocol variables to gather reliable and applicable data. Regular protocol reviews are crucial for effective
optimisation, to include all technical updates. Radiographers are central to the optimisation process combining their technical
expertise with patient-centred care expertise.
Limitations: No experimental studies were conducted.
Ethics committee - additional information: No information was provided by the submitters.
Enhancing radiation safety for nuclear medicine radiographers: the impact of introducing the automated
radiopharmaceutical preparation and administration system (7 min)
Serena Pichierri; Milan / Italy
381
Author Block: S. Pichierri, G. R. Bonfitto, A. Roletto, L. Bombelli, E. Scaramelli, S. V. Fasulo, A. Savi, M. Olivieri, D. Catania; Milan/IT
Purpose: In Nuclear Medicine (NM), radiopharmaceutical doses have traditionally been administered in unit doses, divided and
administered manually by personnel. The recent shift in many hospitals towards automated preparation and injection systems aims to
minimize staff exposure to radiation. However, studies on their efficacy and potential benefits to radiographers are still limited. The
aim of this preliminary study is to investigate the radiation exposure of NM staff involved in automated preparation and injection of
(18)F-fluorodeoxyglucose (FDG), compared to manual administration.
Methods or Background: From April to July 2023, NM radiographers' radiation exposure was recorded with electronic dosimeters
during (18)F-FDG administration with the automated preparation and injection system and compared with the exposure during
manual preparation and administration.
Results or Findings: By analysing the radiation exposure recorded with the electronic dosimeter provided to the radiographers, it
was found that the individual whole-body radiation exposure for radiographers for each patient averaged 4,03 μSv ± 1,2 during the
use of the automated preparation and injection system and it averaged 6,93 μSv ± 1,5 during manual preparation and administration.
Whole-body exposure was significantly reduced with automated administration by 41,85% compared to the value associated with
manual administration (p < 0,05).
Conclusion: Based on preliminary results collected in our hospital, the use of an automated preparation and injection system could
greatly reduce the radiation exposure of NM radiographers involved in the administration of (18)F-FDG. The results of this study will
be used by our team to increase the use of the automated radiopharmaceutical preparation and administration system and more
deeply evaluate its potential benefits in terms of reduction of radiation exposure on the whole NM staff.
An investigation into radiographers' ability to identify developmental dysplasia of the hip (DDH) on radiographs and
factors related to observer performance (7 min)
Author Block: N. G. Carr1, A. England1, J. D. Thompson2, N. Moore1, R. Young1, M. F. F. McEntee1; 1Cork/IE, 2Barrow-in-Furness/UK
Purpose: DDH refers to hip joint dislocation or predisposition in young children. Early detection and management are essential for
effective treatment and the prevention of long-term effects. The aim of this research was to look into radiographers' abilities to detect
DDH on x-ray images and identify factors that influence their performance.
Methods or Background: At the European Congress of Radiology, radiographers were given a structured questionnaire which
included a test bank of images. There were 25 paediatric cases in the questionnaire, and participants were asked to diagnose DDH
and rate the image quality. Data were analysed to determine the sensitivity, specificity, accuracy, and area under the receiver
operating characteristic (ROC) curve.
Results or Findings: The image set was reviewed by 30 radiographers. The average sensitivity for diagnosing DDH was 56.2%, with
a specificity of 40.2%. With an area under the curve of 0.6522, the ROC curve study demonstrated slightly better performance than
chance. Collimation, lower limb position, and the absence of patient rotation were identified as key factors influencing participants'
diagnostic ability.
Conclusion: The study discovered that radiographers' ability to recognise developmental dysplasia of the hip (DDH) on radiographs
varies and is impacted by characteristics such as training, experience, and technological consideration. Additional training and
standard approaches are required to increase their diagnostic performance in DDH. DDH imaging training programmes for
radiographers can improve their skills and may lead to better patient care.
Limitations: Given the nature of the study, participants would have expected a prevalence of DDH within the image bank.
The impact and effect of imaging referral guidelines on patients and radiology services: a systematic review (7 min)
Yi Xiang Tay; Singapore / Singapore
382
Author Block: Y. X. Tay1, S. J. Foley2, R. P. Killeen2, M. E. Ong1, J. McNulty2; 1Singapore/SG, 2Dublin/IE
Purpose: The objective of this systematic review was to offer a comprehensive overview and explore the associated outcomes from
imaging referral guidelines for various key stakeholders, such as patients and radiologists.
Methods or Background: An electronic database search was conducted in Medline, Embase and Web of Science to retrieve
citations published between 2013 and 2023. The search was constructed using medical subject headings and keywords. Full-text
articles and reviews written in English were included. The quality of the included papers was assessed using the mixed methods
appraisal tool. A narrative synthesis was undertaken for the articles selected.
Results or Findings: The search yielded 4,384 records. Following abstract, full-text screening, and removal of duplication, 31
studies of varying levels of quality were included in the final analysis. Before-and-after studies, retrospective data analysis,
randomised controlled trials, interrupted time series, and cohort studies were included. Imaging referral guidelines from the American
College of Radiology were most commonly used. Clinical decision support systems were the most evaluated mode of intervention,
either integrated or standalone. Interventions showed reduced patient radiation doses and waiting times for imaging. There was a
general reduction in radiology workload and utilisation of diagnostic imaging. Low-value imaging utilisation decreased with an
increase in the appropriateness of imaging referrals and ratings and cost savings. Clinical effectiveness was maintained during the
intervention period without notable adverse consequences.
Conclusion: The use of evidence-based imaging referral guidelines is important to improve the quality of healthcare and outcomes
while reducing healthcare costs. Imaging referral guidelines could be an essential component of improving the value of radiology in
the healthcare system.
Limitations: Studies included in this analysis reflect a significant degree of heterogeneity in terms of their research design, study
population, interventions utilised, and outcomes evaluated.
Funding for this study: Funding was received from the SingHealth Talent Development Fund and via the Singapore General
Hospital Scholarship.
Ethics committee - additional information: Ethical review was not required as this study constitutes a systematic review of the
available literature.
383
HW 10Sb - Deciphering wake-up stroke: MRI and CT insights for diagnosis and treatment

CME Credits: 1

1. To explore strokes that occur during sleep and their diagnostic and therapeutic implications using MRI.
2. To learn how to analyse strokes that occur during sleep and their impact on therapeutic decisions using CT.

Aleksandra Zoran Aracki-Trenkic; Nis / Serbia
384
RPS 1005 - Harnessing AI for enhanced reporting and workflow
CME Credits: 1.5
Moderator:
Elmar Kotter; Freiburg / Germany
Does patient education level impact comprehension of radiology reports: can AI level the playing field? (7 min)
Mohammed Bilal Aziz; Blackburn / United Kingdom
Author Block: M. B. Aziz1, R. Husam Al-Deen2, M. H. Chowdhury3, M. I. K. Inayat2, H. Ahmed2, B. Syed2, H. M. Khan2, A. Pervez2, S.
Syed2; 1Blackburn/UK, 2London/UK, 3Basildon/UK
Purpose: This study investigates the impact of artificial intelligence in improving the comprehension of radiology reports for readers
based on their educational background. Radiology reports serve as vital communication between healthcare professionals and
patients, yet their comprehensibility can vary based on the patients' educational attainment.
Methods or Background: 40 musculoskeletal MRI reports, 20 each written by consultant radiologists and generative AI (ChatGPT),
were evaluated by 10 participants across multiple educational backgrounds, from secondary education, university graduates, medical
students to qualified doctors. AI-generated reports were produced from radiologist-authored reports with a standardised prompt for
comprehension for the layman without removing detail. Comprehension was evaluated using a standardised metric in a Likert scale,
and a comparison between the reports with comprehension by educational stratification was performed.
Results or Findings: Our study showed, for radiologist-authored reports, a combined understandability and readability average of
3.02 out of 5 (higher values indicate better comprehension) for secondary school participants, 3.08 for non-medical university
graduates, 3.39 for medical students, and 3.62 for doctors: as the level of medically-related education increases, individuals achieved
a higher mean for comprehension. When compared to AI-generated reports, the mean scores for comprehension were 3.95 for
secondary school, 3.46 for university graduates, 3.71 for medical students, and 3.08 for doctors, demonstrating no pattern between
the mean scores of individuals of varying educational levels – a level playing field.
Conclusion: AI-generated reports demonstrated better comprehension among recipients across the educational spectrum,
highlighting the potential of AI to remove inaccessibility to conversations surrounding a patient's health and allow patients to make
informed medical decisions.
Limitations: The limitation of the study is the sample size consisted of 10 individuals, requiring further research into the applicability
of AI in enhancing patient access to radiology reports.
Ethics committee - additional information: Not considered research as per MRC decision tool.
ChatGPT-4 makes cardiac MRI reports easy to understand: a feasibility study (7 min)
Babak Salam; Bonn / Germany
385
Author Block: B. Salam, D. Kravchenko, L. Weinhold, S. Nowak, A. M. Sprinkart, U. I. Attenberger, D. Kütting, J. A. Luetkens, A. Isaak;
Bonn/DE
Purpose: This study aimed to evaluate the ability of Chatbot Generative Pre-trained Transformer 4 (ChatGPT-4) to transform cardiac
MRI reports into comprehensible text for medical laypersons.
Methods or Background: ChatGPT-4 was used to generate three simplified versions of 20 various cardiac MRI reports using the
same prompt (n=60). Two cardiovascular radiologists evaluated factual correctness, completeness of relevant findings, and serious
misinformation with potential harm (total ratings, n=360), while medical laypersons evaluated the understandability of both versions
(total ratings, n=200 and n=600, respectively) on a Likert scale (1 “strongly disagree”, 5 “strongly agree”). Readability grade level of
reports were measured using the Automated Readability Index. Mann-Whitney U test and Intraclass Correlation Coefficient (ICC) were
performed.
Results or Findings: ChatGPT-4 reports were generated on average in 52 sec (8–78 sec). The median reading grade levels of the
ChatGPT-4 versions were significantly lower (10 [9-12] vs 5 [4-6]; p<.001) and easier to understand for laypersons than original
reports (1 [1-1] vs 4 [4-5]; p<.001). Radiologists’ ratings of the ChatGPT-4 versions reached a median of 5 (5-5) for all three
categories with “strong agreement” for factual correctness in 92% and completeness of relevant findings in 84% of the reports. Test-
retest agreement for layperson understandability between the three simplified reports generated from the same original report was
moderate (ICC: 0.54; p<.001). Interrater agreement between radiologists was high (ICC: 0.92; p<.001).
Conclusion: ChatGPT-4 can transform complex cardiac MRI reports into more understandable, layperson-friendly language without
compromising factual correctness or completeness. That can help convey patient-relevant radiology information in an easy-to-
understand manner.
Limitations: Exploratory study design. Relatively small sample size of medical laypersons. During the questionnaire completion
process, participants may have experienced a learning effect as they read through the simplified reports, potentially influencing their
subsequent assessment of understandability.
Ethics committee - additional information: Because of the use of fictitious, unidentifiable data, approval of an institutional review
board was not required.
Cost-consequence analysis of artificial intelligence-assisted image reading in lung cancer screening (7 min)
Harriet Louise Lancaster; Groningen / Netherlands
Author Block: H. L. Lancaster1, K. Togka1, X. Pan1, M. Silva2, D. Han1, M. Oudkerk1; 1Groningen/NL, 2Parma/IT
Purpose: This modelling study aimed to estimate clinical and costs-consequences of a hypothetical AI-assisted image reading
solution in LCS in the Netherlands, compared to image reading without AI. Lung cancer screening (LCS) with LDCT detects lung cancer
earlier and leads to a reduction in lung cancer mortality by 20-24% (as shown in the NLST and NELSON RCTs). However, implementing
LCS may exacerbate the workload of radiologists. Artificial intelligence(AI) exhibits promising outcomes in the detection,
segmentation, and classification of lung nodules for LCS. Despite encouraging findings, AI assisted image reading is rarely used within
clinical practice.
Methods or Background: A cost-consequence analysis was conducted, capturing costs and effects of different LCS scenarios at
baseline from a healthcare perspective. Essential model inputs included; eligible population, screening population, image reading
time by radiologists, average weighted time, image reading time by AI, costs, screening effectiveness without AI, and discrepancies in
image reading. Control scenario: LCS without AI-assisted image reading. Two radiologists independently read all CT scans. Scenario A:
LCS with AI as a parallel-reader. AI read all CT-scans in parallel with a radiologist and the discrepant results assessed by a consensus
radiologist. Scenario B: LCS with AI as a first-reader. AI read all CT-scans first, then a radiologist confirmed positive scans and
identified false-positive classifications.
Results or Findings: LCS with AI-assisted image reading has the potential to reduce image reading costs by 37% and 73%, in
Scenario A and B respectively (total reading costs [Control: €29,676,879; Scenario A: €18,704,843; Scenario B: €8,146,251]).
Additionally, utilising AI as the first reader may reduce the radiologists’ workload.
Conclusion: The incorporation of AI-assisted image reading into LCS yields substantial reductions in costs associated with image
reading. Our findings support AI utilisation in LCS to alleviate constraints on healthcare resources.
Ethics committee - additional information: Performed using published data and expert opinions.
Deep learning assisted curation of the CANDID-III dataset with free-text reports (7 min)
Sijing Feng; Mosgiel / New Zealand
386
Author Block: S. Feng1, Q. Liu2, D. Ritchie3, B. K. J. Wilson3; 1Mosgiel/NZ, 2Auckland/NZ, 3Dunedin/NZ
Purpose: This studay aimed to curate the CANDID-III dataset, which consists of adult chest radiographs with comprehensive labels
derived from both manual and AI-assisted annotation.
Methods or Background: The CANDID-II dataset is an in-development chest radiograph dataset containing 33,486 anonymised free-
text radiological reports. CANDID-III inherited the same 45 radiological labels from the CANDID-II dataset, which were mapped to
UMLS ontology for standardisation, forming the manually labelled portion of the CANDID-III dataset. An ensemble transformer-based
label extraction model was trained and validated on the CANDID-II dataset in an 80:20 proportion. The model was then used to
automatically label the remaining CANDID-III dataset. An evaluation set of 552 reports was assessed by selected annotation team
members. Label-specific ‘mention’ F1 scores were calculated for the final ensemble model, with ‘not mentioned’ as negative and
‘indeterminate, absent, present’ as combined positive classifications.
Results or Findings: The completed CANDID-III dataset contains 322,473 images and 220,977 anonymised free-text radiological
reports from 94,210 unique patients (1:1.04 M:F ratio). AI-assisted annotation was performed on 88% of the CANDID-III dataset. For
the AI-assisted annotation portion of the CANDID-III dataset, the labelling model has a macro-F1 score of 0.88 and micro-F1 score of
0.94 across all findings. Seven labels are shared with CheXpert, with F1 scores ranging from 0.93 to 1.0. F1 scores for 30 CANDID-III
labels are above 0.90, while 8 labels range between 0.80 and 0.90.
Conclusion: The CANDID-III dataset adds numerous new clinically significant radiological annotations that are labelled to a high
accuracy. It contributes to the repertoire of publicly available chest radiograph datasets for AI development. Instructions to access the
dataset can be accessed at DOI: 10.17608/k6.auckland.22726004.
Limitations: Single institution dataset, radiologists' opinion is used as label ground truth rather than objective quantitative
measures.
Ethics committee - additional information: The study was approved by the University of Otago Human Ethics Committee.
Sixteen thousand and counting: performance of an artificial intelligence tool for identifying common pathologies on
chest radiographs and report prioritisation (7 min)
Carolyn Horst; London / United Kingdom
Author Block: C. Horst1, N. Ruechuseth1, C. Allwin1, V. Naidu1, Y. Zhu2, R. O'Shea1, C. Goncalves1, M. Narbone1, V. Goh1; 1London/UK,
2
Warwick/UK
Purpose: Chest radiograph (CXR) artificial intelligence (AI) tools may streamline reporting times and improve patient outcomes
through decision-support functionalities. However, clinical uptake has been limited and a better understanding of their accuracy at
different probability thresholds for different use cases is required.
The objective of this study is to better understand the accuracy at different probability thresholds for different AI use cases. Chest
radiograph (CXR) artificial intelligence (AI) tools may streamline reporting times and improve patient outcomes through decision-
support functionalities, however, clinical uptake has been limited.
Methods or Background: 16996 CXRs were retrospectively scored from 0-100 probability by an AI tool for 8 common pathologies.
Corresponding historical free text reports were processed by a natural language processing (NLP) model to establish ground truth.
Sensitivities and specificities for the eight findings were calculated for four positive AI score thresholds (5, 15, 30 and 45). A
composite label of 'normal' was created, where none of the individual labels were above the given probability threshold, and
sensitivities and specificities calculated.
Results or Findings: Per-finding sensitivities ranged from 0.46-0.94 and specificities from 0.53-0.99, depending on pathology and
positive threshold. For ‘normal’ CXRs, sensitivities ranged from 0.51-0.82, and specificities from 0.82-0.95.
Conclusion: Our analysis demonstrates the importance of acceptable thresholds for ‘positive’ findings for different pathologies. A
very high sensitivity may be appropriate for emergency findings e.g. pneumothorax, with a sacrifice in the specificity. Conversely, a
high specificity is preferable to triage low risk studies for reporting without missing actionable pathology. Sensitivities and specificities
at the approved threshold of 15 provide a balance of sensitivity and specificity. Our study demonstrates a flexible approach to using
AI for CXR analysis of common abnormalities, and the possibility of using the tool for identifying 'normal' radiographs for triaging
purposes.
Limitations: There may be inaccuracies in the NLP outputs that have not been controlled for, and the AI definitions of certain
pathologies may not align with radiologists’ local reporting practices. This effect was partially mitigated by the large number of
analysed cases.
Ethics committee - additional information: Retrospective study.
Glioblastoma patient monitoring using a large language model: accurate and effective summarisation of radiological
reports with GPT-4 (7 min)
Robert Angelo Terzis; Cologne / Germany
387
Author Block: R. A. Terzis, K. R. Laukamp, J-M. Werner, N. Galldiks, S. Lennartz, D. Maintz, M. Schlamann, M. Schoenfeld, J. Kottlors;
Cologne/DE
Purpose: The purpose of this study was to evaluate this possibility, particularly focusing on the capacity of LLMs to extract
meaningful information from complex textual input. Monitoring of glioblastoma patients involves multiple MRI scans, making the
process complex and resource-heavy. The advent of large language models (LLMs) presents an opportunity to facilitate physician
support by summarising radiological results and disease tracking data.
Methods or Background: We retrospectively included 225 examinations from 45 patients with biopsy-confirmed glioblastoma who
were treated at our institution. The large language model, GPT-4, was supplied with the five most recent MRI reports, including clinical
information in text form. The model's task was to synthesize the disease course, present the current state, and produce the R-code
for a suitable graphic representation. Summaries generated by GPT-4 were evaluated by two expert neuro-oncologists experts with
>20 and >8 years of experience, respectively. The evaluation categories included: (1) accuracy and logical-semantic representation,
determined by assessing four distinct items on a binomial scale of "yes" or "no"; (2) overall quality; and (3) utility in patient
monitoring and therapeutic decision-making, assessed using a 5-point Likert scale, with higher scores indicating more favourable
results.
Results or Findings: The summaries from GPT-4 matched the expert consensus on the disease progression 86.7% of the time.
GPT-4's disease course summaries received a median score of 4 in terms of quality and were perceived to have a median utility score
of 3.
Conclusion: GPT-4 effectively outlined the disease progression with significant precision, value, and relevance for clinicians. Our
results underline the potential of large language models for radiological and medical workflow optimisation.
Limitations: Limitations are reliance on text-only data, the GPT-4 model's knowledge cutoff in 2021, the "black-box" problem and a
single-center linguistic focus.
Funding for this study: This study was funded in part by the German Federal Ministry of Education and Research Network of
University Medicine 2.0 (Grant no. 01KX2121).
Ethics committee - additional information: This study received ethical approval, and informed consent was waived due to the
retrospective design. No patient-identifying information was provided to the artificial intelligence.
Initial experience of a fully digital workflow for radiological evaluation in clinical trials (7 min)
Martin Scott; Uppsala / Sweden
Author Block: M. Scott1, J. Burwick Nyberg1, T. Sundin1, M. Gelotte1, P. Eckerbom1, J. Wikström1, P. Liss1, T. Bjerner2; 1Uppsala/SE,
2
Linköping/SE
Purpose: The aim of this project was to use a research PACS and set up a fully digital workflow for radiologic evaluation in clinical
trials. Radiological evaluation of tumour response during oncologic treatment is an important task for many radiology departments.
Reporting of such evaluation has previously been and is still often documented on paper and not digitally in the PACS (Picture
Archiving and Communication System).
Methods or Background: This project was carried out at the Department of Radiology, Uppsala University Hospital. Studies to be
radiologically evaluated in a research trial were pseudonymised using RSNA CTP and exported from the clinical Philips Vue PACS to an
external server, the research PACS, consisting of another Philips Vue PACS with some adaptations to its configuration. A notification of
a new study to evaluate was then sent by email to the reading radiologist by the CTP. A structured report template was set up in the
research PACS where findings were marked and hyperlinks were included. The report can then be read, and the hyperlinks used by
the oncologist in the Philips VueMotion web-interface of the research PACS. The report also includes graphs visualising lesion
development over time.
Results or Findings: Since the start of the digital workflow in September 2023 it has rapidly been adopted by the research nurses
and the radiologists involved and the workflow is now considered more efficient and consistent. For the oncologist involved in the
study it is a great advantage to be able to easily see the measurements and the graphs in the web viewer. Time is saved by reduced
paper handling.
Conclusion: A digital workflow significantly improves the handling of oncology studies that includes radiological evaluation of tumour
response to treatment.
Ethics committee - additional information: This is not a clinical study.
Reshaping CT imaging workflow through intelligent AI orchestration using RAPID: radiologic automated processing for
image distribution (7 min)
René Hosch; Essen / Germany
388
Author Block: R. Hosch, V. Parmar, J. Kohnke, K. A. Borys, K. Arzideh, G. Baldini, J. Haubold, L. Umutlu, F. Nensa; Essen/DE
Purpose: The purpose of this study was to introduce RAPID, an algorithm for swiftly and automatically orchestrating images based on
detected anatomical landmarks and body regions.In the rapidly evolving medical AI field, radiologists are incorporating AI models into
clinical practice, aiming for enhanced efficiency and workflow optimisation. This necessitates the implementation of an "Orchestrator"
capable of automatically directing images to appropriate AI models without manual intervention. Existing CT solutions predominantly
rely on DICOM tags, which offer limited and often unreliable information like SeriesDescription.
Methods or Background: 13,211 abdominal and 6,789 whole-body CT scans from 20,000 patients (42.75% female) were used.
Topograms from these scans were employed for three tasks: classification (torso, head-neck, hands, legs), region detection (head,
brain, pericardium, thorax, abdomen), and organ detection (lung, heart, spine, liver, kidneys, spleen, stomach, colon, pancreas, brain,
hip). Series specific organ and body region segmentations generated using the Body and Organ Analysis algorithm (BOA) were
mapped onto topograms using DICOM geometry as ground truth. YOLOv8 models were trained for classification and object detection
and evaluated using F1-score and mean Average Precision (mAP0.5).
Results or Findings: Classification achieved a weighted F1-score of 0.92. Region detection reached 0.96 mAP, while organ detection
scored 0.94 mAP. After topogram-based robust classification and detection, orchestration rules were established to automatically
route series to suitable AI models if they met the model’s prerequisites.
Conclusion: RAPID accurately and efficiently locates body regions and organs in CT scans using topograms. These landmarks
facilitate series orchestration for AI applications. RAPID employs "deep content inspection" for precise routing decisions, prioritizing
image data over manual entered meta data.
Limitations: The trained models should be evaluated on external datasets. In addition, the number of relevant landmarks for object
detection should be extended.
Ethics committee - additional information: This study adhered to all guidelines defined by the approving institutional review
board of the investigating hospital. The Institutional Review Board waived written informed consent due to the study's retrospective
nature. Complete anonymisation of all data was performed before inclusion in the study.
A novel radiology communication tool to reduce workflow interruptions: clinical evaluation of RadConnect (7 min)
Sandra Vosbergen; Eindhoven / Netherlands
Author Block: M. Sevenster1, K. Hergaarden2, O. Hertgers2, N. Kruithof2, J. Roelofs2, S. Romeijn2, D. D. Nguyen1, S. Vosbergen1, H. J.
Lamb2; 1Amsterdam/NL, 2Leiden/NL
Purpose: The objective of this study was to test the hypothesis that a novel asynchronous communication tool (RadConnect) reduces
radiologist workflow interruptions. Effective stakeholder communication across the imaging value chain is a crucial responsibility of
radiologists. However, typically communication tools are used that were not created specifically for the unique needs of imaging. This
contributes to frequent radiologist interruptions.
Methods or Background: We conducted a difference-in-difference before-after study. Before adoption of RadConnect, technologists
used three conventional communication methods to consult radiologists (in-person, telephone, general-purpose enterprise chat
[GPEC]). After adoption, participants used RadConnect as a fourth. Technologists manually recorded every radiologist consult request
related to neuro and thorax CT scans in the 40 days before and 40 days after adoption of RadConnect. Telephone traffic volume to
section beepers was obtained from the hospital telephone system. The abdomen beeper was included as control group. The value and
usability experiences were collected through an electronic survey and structured interviews.
Results or Findings: Adoption of RadConnect resulted in 53% reduction of synchronous (in-person, telephone) consult requests:
from 6.1±4.2 per day to 2.9±2.9 (P < 0.001). There was a 77% decrease (P < 0.001) in telephone volume to the neuro and thorax
beepers, while no significant volume change was noted to the abdomen beeper (control). The positive impact of RadConnect on
workflow interruptions was not solely observed through statistical analysis, but was also confirmed through the survey (46% response
rate) and interviews.
Conclusion: RadConnect significantly reduced workflow interruptions. RadConnect differentiates from a general chat application by
role-based interaction and a prioritized worklist overview, which was valued by study participants. Future iterations of RadConnect
can potentially contribute to a more focused work environment.
Limitations: This was a single-centre study with use limited to CT scans for select sections.
Funding for this study: Funding was received from Philips.
Ethics committee - additional information: The Internal Committee for Biomedical Experiments of Royal Philips (Amsterdam, The
Netherlands) approved this study (ICBE-S-000556) and it was registered in clinicaltrials.gov (NCT05540444). Protocol review was
waived by the Medical Ethics Committee of the Leiden University Medical Center (N22.056)
Enhancing radiology workflows through efficient x-ray image-based orchestration and classification (7 min)
Judith Kohnke; Rheurdt / Germany
389
Author Block: J. Kohnke, R. Hosch, J. Haubold, V. Parmar, K. A. Borys, K. Arzideh, L. Umutlu, F. Nensa; Essen/DE
Purpose: The purpose of this study was to present a classification algorithm to classify up to 12 different X-ray procedures in
milliseconds. In radiology, the X-ray modality consistently stands out as the predominant diagnostic procedure in terms of utilization.
With the exponential growth in radiological examinations and the concurrent proliferation of artificial intelligence (AI) integrations,
there is an emergent demand for an image-based routing system. This system should be adept at automated image orchestration
while also being proficient in precise image classification, ensuring the optimisation of data quality and supporting automatic AI
workflows.
Methods or Background: An internal dataset of 15,502 x-rays encompassing various anatomical regions for this study was
collected which contains the following classes: knee (n=1,676), pelvis (n=692), foot (n=1,686), ankle (n=1,664), wrist (n=1,620),
thigh (n=1,183), hips (n=1,496), thorax (n=893), shoulder (n=1,633), thorax lying (n=639), lumbar spine (n=813), and hands
(n=1,507). Each class was split using an 80/20 approach for the initial training process. For image classification, the YoloV8 algorithm
was employed, and F1-score, sensitivity, and specificity metrics were used for evaluation.
Results or Findings: The algorithm demonstrated strong performance with an overall sensitivity of 0.997 and an overall specificity
of 0.970. In addition, the model reached an overall F1-score of 0.970, highlighting a robust classifier performance.
Conclusion: The presented algorithm shows an accurate and reliable classification performance which could benefit X-ray
orchestration and data quality in radiology.
Limitations: The trained algorithm should be expanded using more detailed classes for specific X-ray types and evaluated on
external data.
Ethics committee - additional information: This study adhered to all guidelines defined by the approving institutional review
board of the investigating hospital. The Institutional Review Board waived written informed consent due to the retrospective nature of
the study. Complete anonymisation of all data was performed before inclusion in the study.
Diagnostic accuracy and time efficiency of a novel deep learning algorithm for the assessment of intracranial
haemorrhage: first results (7 min)
Christian Booz; Frankfurt a. Main / Germany
Author Block: C. Booz1, G. M. Bucolo2, V. Koch1, L. D. Gruenewald1, L. S. Alizadeh1, A. Gökduman1, T. D'Angelo3, T. J. Vogl1, I. Yel1;
1
Frankfurt a. Main/DE, 2Barcellona Pozzo di Gotto/IT, 3Messina/IT
Purpose: The objective of the study was to evaluate the diagnostic accuracy and time efficiency of a deep learning-based pipeline
using a Dense U-net architecture for the assessment of intracranial hemorrhage (ICH) in unenhanced head CT scans.
Methods or Background: This retrospective study included 502 CT scans of 502 patients (mean age, 70 ± 13 years; 248 men and
254 women) who had undergone an unenhanced head CT scan for the assessment of ICH. All CT scans were analysed by the
algorithm and a board-certified radiologist independently for the presence of ICH. In case of ICH presence, ICH had to be defined as
intraparenchymal hemorrhage (IPH), intraventricular hemorrhage (IVH), subarachnoid hemorrhage (SAH), subdural hemorrhage (SDH)
and epidural hemorrhage (EDH). Additionally, the time until first temporary diagnosis of ICH was measured. Three board-certified
radiologists analysed the CT scans in consensus reading sessions to establish the standard of reference for hemorrhage presence and
classification.
Results or Findings: The reference standard revealed a total of 554 different ICH presences (IPH, n=172; IVH, n=26; SAH, n=163;
SDH, n=178; EDH, n=15). The algorithm showed a high diagnostic accuracy for the assessment of ICH with a sensitivity of 92%,
specificity of 95% and an accuracy of 93%. Concerning the most frequently present different ICH types in this study, the sensitivity
was 92%, 93% and 93% (IPH, SAH and SDH, respectively), and the specificity was 95%, 96% and 95% (IPH, SAH and SDH,
respectively). Regarding analysis time, the algorithm was significantly faster compared to the temporary report of the assigned
radiologist (15 ± 2 s vs 277 ± 14 s, p < 0.001).
Conclusion: A novel deep learning algorithm provides high diagnostic accuracy combined with time efficiency for the identification
and classification of ICH in unenhanced CT scans.
Limitations: Single-centre retrospective study.
Ethics committee - additional information: This study was approved by the local IRB.
390
CUBE 11 - The era of the robots

Moderator:

The era of the robots (28 min)

1. To offer an overview of current robotic technology interventions in oncology.
2. To illustrate the benefits of using robotic assistance.
3. To describe the outcomes of robotic interventions.
391
RPS 1001 - Characterisation of upper GI-tract malignancies and beyond
Categories: GI Tract, Imaging Methods, Multidisciplinary, Oncologic Imaging

CME Credits: 1.5
Moderator:
Dietmar Tamandl; Vienna / Austria
A proposal for a new fluoroscopy severity assessment in achalasia: the IVA score (7 min)
Giovanni Fontanella; Avellino / Italy
Author Block: G. Fontanella1, S. Borrelli2, B. Brogna3; 1Benevento/IT, 2Mirabella Eclano/IT, 3Avellino/IT

Purpose: The aim of the study was to establish a quali-quantative fluoroscopic severity assessment for achalasia, comparable to the
equivalent clinical Eckhard scoring system.
Methods or Background: From September 2020 to August 2022, 69 patients already diagnosed with achalasia and scored with ESS,
were recruited and evaluated with our fluoroscopy barium protocol. The AP sequence was used to divide the esophagus into nine
segments, according to Brombart's classic description, plus the gastro-esophageal junction. Three scoring items were chosen, after a
profiling study of achalasia, to depict the features, some mutually exclusive, of the three clinical subtypes: lumen dilation, stasis,
spasm. Each esophageal segment was scored for the three items (1 point, item present; 0 points, no item), the IVA score was
calculated by summing points up until a maximum of 20 points for each subtype was reached. IVA scores were then normalised on a
0-12 scale to be compared to ESS.
Results or Findings: IVA and ESS scores were not found to be statistically diverging in 60/69 patients (86.95%, p=0.05). IVA scores
were diverging and superior to ESS in 6/69 patients (8.69%); in this group of patients, the ESS' 'chest pain'/'weight loss' items were
found to be biasing factors. IVA scores were inferior to ESS in just 3/69 patients (4.34%). In all the patients with a diverging IVA score
(9/9), ESS scores were found to be lower than 6/12.
Conclusion: IVA score was found to be consistent and compatibile with ESS scores, especially in patients with moderate to severe
achalasia. The apparent superiority of imaging scores in a small proportion of patients might instead be used as a revealing tool to
call out patients in which the ESS does not reflect the disease's severity, due to internal biases.
Limitations: This was a monocentric study and there was a limited number of patients.
Ethics committee - additional information: This study was approved by the FBF BN ethical committee.
Early tumour shrinkage as a predictor of survival in patients with advanced esophageal squamous cell carcinoma
treated with first-line checkpoint inhibitors (7 min)
Mu Wan Ling; Zhengzhou / China
392
Author Block: M. W. Ling, Y. Zhou, J. Gao; Zhengzhou/CN
Purpose: Early tumour shrinkage (ETS) is a promising parameter for assessing treatment responses. Our study hypothesised that an
ETS with an optimal cut-off value was an imaging biomarker for advanced esophageal squamous cell carcinoma (ESCC) treated with
first-line immunotherapy.
Methods or Background: We retrospectively enrolled 129 patients with unresectable locally advanced ESCC treated with first-line
immunotherapy between 2019 and 2021. ETS was defined as the relative change in the sum of the target lesions' longest diameters
at the first evaluation compared with that at baseline. Multivariate analyses were conducted to identify the significant prognostic
variables in progression-free survival (PFS) and overall survival (OS).
Results or Findings: The median value of ETS was 29.5%. An ETS with a 10% cut-off value was statistically significantly associated
with PFS in the univariate analysis (hazard ratio [HR]: 2.26; 95% confidence interval [95% CI]: 1.21-4.24; p = 0.009). Besides, in the
univariate analysis, the longest diameter, maximum short diameter, central necrosis on enhanced computed tomography, enhanced
pattern, and ETS values were statistically significant predictive factors for OS. In the multivariate analysis, ETS with a 10% cut-off
value was an independent predictive factor for OS (HR: 3.14; 95% CI: 1.45-6.83; p = 0.004).
Conclusion: ETS is associated with survival outcomes in patients with advanced ESCC treated with immunotherapy. Early tumour
size shrinkage of at least 10% can be regarded as a promising biomarker predictor for PFS and OS.
Limitations: First, there might have been some bias. Second, the ETS was not sufficient to fully reflect the dynamic process of
tumour response to treatment, including the combination of tumour response and treatment time points. The tumour size reduction
pattern, morphological response based on imaging, and other factors might be related to the patient's prognosis.
Funding for this study: Funding was provided by the Youth Project of the Henan Natural Science Foundation (no. 212300410271)
and the Youth Project in Medical Science and Technology of Henan Province (no. SBGJ202003016).
Ethics committee - additional information: Ethics committee approval was provided by the first affiliated hospital of Zhengzhou
University.
Radiographic characterisation of enlarged lymph nodes in locally advanced esophageal squamous cell carcinoma
(ESCC) patients treated with neoadjuvant immunotherapy (7 min)
Chenyi Xie; Guangzhou / China
Author Block: C. Xie1, Z. Ning1, Y. Gao2, J. Chen1, Q. Zhang1, Z. Liu1, Y. Hu1; 1Guangzhou/CN, 2Jinan/CN
Purpose: Neoadjuvant immunotherapy has emerged as a promising therapeutic approach for locally advanced ESCC. This study aims
to systematically evaluate suspicion concerning lymph node involvement by synthesising well established imaging observations
based on routine workup and reassessment CT images.
Methods or Background: In our retrospective study, we enrolled a total of 100 patients diagnosed with locally advanced ESCC. CT
images were meticulously evaluated by experienced radiologists, and enlarged LN were further analysed for size measurement (long-
axis diameters, short-axis diameters, and corresponding ratios) and morphological appearances (shape, enhancement pattern, the
completeness of the extracapsular border, the presence of fatty hilum, necrosis, fusion, and conglomeration) for prediction of lymph
node metastasis (LNM). Subsequently, we compared clinicopathological characteristics between enlarged LN groups presenting
different radiological features for exploration of their potential biological significance.
Results or Findings: Neoadjuvant immunotherapy yielded a radiologically enlarged LN in 27/100 (27%) of patients. We observed a
significant increase in the size of the long-axis diameter (63% vs 39%, p = 0.041) in pathologically negative enlarged LN than
malignant LN. A cut-off value of >40% change in lymph node long-axis size was established as a statistically significant discriminator
of LNM (AUC = 0.747). Reactive enlargements of LNs are more likely to occur in patients with favourable prognostic and predicative
biomarkers (PD-L1 positivity, lower expression of ECD, EGFR, and CD44V6) .
Conclusion: Our study has contributed to our understanding of the correlation between CT-based morphological features and
enlargements of LNs, potentially addressing a gap in current knowledge. Our findings indicate the observed radiological appearance
of cancer progression in lymph nodes may actually be attributed to a special response pattern following neoadjuvant immunotherapy.
This step is crucial for ensuring the development of an appropriate clinical treatment plan.
Limitations: The sample size was limited.
The significance of small lymph nodes on CT for poorly cohesive advanced gastric cancer (7 min)
Gyeongme Cho; Seoul / Korea, Republic of
393
Author Block: G. Cho; Seoul/KR
Purpose: The purpose of this study was to compare the size of metastatic lymph nodes on preoperative CT in poorly cohesive
advanced gastric cancer with other types of advanced gastric cancers.
Methods or Background: Literature concerning poorly cohesive gastric cancer is scarce and lymph node metastasis is a well known
prognostic factor in gastric cancer patients. So, herein, we evaluated poorly cohesive AGC with emphasis on the difference in size of
metastatic LNs, compared with other types of AGC.
We retrospectively included AGC patients, who underwent gastric cancer surgery at Chung-Ang University Hospital from February
2018 to May 2023. Two abdominal radiologists independently reviewed abdomen CT scans and evaluated the largest size of visible
LNs on each gastric LN stations. Measurable LNs (Defined as >3 mm SD) were matched with full pathology report on electronic
database records and the metastasis status was determined. We evaluated the size difference of metastatic LNs using independent t-
test.
Results or Findings: A total of 140 patients (median age, 67 years [IQR, 58-77 years]; 92 men) were evaluated. Poorly cohesive
cancer was present in 27 patients (19.7%). Total number of the measurable LNs was 425. 216 out of 425 LNs were matched as
malignant based on pathology report. The size of metastatic LNs in poorly cohesive AGC was significantly smaller than metastatic LNs
in other types of AGC (p<0.001, geometric mean size: 5.774 vs 7.613).
Conclusion: The size of metastatic LNs in poorly cohesive AGC is significantly smaller than metastatic LNs in other types of AGC.
Therefore, lowering the size threshold to 5 mm could improve preoperative CT evaluation of metastatic LNs in poorly cohesive AGC.
Limitations: The study is a retrospective single-centre study. We evaluated only the largest metastatic LNs per station rather than
each and every visible lymph node.
Ethics committee - additional information: This study was approved by the Chung Ang University Hostpital International Review
Board.
Quantitative DCE parameters combined with apparent diffusion coefficient to evaluate molecular typing of gastric
cancer (7 min)
Yan Liangliang; Zhengzhou / China
Author Block: Y. Liangliang, J. Li, J. Qu; Zhengzhou/CN

Purpose: To explore the feasibility of quantitative DCE parameters derived from DCE-MRI combined with ADC values to predict the
molecular typing of gastric cancer (GC).
Methods or Background: 43 patients were enrolled in this retrospective study. Mean values, 10th, 25th, 50th, 75th, 90th percentile
values of quantitative DCE parameters (Ktrans, Kep, Ve) and ADC values were manually extracted. The specimens were performed
with five biomarkers, including EBER in situ hybridisation, MLH1, PMS2, E-Cadherin and P53. According to the different expression
results, they were divided into five molecular types. The aberrant E-cadherin group and aberrant P53 group were combined into a
high-grade malignant group, and the other three groups were combined into a low-grade malignant group. The quantitative DCE
parameters or ADC values between two malignant groups were compared.
Results or Findings: There were significant differences in Ktrans mean, Ktrans 25%, Ktrans 50%, Ktrans 75%, Ktrans 90%, Kep
mean, Kep 10%, Kep 25%, Kep 50%, Kep 75%, Kep 90%, Ve 10%, Ve 25% and ADC between two malignant groups of GC (p = 0.006,
0.044, 0.007, 0.007, 0.009, 0.004, 0.032, 0.024, 0.004, 0.005, 0.016, 0.021, 0.028, 0.018, respectively). Ktrans 90% and ADC were
independent risk factors for predicting the two malignant groups. The AUC values, sensitivity, specificity, positive predictive value,
and negative predictive value of Ktrans 90%, ADC, Ktrans 90% + ADC were 0.720, 0.688, 0.798; 44.4%, 88.9%, 88.9%; 88.0%, 52.0%,
60.0%; 72.7%, 57.1%, 61.5%; 68.7%, 86.7%, 88.2%, respectively.
Conclusion: Quantitative DCE parameters combined with ADC values can assess different malignant groups based on molecular
types of GC, which may provide new directions for the evaluation of GC.
Limitations: Firstly, the sample size is too small. Secondly, this study did not classify molecularly type of GC according to TCGA.
Ethics committee - additional information: This study was approved by the Ethics Review Committee of Henan cancer hospital.
Differential analysis of apparent diffusion coefficient values based on primary tumour and perigastric lymph node for
distinguishing N stages of gastric cancer (7 min)
Yan Liangliang; Zhengzhou / China
394
Author Block: Y. Liangliang, J. Li, J. Qu; Zhengzhou/CN
Purpose: The aim of this study was to explore the differences in ADC values based on the primary tumour and perigastric lymph
nodes for distinguishing N stages of gastric cancer (GC).
Methods or Background: 160 GC patients from April 2019 to April 2022 were enrolled in this retrospective study. ADC values and
relative ADC values (ADCT, rADCT, ADCLN, rADCLN) based on primary tumours and perigastric first station lymph nodes were
measured separately. The ANOVA or Kruskal Wallis test was used to compare differences in ADCT, rADCT, ADCLN, rADCLN values
between different N stages. ROC curves were used to determine the optimal parameters and diagnostic efficacy for predicting N0 + 1
and N2 + 3 stages.
Results or Findings: There were significant differences in ADCT, rADCT, ADCLN, rADCLN values to distinguish different N stages of
GC (p values of <0.001, 0.023, <0.001, <0.001, respectively). The AUC values, sensitivity, specificity, positive predictive value, and
negative predictive value of ADCT, rADCT, ADCLN, and rADCLN values in predicting N0 + 1 and N2 + 3 stages of GC were 0.714,
0.632, 0.739, 0.743; 87.7%, 46.2%, 52.3%, 83.1%; 49.5%, 72.6%, 80.0%, 57.9%; 54.3%, 53.6%, 64.2%, 57.4%; 85.5%, 66.3%, 71.0%,
83.3%, respectively.
Conclusion: Both ADC values and rADC values can be used to distinguish different N stages of GC, and rADC values based on
perigastric lymph nodes have the highest diagnostic efficacy in predicting N0 + 1 and N2 + 3 stages of GC.
Limitations: Firstly, this study is a single-centre retrospective study, which may bring some bias to the results. Secondly, as this
study mainly evaluates whether there are differences between ADCT and ADCLN values in distinguishing N stages of GC, we did not
evaluate T stage and clinical stage, which will be our future research direction.
Ethics committee - additional information: This study was approved by the Ethics Review Committee of Henan cancer hospital.
Predioction of the signet ring cells percentage in diffuse-type gastric carcinoma: comparison between morphological
CT analysis and radiomics (7 min)
Iacopo Capitoni; Siena / Italy
Author Block: I. Capitoni, G. Bagnacci, L. Ferradini, N. Di Meglio, A. Perrella, L. Volterrani, M. A. Mazzei; Siena/IT
Purpose: Recent evidence has shown that in patients with diffuse-type gastric carcinoma (GC), a high percentage of signet ring cells
(SRC) represents a positive prognostic factor (reduced 5-year mortality risk by four times).
The aim of this study was therefore to predict the percentage of SRC in GC through radiomics and morphological criteria applied to
staging CT scans.
Methods or Background: 44 patients were selected based on the following inclusion criteria: (1) re-evaluation of the percentage of
SRC through histopathological examination and (2) good quality of preoperative staging CT scans.
The 10% cut-off of SRC was considered to classify patients into PC-NOS (pure poorly cohesive) and PC-SRC (with signet ring cell
component). The CT images were assessed by two readers with different levels of experience, who evaluated 12 dichotomous criteria
and post-contrast enhancement: the readers also segmented the tumours, and a radiomic analysis was performed using specific
software (pyradiomics).
Results or Findings: Among the two groups (SRC<10% and SRC>10%), significant differences were found in the distribution of six
dichotomous variables and in the post-contrast behavior (ΔHU late-portal 21.5±24.9 VS -26.1±27.2, p=0.001). From the radiomic
analysis, 81 out of 106 variables were found to be reproducible between the two readers (ICC>0.75), and the developed model,
including three variables, had an area under the curve of 0.726.
Conclusion: Patients with different percentages of ring-shaped cells can be accurately identified on CT scans. A larger patient cohort
is desirable in the future, considering the prognostic impact of this information, which can only be extracted from a comprehensive
analysis of the primary neoplasm.
Limitations: This was a monocentric study.
Ethics committee - additional information: Informed consent was obtained from all individual participants included in the study.
Mystery of lymphoma in gastrointestinal tract (7 min)

Farwa Mohsin; Karachi / Pakistan
395
Author Block: F. Mohsin; Karachi/PK
Purpose: It is important from a prognostic point of view to distinguish primary GI lymphoma from secondary extra-nodal involvement
by disseminated nodal disease. Although it has a vast variety of imaging features and exact diagnosis relies on histopathologic
analysis, certain imaging appearances like a bulky mass with diffuse infiltration, fat planes preservation, no signs of obstruction,
multiple site involvement, associated bulky lymphadenopathy can strongly suggest the diagnosis.
Methods or Background: Primary gastrointestinal (GI) lymphoma is an uncommon disease but is the most frequently occurring
extra-nodal lymphoma (10–30%) and is almost exclusively of non-Hodgkin type. The stomach, small bowel, large bowel, and
oesophagus are involved in decreasing order of frequency. Risk factors for the development of gastrointestinal lymphoma include
H/pylori, immunosuppression, Celiac disease, IBD and HIV.
Results or Findings: In the stomach, they typically demonstrate marked thickening of the wall with homogeneous enhancement
and submucosal spread.
In the small bowel, distal ileum is classically the most common site because of the greater amount of lymphoid tissue in this portion of
the bowel. Typical presentation is a thick walled infiltrating mass with aneurysmal dilatation without obstruction which occurs due to
tumour invasion into the muscularis propria causing destruction of its intramural autonomic nerve plexus.
In the large bowel, they usually appear as bulky polypoid masses on CT, larger than the ones that can be encountered in colorectal
adenocarcinomas and may extend beyond the bowel wall, thus presenting as enormous peritoneal masses, that can also be
cavitated. Colonic lymphoma usually involves a longer segment, moreover, usually located near the ileocaecal valve and grows into
the terminal ileum, not invading or obstructing neighbouring viscera.
Conclusion: It is important to decipher the mystery of primary GI lymphomas on the ground of radiology for better management.
Preoperative prediction of gastric adenocarcinoma lymph node metastasis using DLCT-derived first order features of
effective atomic number maps (7 min)
Ma Luo; Guangzhou / China
Author Block: M. Luo1, Y. Liao1, W. Deng2, C. Xie1; 1Guangzhou/CN, 2Shanghai/CN

Purpose: Accurate diagnosis of lymph node (LN) involvement in gastric cancer (GC) remains a challenge with conventional CT. This
prospective study assesses the utility of first order features derived from effective atomic number (Zeff) maps obtained from dual-
layer spectral detector CT (DLCT) in preoperative LN metastasis diagnosis, compared to conventional CT images.
Methods or Background: From July 2021 to February 2022, we enrolled patients with adenocarcinoma scheduled for gastrectomy.
Regional LNs were identified on preoperative DLCT and matched during surgery using carbon nanoparticle solution, guided by
anatomic landmarks on preoperative images. We computed 18 first-order features from Zeff maps and conventional CT images across
unenhanced, arterial, venous, and delay phases. DLCT quantitative features were assessed using Wilcox rank-sum tests and
backward stepwise regression. Logistic regression models were constructed for predictive purposes, and Receiver Operating
Characteristic (ROC) curves were used to compare DLCT Zeff maps with conventional CT images.
Results or Findings: The study included nine patients with 49 successfully matched LNs (14 metastatic, 35 non-metastatic). In the
discrimination of metastatic status based on individual features, the highest area under the curve (AUC) of 0.776 (95% CI: 0.624 -
0.929) was obtained from the energy of the venous phase. Among the different phases, the combined model based on Zeff features
from the delay phase achieved the highest AUC of 0.855 (95% CI: 0.731 - 0.979). Compared to the conventional CT model, the
combined Zeff model exhibited a higher AUC and accuracy (0.825 vs 0.855, 80.00% vs 84.40%) for LN diagnosis.
Conclusion: Incorporating first-order parameters derived from Zeff maps from DLCT improves the preoperative diagnostic efficacy
for LN metastasis in patients with gastric adenocarcinoma.
Limitations: This was a small-sample study.
Response to neoadjuvat treatment in metastatic gastric cancer: proposal of new radiological criteria from a
prospective Italian registry (7 min)
Giulio Bagnacci; Siena / Italy
396
Author Block: G. Bagnacci1, A. Perrella1, N. Di Meglio1, L. Funicelli2, F. Pittiani3, A. Veltri4, P. Morgagni5, G. Mura6, M. A. Mazzei1;
1
Siena/IT, 2Milan/IT, 3Brescia/IT, 4Orbassano/IT, 5Forli/IT, 6Arezzo/IT
Purpose: The introduction of neoadjuvant therapy (NAT) for gastric cancer (GC) has led to the need for radiologists to assess
response to treatment using CT. The shortcomings of the RECIST 1.1 criteria in assesing NAT response for gastric cancer are well
known. Currently, prognostic predictions rely heavily on TNM staging information obtained from staging CT.
Methods or Background: A prospective Italian registry ("METAGASTRO"), focusing on patients with metastatic gastric cancer (GC),
is currently ongoing. Our subanalysis included patients who had undergone at least one cycle of neoadjuvant chemotherapy and had
both staging and restaging CT scans reviewed by experienced radiologists. We selected 124 patients from six centres affiliated with
the GIRCG (Italian Research Group for Gastric Cancer). For each patient, detailed data were collected on approximately 90 variables
related to T, N, and M parameters in both staging and restaging.
Results or Findings: As expected, patients without peritoneal involvement or with fewer than two hepatic metastases or isolated
pathological paraortic lymph nodes demonstrated significantly better survival (p=0.001). RECIST 1.1 offered no prognostic insights (p
= 0.233). New criteria, incorporating different lymph node thresholds and Peritoneal Cancer Index (PCI), resulted in effective
stratification (p <0.001). The combination of new response criteria with oligometastatic status provided optimal prognostic
stratification (p <0.001), with the most favourable group exhibiting a median survival of 41.3 months. Interestingly, non-
oligometastatic patients displaying a partial response had a prognosis similar to stable oligometastatic patients.
Conclusion: The combination of initial staging and new criteria for response provided satisfactory prognostic stratification for
patients affected by metastatic GC.
The prognostic value of CT scans can be improved significantly with the expertise of skilled radiologists.
Limitations: The small sample size, limited interreader agreement as well as slight variation in time interval between staging and
restaging CT were identified as limitations.
The prospective study was approved by the Ethical committee "Comitato Etico Regionale per la Sperimentazione
Clinica della Regione Toscana, Sezione AREA VASTA SUD EST", protocol number 13082_2018, 21/05/2018.
The utility of computed tomography features and histogram texture analysis parameters as diagnostic tool in
preoperative differentiation of high-risk gastrointestinal stromal tumours (7 min)
Milica Mitrovic; Belgrade / Serbia
Author Block: M. Mitrovic, J. Kovac, L. Lazic, A. Jankovic, D. Šaponjski, S. Milosevic, K. Ebrahimi, D. Mašulović, A. Djuric-Stefanovic;
Belgrade/RS
Purpose: The aim of the study is to determine the morphological characteristics of the tumour obtained by the analysis of the
conventional computed tomography examination and texture analysis parameters, which may be useful as imaging biomarkers for
the preoperative prediction of high-risk gastrointestinal stromal tumours.
Methods or Background: This was a prospective cohort study that was carried out in the period from 2020 to 2023. The study
included 79 patients who underwent CT examination and texture analysis, surgical resection of a lesion that was suspicious for GIST,
as well as pathohistological and immunohistochemical analysis.
Results or Findings: Textural analysis pointed out Min Norm (p=0.032) as a histogram parameter of the first order statistically
significant in the prediction of HR GIST, while Min Norm (p=0.007), Skewness (p=0.035) and Kurtosis (p=0.003) showed significance
in predicting high grades of this tumour. Univariate regression analysis identified tumour diameter, margin appearance, growth
pattern, lesion shape, structure, mucosal continuity, presence of enlarged feeding or draining vessel (EFDV) and Max Norm as
significant predictive factors for HR GIST. Multivariate regression analysis extracted interrupted mucosa (p <0.001) and presence of
EFDV (p <0.001) as independent predictive CT features for HR GIST with an AUC of 0.878 (CI: 0.797-0.959), sensitivity of 94%,
specificity of 77% and accuracy of 88% in predicting HR GIST.
Conclusion: The morphological characteristics of the tumour detected by conventional CT examination still have the greatest value
in the preoperative stratification of the metastatic risk of gastric GIST. The incorporation of texture analysis into the basic imaging
protocol may further improve the preoperative assessment of risk stratification.
Limitations: Our study did not include a follow-up of the involved patients.
Funding for this study: This research received no external funding.
Ethics committee - additional information: Our research was permitted by an Ethical Committee of the School of Medicine,
University of Belgrade, code 1322/II-6, and written informed consent was obtained from all patients.
Percutaneous transhepatic balloon-assisted (PTB-A) embolisation with ethylene-vinyl alcohol copolymer (EVOH) of
duodenal stump fistula after gastrectomy for benign and malignant disease (7 min)
Claudio Sallemi; Brescia / Italy
397
Author Block: C. Sallemi, F. Bodini, F. Rosella; Brescia/IT
Purpose: Duodenal stump fistula (DSF) is one of the most serious complications following gastrectomy, with a high risk of morbidity
and mortality and a long period of hospitalisation.
When conservative management fails, percutaneous transhepatic biliary drainage is useful to reduce duodenal pressure and fistula
output by aspirating bile. However, it needs a long healing time and often needs to be followed by other treatments, such as
percutaneous or endoscopic injection of glue.
The aim of this study was to explore the feasibility and efficacy of a novel technique of percutaneous DFS embolisation with ethylene
vinyl alcohol copolymer (EVOH) combined with an occlusion balloon in the duodenal stump for the treatment of DSF after gastrectomy
for malignant or benign disease.
Methods or Background: From 2018 to 2023, 11 consecutive patients underwent PTB-A embolisation with EVOH for the treatment
of DSF. Clinical and technical success, morbidity and mortality were analysed. Fistula recurrence was also evaluated.
Technical success was defined as the absence of contrast media extravasation from the duodenal stump at fistolography after the
embolisation. Clinical success was defined as no leakage from the percutaneous tract of the fistula 48 hours after the embolisation.
Recurrence was defined as a fistula that recurred after clinically complete healing.
Results or Findings: Technical success was achieved in all cases. Clinical success was obtained in 8/10 patients in a single
treatment. 2/10 patients were re-treated to achieve complete fistula healing. No procedure-related complications and mortality were
recorded. No relapse of fistula occurred during follow-up.
Conclusion: PTB-A embolisation with EVOH of DSF after gastrectomy is a feasible and safe procedure and seems to be effective to
achieve complete healing of the fistula.
Ethics committee - additional information: IRB approval was not waived; written informed consent for PTB-A embolisation was
obtained from patients who signed a specific institutional procedure-related consent valid for retrospective observational studies.
398
RPS 1002 - What is new in breast ultrasound?
Categories: Breast, Imaging Methods, Interventional Radiology

CME Credits: 1.5
Moderator:
Andrew Evans; Dundee / United Kingdom
False negative and false positive cases after implementation of an automated breast ultrasound system for breast
cancer screening: a retrospective study in 1219 women (7 min)
Elizabet Nikolova; Zurich / Switzerland
Author Block: E. Nikolova, J. K. Weber, G. Zanetti, J. Wieler, T. Frauenfelder, M. Marcon; Zurich/CH

Purpose: To investigate the false positive (FP) and false negative (FN) rate over three years after implementation of an automated
breast ultrasound (ABUS) system.
Methods or Background: In this IRB-approved retrospective study all women undergoing ABUS examination in our department
between October 2015 and October 2018 have been revised (1995 women in total, mean age±SD, 48.9±13 years). We included all
women who have undergone a follow-up of at least 24 months after ABUS examination or with BI-RADS category 4 or 5, who have
undergone histological evaluation. FP cases were defined as cases classified BI-RADS 3, 4 or 5 in ABUS who have shown to be benign
lesions during follow-up or after biopsy. FN cases were defined as all cases of cancer diagnosed during follow-up and that were
retrospectively already visible in the ABUS examination. Descriptive statistics were used.
Results or Findings: 1219 women (51.9±11.0 years) were included and classification after ABUS examination was: BI-RADS 1
n=136 (11.1%), BI-RADS 2 n=890 (73.0%), BI-RADS 3 n=161 (13.2%), BI-RADS 4 n=15 (1.2%) and BI-RADS 5 n=17 (1.4%). Patients
had a mean follow-up±SD of 54.3±16.7 months. A woman was diagnosed with an invasive carcinoma 14 months and another 24
months after ABUS examination classified BI-RADS 2; both lesions were already visible in previous ABUS and mammography and the
"retraction phenomenon sign" was visible in the ABUS-coronal reconstruction. A malignancy was also diagnosed in 3/161(1.9%) BI-
RADS 3 lesions, 7/15 (46.6%) BI-RADS4 lesions and 17/17 lesions (100%) BI-RADS 5.The FN rate was 6.9% (2/29 cases). The FP rate
was 13.9% (166/1190 cases). Among the 14 malignant lesions, in one case the lesion was only visible in ABUS and not in
mammography (7.1%); in three cases lesion was visible in both but multifocality only in ABUS (21.4%).
Conclusion: We found a FN rate of 6.9% and a FP rate of 13.9%. To avoid false negative cases ABUS coronal reconstruction should
be accurately evaluated.
Ethics committee - additional information: Ethics committee approval was received; KEK 2016-00064.
Learning curve over three years after implementation of an automated breast ultrasound system in an academic
radiology department (7 min)
Elizabet Nikolova; Zurich / Switzerland
399
Author Block: E. Nikolova, J. K. Weber, G. Zanetti, J. Wieler, T. Frauenfelder, M. Marcon; Zurich/CH
Purpose: The aim of this study was to investigate the learning curve after implementation of an automated breast ultrasound system
(ABUS) over three years.
Methods or Background: In this IRB-approved retrospective study we included all women undergoing ABUS examination in our
department between October 2015 and October 2018 who have undergone a follow-up of at least 24 months after ABUS examination
or with BI-RADS category 4 or 5, who have undergone histological evaluation. The number of additional handheld ultrasound exams
(AHHUS) for better definition of an ABUS finding and number of false positive (FP) cases were noted. FP were defined as cases
classified BI-RADS 3, 4 or 5 in ABUS who have shown to be benign lesions during follow-up/after biopsy. AHHUSs and FPs were
compared in the first 6 months versus second 6 months, yearly over the three years and considering presence/absence of previous
ultrasound exam (UE). Chi-square test was applied.
Results or Findings: A total of 1223 women (mean age±SD, 51.9±11.0 years) were included: 346 (28.3%), 597 (48.8%) and 280
(22.9%) exams were performed respectively in the year 1, 2 and 3. 288/1223 (23.5%) had no previous UE. FP cases were 13/45
(28.9%) in the first 6 months and 70/301 (23.3%) in the second 6 months (p=.454). Cases with AHHUS were 16/45 (35.6%) in the first
6 months and 66/301 (21.9%) in the second 6 months (p=.038). Over three years FP were 83/346 (24.0%), 76/597 (12.7%) and
14/280 (5.0%) for year 1,2 and 3 respectively (p <.001). AHHUS were 82/346 (23.7%), 57/597 (9.5%) and 19/280 (6.8%) for year 1,2
and 3, respectively (p <.001). FP and AHHUS were more frequent in women without previous ultrasound exam (FP:21.2% versus
12.0%, p<.001 and AHHUS: 17.7% versus 11.4%, p <.001).
Conclusion: After ABUS implementation FP and AHHUS gradually reduced, especially after the first year; FP and AHHUS are more
frequent in women without previous ultrasound exam.
Ethics committee - additional information: Ethics committee approval was given; KEK 2016-00064.
Added detection potential of ultrasound in the diagnostic assesment of breast cancer: results from personalised,
integrated, network, knowledge (PINK) study: an Italian longitudinal multicentric study (7 min)
Ludovica Anna Incardona; Florence / Italy
Author Block: L. A. Incardona1, M. Franchini2, P. Belli3, G. M. Giuseppetti4, G. P. Scaperrotta5, E. Cassano3, C. Di Maggio6, E.

Montrucchio7, J. Nori1; 1Florence/IT, 2Pisa/IT, 3Rome/IT, 4Ancona/IT, 5Vimodrone/IT, 6Padua/IT, 7La Spezia/IT
Purpose: The P.I.N.K. study seeks to evaluate the enhanced diagnostic precision in breast cancer screening across Italy, achieved
through various imaging technology combinations, with a particular emphasis on the added diagnostic value of Ultrasound (US) in
detecting breast lesions, especially those undetected by Mammography (MX).
Methods or Background: Initiated in 2018, the study engaged 30,023 patients aged 40 and above, undergoing MX and at least US
and/or Tomosynthesis (TS). Inclusion criteria encompassed patients with histopathologically-confirmed B3-B5 lesions and a positive
US assessment (U4-U6), utilizing a complete diagnostic pathway of MX+TS+US or MX+US+TS. The primary objective was to ascertain
the additional detection rate of US compared to other imaging techniques, stratifying cases detected solely through US by the MX risk
category and age at diagnosis/mammographic breast density.
Results or Findings: Of the recruited, 883 had proven breast B3-B5 lesions, with 408 and 27 undergoing MX+TS+US and
MX+US+TS pathways respectively. US added detection rate in the MX+TS+US pathway was 14.7%, increasing to 21.5% in benign MX
(R2) cases and decreasing to 12% in negative MX (R1) cases. US demonstrated an elevated diagnostic rate in the youngest (40-49
years) and oldest (over 70 years) subgroups, and a higher detection rate (26.1%) in women with dense breasts.
Conclusion: The findings suggest that US could be a significant addition to the current breast cancer screening approach, especially
where MX may not detect lesions, potentially improving early detection and treatment, thereby enhancing patient outcomes.
Can the application of the Koios decision support system optimise diagnostic performance of hand-held breast
ultrasound? A prospective study (7 min)
Giovanni Irmici; Milan / Italy
400
Author Block: G. Irmici1, C. Depretto1, A. Cozzi2, G. Della Pepa1, E. D’Ascoli1, C. De Berardinis1, A. Bonanomi1, S. Marziali1, G. P.
Scaperrotta1; 1Milan/IT, 2Lugano/CH
Purpose: The aim of this study was to investigate if an artificial intelligence-based decision-support system (Koios DS) influences
diagnostic performance of hand-held breast US performed by radiologists of different experience and can reduce the number of
unnecessary breast biopsies.
Methods or Background: This prospective monocentric study enrolled consecutive patients referred for breast biopsy in a tertiary-
level centre between May 2022 and January 2023. A junior (2 years of breast imaging experience) and a senior reader (20 years
experience) assigned US BI-RADS categories before and after Koios application. Differences in diagnostic performance, taking biopsy
as the reference standard, were assessed by comparing areas (AUC) under the receiver-operating characteristic curves (DeLong's
test). The number of unnecessary biopsies (i.e. yielding benign results) before and after the application of Koios was compared with
the McNemar's test.
Results or Findings: We enrolled 222 patients (median age 58 years, interquartile range 46–72) with 226 lesions (at biopsy: 137
malignant, 89 benign). The diagnostic performance of the junior reader (AUC 0.785, 95% confidence interval [CI] 0.736–0.836) was
significantly improved (p <0.001) by Koios (AUC 0.864, 95% CI 0.819–0.908). Likewise, the diagnostic performance of the senior
reader (AUC 0.823, 95% CI 0.777–0.869) was significantly improved (AUC 0.868, 95% CI 0.825–0.912, p <0.001). The diagnostic
performance of the junior reader assisted by Koios (AUC 0.864, 95% CI 0.819–0.908) was higher than that of the senior reader alone
(AUC 0.823, 95% CI 0.777–0.869), albeit non-significantly (p=0.053). The application of Koios was able to significantly reduce the
number of unnecessary biopsies both for the junior (from 51.6% to 32.6%, p<0.001) and the senior reader (from 46.1% to 33.7%,
p<0.001).
Conclusion: The Koios decision-support system is able to significantly improve the performance of junior and senior readers and
reduce the number of unnecessary biopsies.
Limitations: This was a single-centre study within a high-experience setting.
Management of radiological and histopathological discrepancies of breast lesions after US-guided biopsy (7 min)
Vanessa Marisi; Vasto / Italy
1 2 2 2 2 2 2 2 2
Author Block: V. Marisi , M. C. Torrione , S. Hannan , F. Ricciardella , G. Piccolo , A. Di Credico , A. Figorilli , M. Muzi , M. Caulo ;
1 2
Vasto/IT, Chieti/IT
Purpose: The aim of the study was to analyse the concordance between radiological and histopathological characteristics of breast
lesions, demonstrate the high predictive value of the BIRADS (Breast Imaging-Reporting and Data System) classification and asses
managment, surgical treatment and follow-up.
Methods or Background: In this retrospective study, we analysed the discrepant findings between imaging and pathological
findings in patients undergoing breast biopsy between January 2022 and April 2023. 857 patients with 957 lesions undergoing
ultrasound-guided breast biopsy were included in this study. Exclusion criteria were BIRADS4 b having been performed on patients or
biopsies with another first-instance method (stereotaxis or MR-guided biopsy).
Results or Findings: 52 discordant biopsies were evaluated by two experts in breast radiology. 32 classified as BIRADS3 or 4a were
found to be B5 on histological examination (2 infiltrating lobular carcinoma, 8 ductal carcinoma in situ, 16 invasive carcinoma NST, 2
infiltrating ductal carcinoma, 1 papillomatosis, 1 infiltrating papillary carcinoma and 2 invasive G2 carcinoma); 20 biopsies classified
BIRADS4c or 5 were B1 (4), B2 (10) and B3 (6). According to discordance radiological and histopathological findings, 2 patients with
B1 results underwent a new biopsy or VAAB: the result was B5 (1) and B3 (1); 5 patients with B2 result underwent VABB or surgical
excision: the result was B5 (4 patients) and B2 (1 patients).
Conclusion: The BIRADS classification is fundamental in breast cancer diagnosis. When there are discrepancies between imaging
and histological results a discussion by an interdisciplinary team is mandatory: this is crucial in clarifying why they exist and to re-
evaluate the most suitable methodology in continuing the diagnostic process. In our series, 52% of suspicious lesions with a benign
result were found to be malignant after rebiopsy or surgery.
Breast tumour malignancy detection in ultrasound images using a machine learning approach (7 min)
Mostafa Taghipour; Karaj / Iran
401
Author Block: M. Ghelichoghli1, M. K. Tarzamni2, A. Shabanzadeh1, M. Faraji2, A. Akhavan1, M. Taghipour1; 1Karaj/IR, 2Tabriz/IR
Purpose: Breast ultrasound imaging is a widely used non-invasive modality for breast cancer screening and diagnosis. However,
manual interpretation of ultrasound images can be challenging, especially for inexperienced radiologists. Machine learning (ML) has
the potential to assist radiologists in breast ultrasound classification, leading to improved accuracy.
Methods or Background: This study evaluated the performance of machine learning (ML) for breast ultrasound classification. A
dataset of 2090 breast ultrasound images (1140 benign, 950 malignant) was collected from several imaging centers. The images
were acquired using a variety of ultrasound machines and protocols. All images were reviewed by two experienced radiologists to
ensure accuracy in labelling.
To address the class imbalance, the synthetic minority over-sampling technique - edited nearest neighbour (SMOTEEN) was applied.
SMOTEEN performs the SMOTE algorithm (creating synthetic data points for the minority class) and cleans the resulting dataset by
ENN.
An XGBoost classifier was trained and evaluated using a 60-20-20 dataset split. XGBoost is a powerful ML algorithm that is a type of
ensemble learning algorithm, which means that it combines the predictions of multiple weak learners to produce a more robust
prediction.
Results or Findings: The XGBoost classifier achieved an accuracy of 98%, area under the curve (AUC) of 98.2%, precision of 97%,
recall of 98%, and F1 score of 98% on the test set. These results demonstrate the potential of ML for accurate and reliable breast
ultrasound classification.
Conclusion: This study demonstrates the feasibility of using ML to assist radiologists in breast ultrasound classification. The proposed
framework achieved excellent performance on the test set, suggesting that it could improve the breast cancer diagnosis performance
in clinical routines. Further studies are needed to validate the performance of this framework in a larger dataset.
Funding for this study: Funding for this study was received from Med Fanavaran Plus Co.
Response prediction of neoadjuvant treatment in breast cancer patients at baseline ultrasound: comparison of B-mode
and a deep-learning model (7 min)
Panagiotis Kapetas; Vienna / Austria
Author Block: P. Kapetas1, C. Fürböck1, R. Aggarwal2, B. Altuwayjiri3, P. Clauser1, T. H. Helbich1, G. Langs1, P. A. Baltzer1; 1Vienna/AT,
2
Southend-On-Sea/UK, 3Riyadh/SA
Purpose: The study aimed to evaluate whether pre-therapeutic breast ultrasound (US) can serve in the prediction of breast cancer
(BC) patients non-responding to neoadjuvant chemotherapy (NAC) and to compare two different models.
Methods or Background: This retrospective, IRB-approved study included 245 patients with histologically confirmed BC undergoing
NAC. The dataset was divided into a training (165 cases) and a validation set (80 cases). A representative B-mode US image of each
tumour from the pre-treatment examination was selected. Two experienced breast fellows independently evaluated the lesions using
standard BI-RADS descriptors. Logistic regression was used to identify independent predictors of NAC response in the training set and
create a model. Additionally, a Resnet18-based neural network with Dropout layers to decrease the amount of overfitting was trained
to predict the treatment outcome. The performance of both models was evaluated on the validation set using descriptive statistics.
Postoperative histology was the standard of reference for treatment response.
Results or Findings: 145 patients (59.2%) did not achieve a pathological complete response. From the BI-RADS descriptors, oval or
round shape, microlobulated or spiculated margin and the presence of calcifications or edema proved to be independent predictors of
pCR. A model using these showed an accuracy, sensitivity, specificity, positive and negative predictive value of respectively 65%,
82%, 41%, 67% and 61% for the prediction of non-responders to NAC. Compared to that, the DL-based model achieved an accuracy,
sensitivity, specificity, positive and negative predictive value of 76%, 88%, 59%, 76% and 78%.
Conclusion: Breast US can accurately predict lack of response to NAC for BC patients prior to its initiation. A DL model using images
from the baseline US examination demonstrates an increased diagnostic performance as compared to standard B-mode BI-RADS
descriptors.
Limitations: Monocentric study, small patient number.
Ethics committee - additional information: Ethics committee - additional information: This study was approved by the Medical
University of Vienna (Ref. Number: 1793/2013)
Ultrasound Doppler twinkling-guided identification of a breast biopsy marker after neoadjuvant systemic therapy:
preliminary data from a phase one clinical trial (7 min)
Christine Lee; Rochester / United States
402
Author Block: C. Lee, G. Hesley, M. Urban, M. Piltin; Rochester, MN/US
Purpose: The purpose of this clinical trial is to evaluate the safety and twinkling signature of the marker after months of NST.
Ultrasound detection of commercial breast biopsy markers is challenging, particularly in the axilla in patients with node-positive
breast cancer responding favorably to neoadjuvant systemic therapy (NST). We have developed a biopsy marker that demonstrates
robust ultrasound color Doppler twinkling.
Methods or Background: In an ongoing, prospective, single-arm trial involving 10 patients with node-positive breast cancer, a
twinkling marker was implanted near the conventional biopsy marker (control) in the positive axillary lymph node before the start of
NST. Ultrasound was performed at baseline, during NST, and preoperatively to assess for twinkling and B-mode conspicuity of the
twinkling and commercial markers. Based on the preoperative ultrasound, the surgeon determined whether I-125 seed localisation of
the positive node was necessary or if intraoperative ultrasound guidance, as part of their surgical practice, would be used to retrieve
the marked positive node.
Results or Findings: Eight of 10 patients with a mean age of 49.5 years (SD 14.1) have enrolled. Each patient has a different
commercial biopsy marker in the positive node. Thus far, four of the eight patients had ultrasounds during NST, and twinkling of the
twinkling marker was readily present on two different ultrasound scanners (GE Logiq and Fujifilm Arietta). No adverse events. Three
patients have completed targeted axillary dissection, and the surgeon successfully used ultrasound twinkling intraoperatively to
identify the positive node. None of the three patients required preoperative axillary radioactive seed localisation.
Conclusion: Early preliminary results are promising for identifying a marker by ultrasound Doppler twinkling after NST, possibly
obviating preoperative localisation.
Limitations: The small study size and involvement of a single surgeon are limitations.
Funding for this study: Funding was received from Mayo Clinic President's Discovery Translation Program.
Ethics committee - additional information: The study received approval from the Mayo Clinic Institutional Review Board, and
written consent was obtained from all participants. This study is registered under ClinicalTrials.gov NCT0554347.
Tanner stage and ultrasonographic breast development in girls diagnosed with premature thelarche and central
precocious puberty; a prospective follow-up in CPP patients receiving "blocking" therapy (7 min)
Roberta Dattoli; Rome / Italy
Author Block: C. Boldrini, R. Dattoli, A. Marra, S. Amodeo, P. Belli, R. Manfredi; Rome/IT

Purpose: This study aims to evaluate, in a population of pediatric patients diagnosed with premature thelarche (PT) and central
precocious puberty (CPP), the correlation existing between the clinical stage of breast development according to Tanner and the
ultrasonographic stage of the mammary gland, both at the diagnosis and after pubertal “blocking” therapy, the latter evaluated in
CPP girls. Under hormonal stimulation from birth until puberty, various changes in the mammary gland occur.
Methods or Background: A population of 42 pediatric female patients with evidence of breast “bump” at an age below the
threshold of eight years was selected. The girls clinically ascertained to have breast development (Tanner stage B2 and subsequent)
underwent routine tests: measurement of basal levels of FSH, LH and E2, GnRH stimulation test, determination of bone age using
hand-wrist X-ray, vitamin D dosage. The ultrasonographic stage was established by taking into consideration the dimensionally
largest breast and assigned from one to five. Patients diagnosed with CPP were candidates for carrying out blocking therapy
(Decapeptyl®); after six months of therapy they were recalled to evaluate again growth speed, bone age, and US breast
development.
Results or Findings: It was observed that the average number of patients at the diagnosis of CPP was stage 3 at the US
examination (versus Tanner stage B2), and that after six months of blocking therapy the average was stage 2 at US. Clinical
evaluation according to Tanner stages tended to overestimate breast development.
Conclusion: The assessment of the ultrasonographic stage of breast development represents a novel method to help clinicians in the
evaluation of little girls affected with PT e CPP, either in the diagnosis or follow-up, and might become routine very soon in the future.
Limitations: Limitated number of patients
Funding for this study: The study did not receive any funding.
Invasive recurrence after breast conserving treatment of ductal carcinoma in situ of the breast between 1989 and
2021: time trends and the role of tumour grade (7 min)
Lucien Duijm; Nijmegen / Netherlands
403
Author Block: L. Duijm1, R. O'Leary2, L. Boersma2, M. van der Sangen3, S. Siesling4, R-J. Schipper3, A. Voogd2; 1Nijmegen/NL,
2
Maastricht/NL, 3Eindhoven/NL, 4Utrecht/NL
Purpose: The aim of this study was to provide updated figures on the risk of invasive ipsilateral breast cancer (iIBC) after breast
conserving surgery (BCS) of ductal carcinoma in situ (DCIS) with or without adjuvant radiotherapy (RT). A second aim was to analyse
the association between DCIS grade and the risk of iIBC following BCS.
Methods or Background: In this population-based, retrospective cohort study, the Netherlands Cancer Registry collected patient
information on 25,719 women who received a new diagnosis of DCIS in the Netherlands from 1989-2021 and who underwent BCS, of
which 19,034 (74%) received adjuvant RT. Kaplan-Meier analyses and Cox multivariable regression models were used in the analyses.
Results or Findings: A total of 1,135 patients experienced an iIBC. The 10-year cumulative iIBC incidence rates for patients
diagnosed in the periods 1989-1998, 1999-2008 and 2009-2021 and undergoing BCS only, were 12.6%, 9% and 5% (P<0.001),
respectively. For those undergoing BCS with RT these figures were 5.7%, 3.7% and 2.2%, respectively (P<0.001). In the multivariable
analyses, DCIS grade was not significantly associated with the risk of iIBC.
Conclusion: Since 1989 the risk of iIBC has decreased substantially. No significant association of DCIS grade with the risk iIBC was
found, stressing the need for more powerful prognostic factors to guide the loco-regional treatment of DCIS.
Limitations: The well known poor reproducibility among pathologists in grading DCIS lesions will haveresulted in misclassification of
tumour grade and is a potential source of information bias.
Ethics committee - additional information: Informed consent was not necessary for this observational study. Patients are
informed that their data is part of the Netherlands Cancer Registry, and there is an optional opt-out procedure available. No approval
for this study was required from a Medical Ethics Committee.
Ultrasound and shear wave elastography for distinguishing benign and malignant axillary lymph nodes in breast
cancer patients (7 min)
Mesut Öztürk; Samsun / Turkey
1 2 2 1 2
Author Block: C. Kalkan , I. K. Bayrak , M. Öztürk ; Denizli/TR, Samsun/TR
Purpose: This study aimed to assess the diagnostic efficacy of ultrasound (US) and shear wave elastography (SWE) in distinguishing
between benign and malignant axillary lymph nodes in patients with breast cancer.
Methods or Background: A total of 121 axillary lymph nodes from 61 breast cancer patients (mean age: 52.4 ± 14.6 years) were
enrolled between May 2019 and August 2020. Lymph nodes were histopathologically diagnosed through core-needle biopsy or
surgical excision. B-mode US features (short axis diameter, short-to-long axis daimeter ratio, presence of echogenic hilum, presence
of asymmetric cortical thickening, cortex thickness) were assessed and SWE measurements (Maximum shear wave velocity
(SWVmax), minimum SWV (SWVmin), median SWV (SWVmedian), mean SWV (SWVmean)) were performed.
Results or Findings: There were 45 (37.2%) benign and 76 (62.8%) malignant lymph nodes. The short axis diameter, short-to-long
axis diameter ratio, and the mean cortical thickness of malignant lymph nodes were significantly higher compared to benign lymph
nodes (p < 0.001 for each). SWVmean, SWVmedian, SWVmax, and SWVmin of malignant lymph nodes were significantly higher than
those of the benign lymph nodes (p < 0.001 for each).
Conclusion: US and SWE were useful for discriminating between benign and malignant axillary lymph nodes.
Limitations: Limited number of patients included.
Ethics committee - additional information: Ethics committee - additional information: This study was approved by the Ondokuz
Mayıs University Clinical Research Ethics Committee under protocol number B.30.2.ODM.0.20.08/686.
404
RTF Quiz - Mona Lisa smile: women's health imaging through the ages
Categories: Cardiac, Emergency Imaging, Genitourinary, Interventional Radiology, Multidisciplinary

CME Credits: 1.5
Come and take part in this special quiz and support a noble cause! Every participant counts as the ESR will donate 1€ to Doctors
Without Borders for each correct answer you give! Each case will give you an opportunity to show your knowledge, and the ESR will
make sure that it gets rewarded by donating to the great Doctors Without Borders.
Moderators:
Quiz Masters (90 min)

405
E³ 1018 - Cardiovascular radiology into the future

CME Credits: 1.5
At the end of this session, participants will be able to know the current status of new developments in cardiovascular radiology. They
should learn to name clinical indications where myocardial strain imaging or 4D flow can be of importance, and they should be able to
reflect on the advantages of photon-counting CT technology for cardiovascular imaging.
Moderator:

Myocardial strain imaging (24 min)

4D flow for cardiovascular diseases (24 min)

Ursula Reiter; Graz / Austria
Photon-counting CT (24 min)

Panel discussion: What are the unmet needs in cardiovascular radiology? (16 min)
406
Meets 10 - Current developments in radiography and radiotherapy: an Italian perspective
Categories: Breast, Education, Interventional Radiology, Professional Issues, Radiographers

CME Credits: 1
Moderators:
Chairpersons' introduction: radiography in Italy (10 min)

Improving patient experiences in breast imaging: experiences of Italian radiographers (15 min)
Chiara Martini; Parma / Italy
1. To promote the importance of recognising and promoting role evolution.

2. To emphasise the critical role of radiographers, showcasing their contributions to patient care and multidisciplinary collaborations.
Education and training in interventional radiology: the evolving role of Italian radiographers (15 min)
Lena Manzo; Torino / Italy
1. To promote the importance of recognising and promoting the role evolution of interventional radiographers in Italy.
2. To emphasise the critical role of interventional radiographers, showcasing their contributions to patient care and multidisciplinary
collaborations.
Innovation in personalised oncology and advanced skills in interprofessional settings: new perspectives of Italian
radiographers (15 min)
1. To emphasise the growing significance of the evolving role of therapeutic radiographers/radiation therapists in Italy.
2. To facilitate the development of new professional frameworks and educational programmes.
3. To promote and highlight the contributions of therapeutic radiographers/radiation therapists within care pathways.
407
MS 10 - Critical leg ischaemia: the role of IR in limb salvage
Categories: Education, Interventional Radiology, Multidisciplinary, Research, Vascular

ETC Level: LEVEL II
CME Credits: 1.5
This interactive session will discuss recent advances in endovascular revascularisation of peripheral vascular disease. Contemporary
evidence of best practices, technological developments and role of various tools and techniques will be discussed in depth to guide
practitioners into safe and effective patient care in a multidisciplinary environment.
Moderator:
MohamadS. Hamady; London / United Kingdom

Patient selection algorithm for open vs endovascular approach (15 min)

Sarah Onida; London / United Kingdom
Drug-eluting technology in infrapopliteal disease, evidence update (15 min)

The role and evidence of new technology in heavily calcified lesions: atherectomy, shockwave (15 min)
Florian Wolf; Vienna / Austria
Multidisciplinary case presentation and discussion (35 min)
Conclusion and take-home message (5 min)

408
RW 10 - How to improve MSK reports
Categories: Musculoskeletal, Professional Issues

CME Credits: 1
Tips on improving your MSK reports (15 min)

James F. Griffith; Shatin / Hong Kong SAR China
1. To learn how to improve report readability.

2. To understand the importance of clear, clinically relevant reports.
James F. Griffith; Shatin / Hong Kong SAR China
1. To appreciate that report writing is a lifelong evolving process that benefits from constructive criticism and recommendation.
409
E³ 1021 - Indeterminate retroperitoneal masses
Categories: Abdominal Viscera, Imaging Methods, Translational Imaging

CME Credits: 1.5
Overview and diagnostic algorithms (45 min)

Dow-Mu Koh; London / United Kingdom
1. To become familiar with focal and diffuse retroperitoneal lesions.

2. To learn the indications of additional imaging modalities.
Special cases and differential diagnosis (45 min)

Wolfgang Gerhard Kunz; Munich / Germany
1. To discuss the appearance of particular retroperitoneal lesions.

2. To understand when MRI could be useful.
410
SF 10 - Total-body PET: technical aspects and clinical applications
Categories: Hybrid Imaging, Molecular Imaging, Nuclear Medicine, Oncologic Imaging, Translational Imaging
CME Credits: 1.5
Moderator:

A physicist's perspective (15 min)

Thomas Beyer; Vienna / Austria
1. To define total-body PET.

2. To highlight the key differences and advances with regard to standard-of-care PET systems.
3. To be alert to flexible imaging protocols based on the higher performance characteristics of total-body PET over whole-body PET.
Impact on clinical PET workflow (15 min)

Axel Rominger; Bern / Switzerland
1. To know the current total-body PET landscape.

2. To get an overview of the opportunities of such systems.
3. To understand the challenges and how to overcome them.
Clinical experience of an early adopter (15 min)

Lorenzo Nardo; Sacramento / United States
1. To describe the main differences between total-body PET/CT and short axial field of view PET/CT.
2. To identify key clinical indications for total-body PET/CT.
3. To consider future research areas enabled by total-body PET/CT.
A quantum jump forward in imaging technology (15 min)

Ronald Boellaard; Amsterdam / Netherlands
1. To describe why total-body or large field of view systems offer a tremendous improvement in sensitivity and image quality.
2. To list to what degree radiation dose and/or scanning time can be reduced using total-body PET, providing new opportunities for
ultra-low dose-ultra fast imaging.
3. To name and identify three major new patient care and research opportunities that total-body PET can offer.
Panel discussion: Total-body PET: why (not) everyone should get one? (25 min)
411
RC 1012 - Imaging the preterm neonate
Categories: Abdominal Viscera, Chest, General Radiology, Neuro, Paediatric

ETC Level: LEVEL I
CME Credits: 1
Moderator:
Willemijn Margriet M. Klein; Arnhem / Netherlands

Willemijn Margriet M. Klein; Arnhem / Netherlands
Typical neuro diagnosis in the preterm neonate (15 min)

Vasileios G Xydis; Ioannina / Greece
1. To become familiar with the imaging approach we use.

2. To learn what is normal for age.
3. To become familiar with the imaging patterns of encephalopathy of prematurity.
Typical cardiothoracic diagnosis in the preterm neonate (15 min)

Baptiste Morel; Tours / France
1. To identify the normal cardiothoracic radiographic aspect.

2. To list three of the most frequent premature cardiothoracic pathologies.
3. To describe the imaging patterns of the main premature cardiothoracic pathology.
Typical abdominal diagnosis in the preterm neonate (15 min)

Josephine Bomer; Oslo / Norway
1. To identify radiographic signs of necrotising enterocolitis.

2. To interpret ultrasound findings in necrotising enterocolitis.
3. To know the complications of necrotising enterocolitis.
4. To name the most important differential diagnosis in necrotising enterocolitis.
Panel discussion: Should every general radiologist be able to recognise the typical preterm neonate diagnoses? (10
min)
412
OF 10T - Career development: teaching smarter for the next generation

CME Credits: 1
Moderator:

What makes a good radiology teacher? Trends and techniques (15 min)
Cindy Chew; Glasgow / United Kingdom
1. To discuss principles and the current state of the art in medical education.
2. To provide examples of best teaching practices for young radiologists.
What makes a good radiology teacher? Traits and training (15 min)
Antonios Tzortzakakis; Stockholm / Sweden
1. To discuss what personal qualities make for a good radiology teacher.

2. To discuss opportunities for your radiologists to learn to be better teachers.
Teaching fellowships: a springboard to being a better educator (15 min)

Ivana Blazic; Belgrade / Serbia
1. To discuss the opportunities from teaching fellowships to become better educators.

2. To discuss how this can benefit young radiologists.
Open forum discussion: Making the most of opportunities as a young radiologist (10 min)
413
IF 10 - Make radiology green again: can radiology be sustainable?
Categories: Contrast Media, Education, Imaging Methods, Management/Leadership, Physics in Medical Imaging
CME Credits: 1.5
Healthcare contributes substantially to the overall greenhouse gas emissions. Among all specialities, radiology is one of the most
energy-intensive - despite the industry's efforts to provide lower-consumption devices. Furthermore, contrast media residuals in
wastewater and medical waste in interventional radiology contribute to radiology's environmental footprint. This session will provide
an overview of radiology's impact in those areas and discuss new approaches to mitigate them.
Moderator:

Switching off for the future: how to lower energy consumption (15 min)
Tobias Heye; Basel / Switzerland
Contrast media recycling: where are we at? (15 min)

Olivier Clément; Paris / France
Waste management in interventional radiology: can we do better? (15 min)

Mark C. Burgmans; Leiden / Netherlands
Sustainability and AI in radiology: problem or solution? (15 min)

Daniel Truhn; Aachen / Germany
Panel discussion: How much impact does radiology have on global warming? (25 min)
414
E³ 1020 - Pain management
Categories: Interventional Oncologic Radiology, Interventional Radiology, Musculoskeletal

CME Credits: 1.5
The session gives an overview of the clinical and technical aspects of pain management in patients with benign and malignant
conditions.
Moderator:
Luca Maria Sconfienza; Milano / Italy

Upper and lower limb joints: steroids and PRP (24 min)
Domenico Albano; Cefalu' / Italy
Spine injections for benign low back pain (24 min)

Antoine Feydy; Paris / France
Pain management in cancer patients (24 min)

Georgia Tsoumakidou; Lausanne / Switzerland
Panel discussion: How to establish the "IR pain clinic" (13 min)
415
ESOR - Soft skills
Categories: Education, Management/Leadership

CME Credits: 1.5
The session aims to bring an understanding of soft skills in radiology. Furthermore, the attendees should be made aware of the
importance of soft skills in radiology leadership and their importance for education in radiology.
Moderators:

What are soft skills (15 min)

Lennart K. Blomqvist; Stockholm / Sweden
Why soft skills are important in radiology clinical practice (15 min)
Why soft skills are important in radiology leaders (15 min)

My experience: Nicholas Gourtsoyannis Teaching Fellowship (10 min)

Hugo Alexandre Meireles Rio Rio Tinto; Lisbon / Portugal
Award of certificate of appreciation to ESOR training centre (5 min)

Award of certificates to ESOR scholars and fellows (20 min)

416
TC 1027 - Neurodegeneration: movement disorders

CME Credits: 1
This educational session will discuss how MRI and molecular imaging techniques can be used to understand the underlying pathology
in patients with neurodegenerative movement disorders such as Lewy body dementias, Parkinson’s Disease, and atypical
parkinsonism. This session will give general radiologists and those interested in neuroradiology and nuclear medicine/molecular
imaging the knowledge needed for appropriate selection and structured assessment of imaging exams in diagnostic work-up of
movement disorders. The session format consists of in part lectures, followed by an interactive case-based discussion.
Moderators:
Kejal Kantarci; Rochester / United States

MR imaging in movement disorders (19 min)

Stephane Lehericy; Paris / France
Molecular imaging in movement disorders (19 min)

Movement disorders: interactive case discussion (19 min)

417
RC 1010 - Muscle imaging
CME Credits: 1
Moderator:

Grading muscle injury (15 min)

Jaime Isern Kebschull; Barcelona / Spain
1. To characterise sport muscle injuries.

2. To present the contemporary imaging classifications used for sports-related muscle injuries.
3. To appreciate the clinical significance of grading muscle injuries.
Myositis and myopathy (15 min)

Ana Isabel Garcia Diez; Barcelona / Spain
1. To list the more frequent inflammatory and hereditary myopathies with their clinical characteristics.
2. To apply the whole-body MRI to assess disease activity and extension of the changes, to identify the target site for biopsy, as
monitoring tool and to evaluate related pathologies.
3. To identify some specific MRI patterns associated with individual phenotypes.
4. To critique some MRI limitations.
5. To integrate functional and quantitative MRI techniques with a potential role in the evaluation of the myopathies.
Muscle tumours and pseudo-tumours (15 min)

Filip M. Vanhoenacker; Bonheiden / Belgium
1. To discuss the imaging semiology of tumour and tumour like conditions of the muscles.
2. To discuss how to differentiate benign from malignant muscle tumours.
3. To summarise the diagnostic strategy.
Panel discussion: What is the ideal imaging algorithm for muscle disorders? (10 min)
418
EIBIR 10 - Implementing lung cancer screening programmes in Europe (SOLACE Project)
Categories: Oncologic Imaging, Professional Issues, Research

CME Credits: 1.5
Moderator:
Oyunbileg von Stackelberg; Heidelberg / Germany

Oyunbileg von Stackelberg; Heidelberg / Germany
Evidence and bottlenecks of lung cancer screening: the vision of SOLACE (10 min)
1. To consolidate knowledge about the evidence.

2. To appreciate challenges in implementation and reaching out to eligible participants.
3. To learn about the requirements for EU4health projects.
The Croatian experience (10 min)

Miroslav Samarzija; Zagreb / Croatia
1. To learn from Europe's first national lung cancer screening programme.

2. To appreciate the potential of a cloud-based IT system and the benefits of complete digitalisation.
3. To understand the role of the general practitioners.
The French experience focusing on women (10 min)

1. To learn about gender differences in lung cancer screening.

2. To appreciate the requirement to build a woman-only cohort.
3. To understand the potential benefits of linking breast and lung cancer screening.
The Hungarian experience focusing on deprived populations (10 min)

Ildiko Horvath; Budapest / Hungary
1. To learn about the preliminary experience from the Hungarian pilot projects.
2. To appreciate specific efforts to reach out to deprived populations.
The Czech experience focusing on participants at very high risk (10 min)
Martina Koziar Vasakova; Praha 4 - Krc / Czechia
1. To learn about the preliminary experience from the Czech pilot projects.
2. To appreciate aspects of higher risk beyond smoking history.
3. To understand how to identify and reach out to participants at very high risk.
Future impact of AI (10 min)
419
Colin Jacobs; Nijmegen / Netherlands
1. To learn about the basics for automated detection of lung nodules.

2. To appreciate the current developments to assess the probability of malignancy.
3. To understand the future role of AI for workflow in lung cancer screening.
Participant's perspective, outreach, and dissemination (10 min)

Pippa Powell; Sheffield / United Kingdom
1. To learn about the expectation of the participants and how to address their needs best.
2. To appreciate the importance of dedicated materials to increase participation rates.
3. To understand the SOLACE strategies for outreach and dissemination.
Panel discussion: How to implement lung cancer screening successfully? (15 min)
420
ST 9 - EFRS Activities within ECR 2024

Overview of EFRS activities at ECR2024.
Moderators:
Interview (30 min)

Louise A. Rainford; Dublin / Ireland
Jonathan Loui Portelli; Msida / Malta
421
AI-SC 4 - Building a compliant data registry for AI research: the way we do it

Moderator:


Emanuele Neri; Pisa / Italy
Hanna Leisz; Heidelberg / Germany
Esther Bron; Rotterdam / Netherlands
1. To discuss methodologies to build a data registry for AI research which is compliant with ethical regulation.
2. To identify pitfalls of this process.
3. To discuss examples of data registries as built-in private and academic institutions.
422
CUBE 12 - Ablation procedures: management

Moderator:

Anaesthesiologic support for percutaneous tumor ablation (14 min)

Filippo Piacentino; Varese / Italy
1. To discuss the types of anaesthesia (general, sedation, local, nerve blockage).

2. To try to define the best anaesthesiologic way to support percutaneous tumor ablation.
The ideal environment for ablation procedures (14 min)

Elena Bozzi; Calci / Italy
1. To discuss the preparation of the angiosuite/CT room and rescue devices (how to choose the environment & why).
2. To discuss the preparation of the patient before performing the procedure.
3. To discuss the staff training to manage ablation procedures.
423
EFRS 4 - Patient engagement & inclusion in Radiotherapy
Categories: Oncologic Imaging, Professional Issues, Radiographers

ETC Level: LEVEL I
This session aims to gain a clear definition of patient engagement in the context of radiotherapy and explore its impact on improving
processes and infrastructure fostering person-centred care, to recognise the critical role of patient involvement in safety management
and enhance radiographers' interaction with patients towards improving patient safety, and morover to highlight the importance of
patient inclusion in radiographers' education and research and explore strategies for implementation.
Moderator:
Anastasia Sarchosoglou; Athens / Greece

Patient involvement in process (re)design (10 min)

Ainars Bajinskis; Riga / Latvia
Patient involvement in decision making (10 min)

Theresa O'Donovan; Co. Cork / Ireland
Patient involvement in safety management (10 min)

Valerio Pisoni; Giussano / Italy
Patient involvement in radiographer education (10 min)

Discussion (10 min)
Closing (5 min)
424
SF 11 - How to teach in an increasingly digital world?
Categories: Education, Professional Issues

CME Credits: 1.5
Moderator:
Chairperson's introduction: transformation of radiology education, technology challenges, and opportunities (5 min)
Are books becoming obsolete? (15 min)

Franz Kainberger; Vienna / Austria
1. To explore whether textbooks are becoming obsolete.

2. To present the benefits and drawbacks of e-learning programmes.
3. To learn how to engage students in active learning.
Extended reality: gadget or tool for imaging education? (20 min)

1. To learn what extended reality means.

2. To recognise the benefits and limitations of extended reality in imaging education and training.
3. To know the role of the different forms of extended reality in the future of imaging education and training.
Clinical diagnostic reasoning teaching using online platforms (20 min)

Berit Verbist; Leiden / Netherlands
1. To explain and illustrate the concept behind clinical reasoning learning using online platforms.
2. To present active learning methods using online platforms.
3. To learn the effectiveness of online clinical teaching.
AI in radiology: curse or threat for training radiology residents (20 min)

Lorenzo Faggioni; Pisa / Italy
1. To explain which are the basic concepts on AI to be included in the radiology training programmes.
2. To learn how to use AI for precision education in radiology.
3. To discuss the necessity of a standardised AI curriculum to prepare trainees for an effective use of AI tools in their future practice.
Panel discussion: What changes are the new educational platforms and AI bringing into the field of education in
radiology? (10 min)
425
E³ 1121b - Imaging of head and neck: focus on salivary glands and eye
Categories: Head and Neck, Imaging Methods, Oncologic Imaging

CME Credits: 1.5
Salivary glands (45 min)

1. To become familiar with the appropriate imaging studies and radiological anatomy of the salivary glands.
2. To illustrate the spectrum of imaging findings.
3. To learn how to avoid interpretation pitfalls on CT, US, and MRI.
Eye and orbit (45 min)

Katharina Erb-Eigner; Berlin / Germany
1. To become familiar with the appropriate imaging studies and radiological anatomy of the eye and orbit.
2. To illustrate the spectrum of imaging findings in neoplastic and non-tumour diseases.
3. To learn how to avoid interpretation pitfalls on CT, US, and MRI.
426
EU 11 - Computational tools for the establishment and use of diagnostic reference levels (DRLs)
Categories: EuroSafe Imaging/Radiation Protection, Physics in Medical Imaging

CME Credits: 1.5
Moderator:
John Damilakis; Iraklion / Greece

Artificial intelligence (AI) for the establishment of DRLs (22 min)

1. To understand why DRLs should be seen as a dynamic tool that follows the development of clinical practice and technology
advances.
2. To appreciate the potential role of AI in establishing DRLs.
3. To learn how AI can assist in dose estimation and optimisation.
The role of dose management systems in the establishment and use of DRLs (22 min)
1. To understand the importance of dose management systems (DMS) in establishing and using dose reference levels (DRLs) for
radiation dose management in radiology practices.
2. To learn about the different types of DMS available and their respective advantages and limitations in implementing DRLs for
different radiological procedures.
3. To gain insights into the best practices for integrating DMS with DRLs to optimise the quality of radiological examinations.
Challenges and solutions in using big data in establishing DRLs for medical imaging (22 min)
Hugues Brat; Sion / Switzerland
1. To understand the importance of big data in establishing DRLs for medical imaging.
2. To identify potential challenges associated with using big data for DRLs, including anonymisation, security, privacy, verification,
data quality, data interoperability, bias, ethical considerations, and cost.
3. To learn about potential solutions to overcome these challenges, such as automated quality control tools, data validation
processes, peer review, collaboration, data sharing, and random sampling.
Panel discussion: Is the use of computational tools important for the implementation of DRLs in clinical practice? (20
min)
427
RW 11 - How to improve abdominal imaging reports
Categories: Abdominal Viscera, Professional Issues

CME Credits: 1
Tips on improving abdominal imaging reports (15 min)

Andrew Plumb; Esher / United Kingdom
1. To write clear reports addressing the clinical question.

2. To communicate effectively and prioritise communication of important findings.
3. To simplify reports.

2. To discuss focusing on clinically relevant findings and optimising report clarity.
428
OF 11T - Back to the basics: imaging for undergraduates
Categories: EuroSafe Imaging/Radiation Protection, General Radiology, Imaging Methods, Physics in Medical Imaging, Students
ETC Level: LEVEL I
CME Credits: 1
Moderator:

Basics of x-ray imaging (12 min)

1. To explain the basic principles of x-ray imaging.

2. To appreciate how x-ray is applied to diagnosis through case examples.
Basics of US imaging (12 min)

1. To explain the basic principles of US imaging.

2. To appreciate how the US is applied to diagnosis through case examples.
Basics of CT imaging (12 min)

1. To explain the basic principles of CT imaging.

2. To appreciate how CT is applied to diagnosis through case examples.
Basics of MR imaging (12 min)

Sofia Gourtsoyianni; Athens / Greece
1. To explain the basic principles of MRI.

2. To appreciate how MRI is applied to diagnosis through case examples.
Panel discussion: Questions you wanted to ask about imaging (8 min)
429
TC 1127 - The next frontier in imaging in neurodegeneration

CME Credits: 1
Moderators:

1. To describe the changing role of anatomical and molecular imaging for patients with neurodegenerative disorders, and the shift
from nosological diagnosis to a biomarker-defined diagnosis.
2. To identify newest developments in PET tracers for specific neurodegenerative diseases.
3. To explain the role of advances in AI and imaging hardware in the diagnosis of neurodegenerative disease.
The future of your dementia imaging practice: MRI (19 min)

The future of your dementia imaging practice: PET (19 min)

Javier Arbizu; Pamplona / Spain
To infinity and beyond: disruptive software and hardware developments for imaging neurodegeneration (19 min)
Ciprian Catana; Charlestown / United States
430
RC 1110 - Update in imaging rheumatology
Categories: General Radiology, Imaging Methods, Musculoskeletal

CME Credits: 1
Moderator:
Monique Reijnierse; Leiden / Netherlands

Monique Reijnierse; Leiden / Netherlands
How to avoid overdiagnosis of SpA on MRI (15 min)

Antoine Feydy; Paris / France
1. To describe the MRI findings of inflammatory diseases of the axial skeleton.

2. To learn about differentials of inflammatory diseases of the axial skeleton.
New insights and developments in imaging of spondyloarthropathy (15 min)

Winston Joseph Rennie; Leicester / United Kingdom
1. To describe how to optimise your protocol.

2. To list the potential role of dual-energy CT for the detection of inflammatory diseases in the axial skeleton.
3. To list the potential role of synthetic CT in inflammatory diseases in the axial skeleton.
Advances in ultrasound of juvenile inflammatory arthropathies (15 min)

Iwona Sudol-Szopinska; Warsaw / Poland
1. To explain the role of ultrasound in inflammatory arthropathies in children and adolescents compared to conventional radiography
and MRI.
2. To list the ultrasound findings and differentials in inflammatory arthropathies in children and adolescents.
Panel discussion: The role of imaging in the diagnosis and characterisation of rheumatologic diseases (10 min)
431
BS 11 - Creating a thriving interdisciplinary team within a radiology department
Categories: Management/Leadership, Multidisciplinary, Professional Issues, Radiographers, Research

CME Credits: 1
Moderator:
Joseph Castillo; Fgura / Malta

Joseph Castillo; Fgura / Malta
Building bridges and exploring the value of other healthcare professionals within the radiology department (18 min)
Lyanne Molenaar; Amsterdam / Netherlands
1. To describe the role and value of other healthcare professionals when reshuffling tasks in a radiology department.
2. To identify the positive outcomes for radiographers and radiologists when reshuffling tasks.
3. To discuss common challenges that arise when trying to integrate other healthcare professionals into radiology teams.
Interprofessional education to enhance patient-centred care (18 min)

Switinder Singh Ghotra; Lausanne / Switzerland
1. To outline the benefits of interprofessional education.

2. To describe a framework used to combine five learning processes within a jointly designed 3-week interprofessional education (IPE)
programme.
3. To explain how an interprofessional training programme focussed on person-centred care enhanced critical thinking through the
development of five skills: interprofessional communication, role clarification, team functioning, conflict management and leadership.
The potential of interdisciplinary research to advance practice, care and safety (18 min)
Helle Precht; Middelfart / Denmark
1. To describe the potential of interdisciplinary collaboration within radiology-specific research.

2. To create a patient-centred focus on research topics when including interdisciplinary colleagues.
3. To analyse examples of research in advanced practice, care, and safety with the later implementation in clinical practice.
432
RC 1105 - Structured reporting: ready for clinical routine?
Categories: General Radiology, Imaging Informatics, Interventional Radiology, Multidisciplinary, Professional Issues
ETC Level: LEVEL II
CME Credits: 1
Moderator:

Why and how to integrate structured reporting in clinical routine? (15 min)
1. To learn about the benefits of structured reporting for diagnosis.

2. To appreciate the benefits of structured reporting in the communication with referrers and patients.
3. To understand the challenges of integrating structured reporting in clinical workflows.
4. To review existing solutions for structured reporting and template collections.
Structured reporting in the era of large language models (15 min)

1. To summarise the current status of structured reporting.

2. To examine how state-of-the-art technologies might influence clinical practice.
3. To critically evaluate current reporting practices and propose improvements.
Is there a place for structured reporting in interventional radiology? (15 min)

Irene Bargellini; Candiolo / Italy
1. To learn about the specifics of using structured reporting pertaining to interventional radiology.
2. To understand the value of structured reporting in the follow-up and adverse events documentation.
3. To appreciate the evolving paradigm in a real-world setting and upcoming advances.
Panel discussion: Where does the radiologist meet the clinician when it comes to structured reporting? (10 min)
433
PC 11 - The radiologist of the future: between the digital twin and the human being
Categories: Imaging Informatics, Multidisciplinary, Professional Issues, Research, Translational Imaging

CME Credits: 1.5
Moderator:

The issue of prediction models in human beings (18 min)

Andrew Julian Vickers; New York, NY / United States
1. To provide a background on prediction models.

2. To provide an overview of available prediction models.
3. To address the challenges of prediction models and digital twins.
Digital twin framework: technical implementation (18 min)

Huan Xuan Nguyen; London / United Kingdom
1. To provide a background on digital twin technology.

2. To provide an overview of the digital twin's technical needs.
3. To address the challenges of the implementation of digital twins in healthcare.
Radiology: bridging the gap between human beings and DT (18 min)
1. To provide an overview of the role of AI and imaging biomarkers.

2. To provide insights on the future development of digital twin in radiology.
3. To highlight the fields of applicability in radiology.
Ethical issues (18 min)

Ilenia Rapisarda; Aci Sant'Antonio, Catania, Sicilia / Italy
1. To provide an overview of the legal issues.

2. To provide an overview of the ethical issues.
3. To provide a roadmap to overcome the challenges.
Panel discussion: Are we ready for the clinical implementation of digital twins? (13 min)
434
E³ 1121a - Emergency ultrasound: beware of pitfalls and artefacts!
Categories: Emergency Imaging, Genitourinary, Imaging Methods

CME Credits: 1.5
Gynaecological/obstetrical emergencies (45 min)

Marijana Basta Nikolic; Novi Sad / Serbia
1. To learn about the pitfalls and artefacts encountered in the US in gyn/obstetrical emergencies.
2. To discuss when the patient should be studied with further imaging modalities.
Acute scrotum: diagnosis is not always straightforward (45 min)

1. To become familiar with different scrotal and non-scrotal causes of acute scrotal pain.
2. To learn about the spectrum of findings in the US in patients with acute scrotal pain.
3. To become familiar with scrotal disease presenting with isolated abdominal symptoms.
435
AWARDS - Abstract Award Ceremony

During the ceremony, the most exceptional EPOS Posters will be offered an award.
Presenting EPOS Poster Awards (45 min)

Presenting Radiographers awards (15 min)

Presenting Research Presentation Abstract Awards (30 min)

436
RC 1102 - Personalised breast cancer screening: where do we stand today?
Categories: Breast
ETC Level: LEVEL I
CME Credits: 1
Moderator:
Sophia Zackrisson; Malmö / Sweden

Sophia Zackrisson; Malmö / Sweden
Risk assessment models (15 min)

1. To provide an update on the existing individualised risk models.

2. To understand the variables used to calculate the risk, which indicates screening recommendation and frequency.
3. To understand how deep learning aids in breast cancer risk assessment.
Personalised screening in intermediate risk (15 min)

Athina Vourtsis; Athens / Greece
1. To understand which group of women are considered at intermediate risk and to review the current screening recommendations for
these women.
2. To appreciate the outcomes and controversies of implementing personalised breast cancer screening.
3. To present emerging methods and future vision for personalised breast cancer screening.
Personalised screening in high-risk women (15 min)

Pascal A.T. Baltzer; Vienna / Austria
1. To understand how to identify women that are considered at high-risk for developing breast cancer.
2. To analyse the different screening guidelines used for the early detection of breast cancer in women at high risk.
3. To learn about the benefits of breast MRI and to adapt the available screening strategies in different scenarios.
Panel discussion: Feasibility and acceptability of personalised breast cancer screening into clinical practice (10 min)
437
Meets 11 - Changing times: adopt, adapt, improve
Categories: Education, Management/Leadership, Professional Issues

CME Credits: 1
Moderators:
Andrea Giovagnoni; Ancona / Italy
From past to the future: new horizons for SIRM (12 min)
1. To present the vision and transformation of a historical, scientific society born in 1913 and now representing more than 12 000
members in Italy.
2. To describe the structure of the board, subspecialty sections and regional groups.
3. To illustrate the societal activities, educational programmes and professional acts.
The SIRM journal, a success story (11 min)

Antonio Barile; L'Aquila / Italy
1. To present the main journal of SIRM: editorial board and section editors.
2. To show the citations, rejection rate and H-index.
3. To describe the developing strategies of the journal.
Diversity, equity, inclusion: a new project for Italian radiologists (11 min)
Stefania Anna Montemezzi; Verona / Italy
1. To show the evolution of society reflected in the scientific and professional areas.
2. To describe how to embrace diversity and inclusion to express radiologists' full potential.
3. To share the Italian initiatives to strengthen regional connections and synergies.
Radiology at the Italian Winter Olympic Games 2026: a SIRM project with the National Olympic Committee (11 min)
Ettore Squillaci; Roma / Italy
1. To present an innovative partnership between radiology and the Italian Winter Olympic Games 2026.
2. To explain how radiology can excel among the branches of sports medicine radiology to become an ally of winter sports athletes.
438
E³ 1119 - CT: a multiparametric technique

CME Credits: 1.5
Moderator:
Uwe Joseph Schoepf; Charleston / United States

Uwe Joseph Schoepf; Charleston / United States
Spectral CT: the new standard of care? (25 min)

Philippe Charles Douek; Lyon / France
1. To describe the different technologies available for spectral imaging.

2. To analyse the impact of spectral technology with regard to radiation exposure and contrast medium dose.
3. To summarise the most relevant current clinical applications.
Photon counting: where we are and where we will go? (25 min)
Victor Mergen; Zurich / Switzerland
1. To describe the current technologies available for photon-counting CT.

2. To analyse the impact of spectral technology with regard to radiation exposure and contrast medium dose.
3. To summarise the most relevant current clinical applications.
AI for optimisation of image quality and contrast medium injection (25 min)
Damiano Caruso; Roma / Italy
1. To understand AI impact in CT protocols for dose reduction.

2. To evaluate the standard of care for contrast media administration.
3. To describe AI advantages in contrast media reduction.
Discussion (10 min)
439
RPS 1106 - Advances in hybrid, molecular and translational imaging
Categories: Hybrid Imaging, Molecular Imaging, Nuclear Medicine, Oncologic Imaging, Research, Translational Imaging
CME Credits: 1.5
Moderator:
Antonio Esposito; Milano / Italy
Assessment of hypoxia-induced neoangiogenesis in breast cancer xenografts using non-contrast-enhanced

multiparametric MRI and fluorescent multiplexed immunohistochemistry (7 min)
Silvester Julian Bartsch; Vienna / Austria
Author Block: S. J. Bartsch1, K. Brožová1, V. Ehret1, J. Friske1, L. Kenner1, K. Kratochwill1, D. Laimer-Gruber1, T. H. Helbich1, K. Pinker-
Domenig2; 1Vienna/AT, 2New York, NY/US
Purpose: Tumour neoangiogenesis is an important hallmark of cancer progression, triggered by alternating selective pressures from
the hypoxic tumour microenvironment. Multiparametric MRI combining blood oxygen level dependent (BOLD) MRI, depicting blood
oxygen saturation, and intravoxel incoherent motion (IVIM) MRI, capturing intravascular and extravascular diffusion, allows insights
into tumour progression, capturing tumour oxygenation and neovascularisation simultaneously. The combination of BOLD- and IVIM-
MRI may provide non-invasive, non-contrast-enhanced imaging biomarkers of neoangiogenesis for the discrimination of breast cancer
(BC) molecular subtypes. We compare our findings to fluorescent multiplexed immunohistochemistry (MP-IHC).
Methods or Background: An in-vivo study on 36 female athymic nude mice, which were inoculated with luminal A, Her2+ and triple
negative BC cells, was conducted. MRI was performed using a 9.4T Bruker BioSpec 94/30USR system. Following IVIM-MR imaging,
tumour oxygenation was measured at 21% oxygen and 100% oxygen for BOLD measurements. Tumours were resected for MP-IHC
staining of CD-31, PDGFR-beta and Hif1-alpha.
Results or Findings: Hyperoxic BOLD MRI discriminates luminal A from Her2+ and triple negative BCs, while the IVIM-derived
parameter fIVIM allows the distinction of luminal A and Her2+ from triple negative BCs. A principal component analysis (PCA) of BOLD
and IVIM-MRI derived parameters reveals differences between triple negative and other BC molecular subtypes. In a PCA of MP-IHC
stains, luminal A BCs clustered separately from other BC molecular subtypes.
Conclusion: Our multivariate analysis of BOLD and IVIM parameters highlights how oxygen delivery is constricted and
neoangiogenesis is increased in triple negative BCs. The combined interpretation with MP-IHC provides a holistic view on hypoxia-
induced neoangiogenesis in BCs. We conclude that non-contrast-enhanced mpMRI using BOLD and IVIM imaging provides promising
imaging biomarkers for an assessment of hypoxia-induced neoangiogenesis in three BC molecular subtypes.
Funding for this study: Funding was provided by the Vienna Science and Technology Fund (WWTF), grant no. LS19-018.
Ethics committee - additional information: This preclinical study was approved by the Austrian Federal Ministry of Education,
Science and Research (project number BMFWF-66.009/0284-WF/V/3b/2017).
Avoiding anaesthesia in oncologic paediatric patients using short acquisition times with FDG total-body PET/CT (7 min)
Clemens Mingels; Sacramento / United States
440
Author Block: C. Mingels, B. A. Spencer, H. Nalbant, M. Rokni, Y. G. Abdelhafez, F. Sen, R. Badawi, L. Nardo; Sacramento, CA/US
Purpose: Given the increased signal collection efficiency, Total-Body (TB) PET/CT allows for protocol flexibility including acquisition
time (AT). This study aimed to define the shortest AT with maintained diagnostic quality to avoid or reduce anaesthesia time in
children.
Methods or Background: Twelve oncologic paediatric patients were injected with 4.24 MBq/kg 18F-FDG and scanned in list mode
for 20 minutes on uEXPLORER TB-PET/CT after 120 min uptake time. 300s, 150s, 75s, and 37s data sets were extracted and
reconstructed with varying iterations (4-6) and filters (4-6mm FWHM Gaussian or Metz) and then compared to the full data set
reconstructed with 4 iterations (20 subsets), 256x256 matrix, no smoothing. All reconstructions were simultaneously displayed and
independently rated from 1 (best) to 4 (worst) for interpretation/diagnosis by three nuclear medicine physicians. Additionally, signal-
to-noise ratio (SNR), tumour-to-background ratio (TBR) and contrast-to-noise ratio (CNR) were calculated using a 30mm diameter
sphere in the right liver lobe and tumour lesions segmented using a 40%-iso-contour. Data sets were compared using Student's t-test
to the reference data set for changes in semi-quantitative measurements.
Results or Findings: Readers selected the reference reconstruction parameters for the 300s and 150s data sets but preferred
additional smoothing for shorter data sets. Liver, blood-pool and bone SNR decreased with shortened AT. TBR and CNR were not
significantly different between 20 min, 300 s and 150 s. Compared to 20 min TBR and CNR reduced significantly in 75 s and 37 s
reconstructions (TBR 20 min: 3.64; 75 s: 2.40, p <0.05 and 37 s: 2.09, p <0.05; CNR 20 min: 11.29; 75 s: 2.36, p=0.01 and 37 s: 1.77,
p=0.01).
Conclusion: 150 s paediatric acquisitions on TB-PET/CT obtained after 120 minutes uptake time show no significant semi-
quantitative image quality reduction compared to 20 min scans and may decrease or avoid the need of anaesthesia.
Limitations: The study featured a small cohort.
Funding for this study: Funding was received from the US National Institutes of Health R01-CA249422;R01-CA206187.
Ethics committee - additional information: The evaluation was approved by the hospital committee.
Ga-68 Trivehexin and Ga-68 Pentixafor: novel radiopharmaceuticals in PET/CT imaging and their comparison with F-18
FDG PET/CT (7 min)
Dinesh Kumar Gauthaman; Chennai / India
Author Block: D. K. Gauthaman, I. M. Elangovan, S. Simon; Chennai/IN

Purpose: The aim of this study was to compare the novel radiopharmaceuticals Ga-68 Trivehexin (αvβ6 integrin imaging) and Ga-68
Pentixafor (CXCR-4 imaging) PET/CT with F-18 FDG PET/CT in terms of target specificity, diagnostic accuracy, and possible theranostic
implications.
Methods or Background: This prospective observational study included nine patients with suspected pancreatic malignancy and
eight patients with suspected multiple myeloma who underwent Ga-68 Trivehexin and Ga-68 Pentixafor PET/CT respectively. All 17
patients underwent F-18 FDG PET/CT and were followed up for the biopsy results.
Results or Findings: Among nine patients with suspected pancreatic malignancy (six males, three females; mean age 56 years), all
(9/9) showed pancreatic lesions in Ga-68 Trivehexin PET/CT, whereas, six (6/9) demonstrated lesions in F-18 FDG PET/CT. Biopsy from
the pancreatic lesions showed adenocarcinoma. The mean SUV max of pancreatic lesions was 6.6 in Ga-68 Trivehexin PET and 5.3 in
F-18 FDG PET. Similarly, among eight patients with suspected multiple myeloma (six males, two females; mean age 64 years), all
(8/8) showed skeletal lesions in Ga-68 Pentixafor PET/CT, whereas, five (5/8) showed lesions in F-18 FDG PET/CT. Bone marrow biopsy
confirmed plasma cell neoplasm. The mean SUVmax of skeletal lesions was 14.9 in Ga-68 Pentixafor PET and 12.1 in F-18 FDG PET.
Conclusion: Ga-68 Trivehexin and Ga-68 Pentixafor are novel diagnostic PET probes in imaging of pancreatic malignancy and
multiple myeloma respectively. They showed superior diagnostic performance compared to F-18 FDG PET/CT. Further studies in larger
populations should be conducted to validate their promising role as theranostic agents.
Limitations: This was a pilot study featuring a small study population.
The target to background ratio was less in the pancreas in Ga-68 Trivehexin PET/CT in a few cases.
Ethics committee - additional information: The hospital ethics committee approved the study and an informed written consent to
participate in the study was obtained from all participants.
Characterisation of international metabolic prognostic index (IMPI) and in its components in CAR T-cell treatment of
lymphoma (7 min)
Michael Winkelmann; Munich / Germany
441
Author Block: M. Winkelmann, V. Blumenberg, K. Rejeski, V. Bücklein, F. Dekorsy, J. Ricke, M. Subklewe, W. G. Kunz; Munich/DE
Purpose: Chimeric antigen receptor T-cell therapy (CART) is approved for relapsed or refractory (r/r) non-Hodgkin lymphoma (NHL).
Recently, the international prognostic metabolic index (IMPI), which is composed of metabolic tumour volume (MTV), age, and Ann
Arbor stage, was introduced. As higher patient age is a positive prognostic factor and higher MTV is a negative prognostic factor in the
context of CART, we compared the predictive value of each component of the IMPI and its prognostic value for progression-free
survival (PFS) and overall survival (OS).
Methods or Background: Consecutive r/r NHL patients and 18F-FDG PET/CT imaging at baseline were selected. Ann Arbor stage
and patient age were calculated at the time of lymphodepletion. MTV was calculated with an absolute SUV cut-off of 4.
Results or Findings: 43 patients were included (37% female, 63% male) with a median age of 66 years and baseline MTV of 276
cm2. Ann Arbor stage was 1 in four patients, 2 in 11 patients, 3 in eight patients, and 4 in patients.
Splitting patients by median IMPI showed minor differences in median PFS and OS, which were statistically not significant.
Dichotomisation by median MTV alone revealed a larger, statistically significant difference in median PFS and a larger nonsignificant
difference in OS. Patients with an older age demonstrated a slightly longer PFS and OS, whereas there was no relevant difference
between the two Ann Arbor risk groups.
Conclusion: For r/r NHL patients undergoing CAR T-cell therapy, IMPI represents a promising tool for risk assessment. However, in
our study MTV alone was superior in PFS and OS stratification. A larger cohort should be used to assess whether MTV alone or in
combination with measurement of tumour dissemination can improve survival prediction.
Limitations: This was a single centre study with a limited patient population.
Funding for this study: Funding for this study was received from Bayrisches Zentrum für Krebsforschung (BZKF) as well as
Förderung für Forschung und Lehre der LMU (FöFoLe).
Ethics committee - additional information: All medical records and imaging studies were reviewed with the approval of the LMU
Munich Institutional Review Board (LMU Ethics Committee, project number 19-817). Informed consent was obtained from all individual
participants included in the study.
First report of 23Na-MRI in kidney cancer to detect differential sodium-accumulation in benign vs malignant tumours (7
min)
Ines Horvat Menih; Cambridge / United Kingdom
Author Block: I. Horvat Menih, J. Birchall, M. McLean, M. Zamora-Morales, A. Bebb, J. Kaggie, A. Warren, G. Stewart, F. A. Gallagher;
Cambridge/UK
Purpose: Renal cell carcinoma (RCC) poses a significant clinical challenge. Accurate early detection is hindered by limitations in
current diagnostic methods. This study investigates the potential of 23Na-MRI to noninvasively differentiate kidney tumour subtypes,
shedding light on sodium regulation in renal cancer.
Methods or Background: Eight patients with kidney tumours underwent 23Na-MRI. Total sodium concentration (TSC) was
calculated based on an 80mM 23Na-phantom. Regions of interest were drawn on TSC maps. Histopathology was determined on renal
mass biopsy or surgical specimen. Immunohistochemical markers (CD31, Ki67) were assessed in a separate patient cohort. RNAseq
counts for NHE3 and Na+/K+-ATPase were obtained from online databases. Statistical analyses were performed in Graphpad Prism.
Results or Findings: 7 renal oncocytomas (RO), 2 chromophobe RCCs (chRCC) and 2 clear cell RCCs (ccRCC) were identified in the 8
participants. 4 participants are undergoing active surveillance, and 4 underwent nephrectomy.
TSC was highest in ccRCC, and lowest in chRCC, while normal tissue and RO had comparable values. The main Na+ transporter in the
kidney, NHE3, was highest in ccRCC and downregulated in chRCC. By contrast, ATPase was highest in chRCC, followed by ccRCC and
RO. As expected, the ccRCC as the most aggressive kidney tumour subtype, exhibited highest vascularity (CD31) and highest
proliferation rate (Ki67), which may have further contributed to the increased sodium signal detected on imaging.
Conclusion: We have shown feasibility for 23Na-MRI to noninvasively detect sodium concentration differences across kidney
tumours, and associated findings with molecular markers of sodium handling. This prepares the ground for future research and shows
potential to improve the clinical management of kidney cancer.
Limitations: It was not possible to quantify intracellular/extracellular sodium. There was no separation of cortex and medulla, and no
correction for cystic/necrotic areas.
Funding for this study: This work is supported by the Cancer Research UK Cambridge Centre (RQAG/119) as well as the NIHR
Cambridge Biomedical Research Centre.
Ethics committee - additional information: The investigation of the differential biology of benign and malignant renal masses
using advanced magnetic resonance imaging techniques (IBM-Renal) was approved by the Cambridge research ethics committee, no.:
22/EE/0136.
Validation of the standardisation framework SSTR-RADS 1.0 for neuroendocrine tumours using the novel
SSTR‑targeting peptide [18F]SiTATE (7 min)
Ricarda Ebner; Munich / Germany
442
Author Block: R. Ebner, A-K. Lohse, M. P. Fabritius, J. Rübenthaler, P. Bartenstein, J. Ricke, F. Grawe; Munich/DE
Purpose: Somatostatin receptor positron emission tomography/computed tomography (SSTR-PET/CT) using [68Ga]-labelled tracers is
a widely used imaging modality for neuroendocrine tumours (NET). Recently, [18F]SiTATE has shown great potential due to its
favourable clinical characteristics. We aimed to evaluate the reproducibility of SSTR-RADS 1.0 for structured interpretation and
treatment planning of NETs using [18F]SiTATE.
Methods or Background: Four readers assessed [18F]SiTATE-PET/CT of 95 patients according to the SSTR-RADS 1.0 criteria at two
different time points. Each reader selected and evaluated up to 5 target lesions per scan. Overall scan score and the decision on
peptide receptor radionuclide therapy (PRRT) were considered. Inter- and intrareader agreement was determined using the intraclass
correlation coefficient (ICC).
Results or Findings: Interreader agreement for identical target lesions (ICC≥85%), overall scan score (ICC≥90%) and decision to
recommend PRRT (ICC≥85%) showed excellent agreement. However, significant differences were observed in recommending PRRT
within ERs (p =0.020) and IRs (p =0.004). Compartment-based analysis demonstrated good to excellent interreader agreement for
most organs (ICC≥74%), except for lymph nodes (ICC≥52%).
Conclusion: SSTR-RADS 1.0 represents an accurate and reproducible framework system for stratifying [18F]SiTATE-PET/CTs as an
alternative for [68Ga]-labelled PET/CTs in NET-imaging. However, excellent interreader agreement on the overall scan score and the
decision for PRRT was observed, there were variations in PRRT recommendations, highlighting the complexity of such decisions,
suggesting the need for multidisciplinary input. Compartment-based assessments demonstrated excellent interreader agreement for
the liver, soft tissue, and skeleton, with varying agreement for lymph nodes, emphasising the importance of functional imaging for
small lesions.
Limitations: The absence of histopathological confirmation for each target lesion and blinding of readers to patients' clinical status
may have influenced inter- and intrareader agreement, indicating the need to integrate clinical information into the interpretation of
[18F]SiTATE-PET/CT scans.
Ethics committee - additional information: The analysis of the data was approved by the institutional ethics board of LMU Munich
(IRB 20-1077).
First application of novel human granzyme B imaging agent in a humanised melanoma mouse model treated with
immune checkpoint inhibitor therapy (7 min)
Priska Summer; Hagenbrunn / Austria
Author Block: P. Summer1, N. Gallon1, N. Bulmer1, S. Prabhu1, P. Heidari2, U. Mahmood2; 1Charlestown, MA/US, 2Boston, MA/US
Purpose: This study aims to evaluate whether a novel human granzyme B PET imaging agent CYT-200 (Cytosite) labelled with 68Ga
can be used as a reliable, non-invasive biomarker for detecting intratumoural granzyme B levels, while evaluating the efficacy of
immune checkpoint inhibitor therapy in vivo.
Methods or Background: Melanoma-bearing humanised NSG mice were randomly assigned to receive either anti-PD1/CTLA4
combined, anti-PD1 alone or saline (controls) 9, 12 and 15 days after tumour implantation. PET imaging was performed prior to
treatment initiation (baseline), such as on days 4, 7 and 14 after the first treatment dose. Tumour-bearing mice were injected with
68Ga-CYT-200 labelled 1h prior to PET/CT imaging. Intra-tumoural T-cell activity was determined by tumour to blood ratio (TBR),
calculated by the standard uptake values (SUVs) of the tumours normalised against the heart blood pool. Treatment response was
assessed by tumour growth measurements over 35 days after the first treatment dose.
Results or Findings: PET imaging on day 4 after treatment initiation showed the highest TBR compared to the baseline (3.08±0.98
vs. 1.22±0.48) with a gradual decrease of the TBRs on day 7 (2.76±1.85), and 14 (1.91±0.23). Anti-PD1/CTLA4 decreased the tumour
size on days 4 and 7 after treatment initiation (336±38 mm3 vs.154±31 mm3 and 131±129 mm3, respectively). However, tumour
growth recurred up to the final time point, 12 days after treatment initiation (181±112 mm3), correlating with the T-cell activity
measured through in vivo PET imaging using CYT-200.
Conclusion: CYT-200 labelled with 68Ga can detect intra-tumoural T-cell activity associated with tumour killing following immune
checkpoint inhibitor therapy in a humanised mouse model for melanoma.
Limitations: The study may have a limited number of mice, which could affect the generalisability and statistical power of the
findings.
Funding for this study: Funding for this study was received by the National Institutes of Health.
Ethics committee - additional information: All animal studies were approved and conducted according to the IACUC guidelines.
Assessment of CAR T-cell therapy response in mice with melanoma using Granzyme B PET imaging (7 min)
Priska Summer; Hagenbrunn / Austria
443
Author Block: P. Summer1, N. Bulmer1, N. Gallon1, S. Prabhu1, P. Heidari2, U. Mahmood2; 1Charlestown, MA/US, 2Boston, MA/US
Purpose: The purpose of this study is to determine whether granzyme B PET imaging can predict treatment response to CAR T-cell
therapy while evaluating the efficacy of CAR T-cell therapy for melanoma.
Methods or Background: Chimeric antigen receptor (CAR) T-cell therapy is a novel cell-based immunotherapy that urgently
requires a reliable tool to determine patient response rapidly and accurately. Four days after melanoma cell implantation, tumour-
bearing NSG mice received either 2x10^6 CAR T-cells (n=9) or vehicle (controls; n=8) intravenously. Treatment response was
evaluated by tumour growth measurements up to 35 days post-treatment. A human granzyme-B-specific agent, CYT-200, labelled
with 68Ga was used for PET imaging to assess intra-tumoural T-cell activity on days 2, 7 and 14 after treatment. Further, liver and
colon uptake were compared between the groups to evaluate granzyme B expression within different organs.
Results or Findings: Tumoural 68Ga-hGZP uptake was significantly greater in the CAR T group 2 (3.1 ± 1.2 vs. 1.1 ± 0.4, P =
0.0017) and 7 (2.0 ± 1.1 vs. 1.1 ± 0.1, P = 0.0111) days after treatment, much earlier than when CAR T treated mice first presented
with significantly lower tumour volumes eleven days after tumour implantation (61.8 mm3 ± 8.7 vs. 287.1 mm3 ± 157.6, P = 0.0455).
Conclusion: Quantitative in vivo imaging of intratumoural granzyme B provides early prediction of treatment response as early as 2
days after treatment. With the recent translation of granzyme B PET imaging to cancer patients treated with checkpoint-inhibitors,
these findings may help inform CAR T response assessment in patients.
Limitations: The study may have a limited number of mice, which could affect the generalisability and statistical power of the
findings.
Funding for this study: Funding for this study was received from the National Insitutes of Health.
Ethics committee - additional information: IACUC approval was obtained.
Are we ready for novel radionuclides in nuclear medicine in Europe: PRISMAP survey perspective (7 min)
Author Block: M. Radzina1, L. Saule1, E. Mamis1, E. Pajuste1, T. E. Coccolios2, T. Stora3; 1Riga/LV, 2Brussels/BE, 3Geneva/CH
Purpose: The study aimed to understand the current status and future perspectives of novel radionuclides in Europe, through a
survey. In order to support the ongoing research across Europe and to facilitate access to novel radionuclides, the PRISMAP
consortium (European medical radionuclides programme) was established.
Methods or Background: A consortium questionnaire was disseminated among 30 countries. We received 114 respondents through
the PRISMAP consortium and user community, professional associations and preclinical/clinical end users in Europe and the current
status of clinical end users in nuclear medicine was identified.
Results or Findings: A total of 40 preclinical/clinical users institutions took part in the survey. Clinical end users currently use the
following radionuclides in their studies: 177 Lu, 68 Ga, 111 In, 90 Y, other alpha emitters, 225 Ac, 64 Cu, and Terbium isotopes.
Radionuclides that would be of interest for users within the next 2 to 5 years are 64 Cu, Terbium radionuclide “family” and alpha
emitters, such as 225 Ac. Active industry involvement in joint outreach activities will provide access to new radionuclides and new
purity grades for medical research. This will enhance clarity and regulatory procedures to foster research with radiopharmaceuticals
and improve the delivered radionuclide data and regulation, along with biomedical research capacity.
Conclusion: The current perspective shows that nuclear medicine specialists/clinical end users from broad parts of Europe are not
only interested in new radionuclides for diagnostics, but also in therapy and technology advancements that confirm their interest in
development. This study was preliminary and should be extended outside the PRISMAP consortium.
Limitations: Not all European countries were covered - the majority of responses came from Western Europe, most notably the
Benelux, France and Italy. More emphasis is needed to reach respondents from South-Eastern Europe.
Funding for this study: This project has received funding from the European Union’s Horizon 2020 research and innovation
programme under grant agreement No 101008571 (PRISMAP). This document reflects only the view of the author(s). The funding
agenc(y/ies) is/are not responsible for any use that may be made of the information it contains.
Ethics committee - additional information: Not applicable.
Micro-CT vs conventional CT: preliminary results of comparing radiomic features in vivo and ex vivo in lung cancer (7
min)
Leonardo Brizzi; Milan / Italy
444
Author Block: L. Brizzi, L. Preda, C. Bortolotto, D. Gioacchino; Pavia/IT
Purpose: Radiomic analysis holds the potential to revolutionise lung cancer management by providing nuanced insights into tumour
characteristics, ultimately aiding in early detection and personalised treatment, thus reducing mortality rates significantly. This study
aims to investigate correlations between radiomic features in images obtained both from conventional spiral-CT and micro-CT scans
of post-surgical anatomical specimens extracted from lung cancer patients.
Methods or Background: Three patients with lung cancer (pT2-3N0-1M0) underwent conventional in vivo spiral-CT before surgery.
After surgery, post-surgical specimens were scanned using both spiral-CT (slice 0.3 mm) and micro-CT (slice thickness 40 μm).
Tumour lesions from in vivo CT and within the specimens were segmented manually and semi-automatically using ITK-SNAP. A
comparative radiomic analysis using Pyradiomics software was conducted, comparing features (shape and II order) from both image
modalities before and after surgery, using ICC for assessing agreement.
Results or Findings: Concordance statistics for the 107 extracted radiomic features between image modalities revealed the
following percentages of agreement: 79% when comparing micro-CT scans and conventional CT scans of the anatomical specimens,
74% when comparing micro-CT scans of the anatomical specimens and conventional in vivo CT scans, and 95% when comparing
conventional CT images of the anatomical specimens with conventional in vivo CT scans. Among all the extracted features the highest
degree of concordance was found in GLCM (Gray Level Co-occurrence Matrix; variation between 0.06-7.97%) and shape-based
features (0.01-0.92%).
Conclusion: These preliminary findings indicate good to excellent concordance of radiomic features extracted from micro-CT and
conventional CT scans, both in vivo and ex vivo. These results establish micro-CT as a possible tool for investigating the biological
basis of radiomic features. In clinical practice, the diagnostic potential of micro-CT could be expanded to the point of performing
"virtual biopsies".
Limitations: The possible limitations are the small sample size, and the feasibility of the study.
Funding for this study: Funding was received from Ricerca Corrente Founding.
Ethics committee - additional information: 1. Patients' consent was acquired by signing standard forms according to hospital
protocol.
Preoperative FDG uptake is associated with histopathological response in patients with operable NSCLC patients
receiving neo-adjuvant immuno-chemotherapy: a prospective single-centre study (7 min)
Daria Kifjak; Vienna / Austria
Author Block: D. Kifjak, M. J. Hochmair, J. Klinger, K. Sinn, A. Hoda, R-I. Milos, A. Haug, H. Prosch, L. Beer; Vienna/AT
Purpose: The aim of this study was to evaluate the association between 18-FDG-PET CT quantitative imaging markers and
histological responses from patients with operable non-small cell lung cancer (NSCLC) treated with neo-adjuvant immuno-
chemotherapy.
Methods or Background: This prospective, single-centre study included 22 patients (7 male, 15 female) with NSCLC who were
treated with neo-adjuvant immuno-chemotherapy and underwent preoperative 18F-FDG-PET-CT. We semi-automatically extracted the
standardised uptake value (SUVmax) and metabolic tumour volume (MTV). The histological results were retrieved from patients’
records. Patients were assigned to either complete pathologic response (cPR) or non-cPR groups. A Mann-Whitney-U-Test was
calculated to determine if there were differences between SUVmax, MTV and histologic response.
Results or Findings: Eleven patients had a cPR, while eleven had a non-cPR. The SUVmax was lower in patients with cPR compared
to those without cPR (median: 4.69 (IQR 5.7) vs. 9.9 (IQR 19.3)), p=0.049. The MTV was not significantly different between the two
groups (median: 16.9 (IQR) 71.4, 29.2 (IQR 110.7)) p=0.0622.
Conclusion: Presurgical SUVmax was associated with cPR in NSCLC patients receiving immuno-chemotherapy. However, cPR can
also be observed in patients with high SUVmax values.
Limitations: This study is a single-centre study with a small sample size.
Funding for this study: Funding was received from the Austrian Federal Ministry for Digital and Economic Affairs, the National
Foundation for Research, Technology and Development and the Christian Doppler Research Association.
Ethics committee - additional information: EC number 1521/2015
Pittsburgh B compound PET volumetric vs SUV analysis for cardiac amyloidosis (7 min)
Anthony Chuprin; Jacksonville / United States
445
Author Block: A. Chuprin, J. Young, M. K. Jain, M. Mahowald, F. Kestel, Amyloid research group; Jacksonville, FL/US
Purpose: Pittsburgh B Compound (PiB) is one of the newer investigational radiotracers studied for use in evaluation of cardiac
amyloid. Potential advantages include the ability to distinguish between types of cardiac amyloid (ATTR vs AL), additional prognostic
information, and monitoring of treatment response. Most if not all studies have quantified cardiac PiB uptake using standardized
uptake value (SUV) metrics. We investigate applications of volumetric analysis of PiB PET and compare it to SUV metrics.
Methods or Background: Retrospective, blinded, case-cohort study on eight patients with biopsy proven ATTR (n=5) and AL (n=3)
amyloidosis with cardiac involvement who had a PiB PET/CT.
Results or Findings: Compared with ATTR cardiac amyloidosis, AL amyloidosis showed higher mean PiB volumetric uptake (246 mL
vs 29.5 mL, p = 0.0005), mean SUVmax (7.3 vs 5.5, p = 0.085) and mean SUVmean (4.3 vs 3.7, p = 0.191. There was no overlap
between ranges of volumetric PiB uptake in cases of cardiac AL amyloidosis and cardiac ATTR amyloidosis (162.6 - 316.1 mL vs. 0 -
60.7 mL), unlike with SUVmax (6.6-8.1 vs. 2.9-8) and SUVmean (2-5.2 vs. 4.1-4.7).
Conclusion: Volumetric PiB PET analysis shows unique application for distinguishing AL from ATTR cardiac amyloidosis. Given
increasingly sophisticated PACS paving the way for more feasible volumetric analysis, we expect volumetric PET analysis to be more
readily available for clinical application. However, further analysis with a larger cohort is needed to confirm these findings and explore
any association with volumetric PiB uptake and clinical outcomes.
Limitations: The limitation of this study is the sample size.
Funding for this study: Funding was received from the Mayo Internal Funding-Clinical Research Operations Group (CROG).
Ethics committee - additional information: This study was approved by the Institutional Research Ethics Board.
446
RPS 1105 - Novel AI models redefining radiology diagnostics
Categories: Artificial Intelligence & Machine Learning, General Radiology, Imaging Methods, Multidisciplinary
CME Credits: 1.5
Moderator:
The use of large language models for first triage decisions for patients at risk for reaction during intravenous contrast
administration: a proof of concept (7 min)
Miriam Dolciami; Rome / Italy
Author Block: G. Avesani1, M. Marin2, M. Dolciami1, L. D'Erme1, A. Perazzolo1, L. Russo1, V. Celli1, B. Gui1, E. Sala1; 1Rome/IT,
2
Gravedona/IT
Purpose: The aim of this study was to determine whether a large language model (GPT 3.5) can provide accurate and valuable
guidance on management for patients at risk for reaction during intravenous contrast administration.
Methods or Background: Six guidelines from various scientific societies were collected, both in English and the local language.
These documents were embedded using OpenAI embeddings within the LangChain framework, creating a database to provide
information to a GPT-3.5 turbo model. We formulated 100 clinical scenarios describing different situations, combining allergic and
renal problems (e.g., moderate to severe allergic reaction and different renal functions) and different types of contrast media (iodine
and gadolinium). We asked the model to give a textual answer for each clinical scenario indicating the correct patient management
following the previously given guidelines. The responses generated by the model were evaluated by a human expert in the field,
considering formal correctness and clinical usefulness. A Likert 5-point scale for each task (correctness and usefulness) was used to
judge the answers (from 5 = correct/safe or very useful to 1 = completely wrong or completely useless for clinical purposes). We
dichotomised the responses with a cut-off of ≥4 to consider the answers acceptable.
Results or Findings: The model's answers were judged formally correct and safe for patients in 95% of scenarios and valuable in
84% of cases. Predominantly, answers deemed not valid were considered too vague to be used.
Conclusion: LLMs have the potential to aid in the clinical management of critical patients. Such models can be very useful for novice
personnel or initial screenings. Better performance might be achieved with fine-tuning and the "tree of thought" techniques.
Limitations: There was limited prompt engineering and fine tuning.
Ethics committee - additional information: No patients data were used.
STP+SC2F: transformer-based transfer learning framework for asymmetry and architectural distortion detection on
mammograms (7 min)
Ma Jie; Shenzhen / China
447
Author Block: Z. Cao, M. Jie, T. Liao, Y. Yang, X. Lin, J. Yuan, L. Ma; Shenzhen/CN
Purpose: The aim of this study was to develop a joint transformer-based framework and a detection model to detect asymmetry (AS)
and architectural distortion (AD) on mammograms.
Methods or Background: Our data is collected from three collaborative hospitals at distinct geographical locations using Siemens
and Giotto equipment following the ACR standard and dated from 2011 to 2018. For training, we use 2,584 mammograms fully
labeled with masses, 526 with ASs, 248 with ADs, and 25,139 mammograms with no lesion. For testing, we collect 128 mammograms
fully labelled with the AS, 102 with the AD, 432 with the mass, and 3,660 with no lesion. We propose a framework with Supervised
mass-transferred pre-training (STP) followed by supervised constrained contrastive fine-tuning (SC2F) to detect ASs and ADs. We first
pre-train AsAdNet to detect the mass, AS, and AD as one class in STP. Then, we use a novel constrained contrastive learning to depart
them in SC2F as fine-tuning. We evaluate the model performance with the free-response operating characteristic when different
detection results are acquired under various thresholds. A detection of AS or AD is correct if it has an loU of over 0.2 with the ground
truth. We evaluate the sensitivity with false-positive-per-image (FPPI) at 0.1, 0.5, 1, and 2.
Results or Findings: Our model has a sensitivity of 0.172, 0.513, 0.662, 0.746, and 0.179, 0.554, 0.676, and 0.753, with FPPI at 0.1,
0.5, 1, and 2 for AS detection and AD detection. This result outperforms all existing methods by a noticeable margin, achieving state-
of-the-art performance for AS and AD detection on this data set with 29,159 mammograms.
Conclusion: We present a transformer-based framework of STP + SC2F with a detection model, AsAdNet, to solve the asymmetry
and architectural detection on mammograms.
Limitations: The system requires GPUs to operate.
Ethics committee - additional information: This project is approved by the IRB; number LL-XJS-2020011. The ethics review and
institute approved this retrospective case-control study; the national review board waived individual informed consent requirements.
Enhancing quick-acquired MRI scans with the DL-based Aikenist framework: a clinical assessment (7 min)
Bhanu K. N. Prakash; Singapore / Singapore
Author Block: C. S. Arvind1, S. S. Bhat2, B. Dikendra3, S. Z. T. Jordan1, A. Amrapuram2, B. K. N. Prakash1; 1Singapore/SG,

2
Bangalore/IN, 3Chennai/IN
Purpose: MRI remains a cornerstone for clinical diagnostics and research. High-resolution MRI, though detailed, necessitates
extended acquisition, increasing patient discomfort and risk of motion artifacts. A frequently used alternative is the quick-acquisition
technique, but it tends to compromise image quality due to noise and diminished contrast.
Methods or Background: This study employed a DL-based Aikenist post-processing enhancement on QuickScan-acquired MR scans
from 30 brain scans (5400 slices) and 32 abdomen subjects (1920 slices). The brain and abdomen data were acquired from different
MRI scanners (GE, Siemens, Toshiba) at different locations using different acquisition protocols, introducing scanner variability.
Results or Findings: Our results highlighted significant improvements in image quality metrics, even accounting for scanner
variability.
For brain scans, the average SNR surged from 28.44 to 42.92 (p <0.001) and CNR from 11.88 to 18.03 (p <0.001). Meanwhile,
abdominal scans experienced an SNR leap from 35.30 to 50.24 (p <0.001) and an impressive CNR ascent from 10,290.93 to
93,767.22 (p <0.001).
Furthermore, in a double-blinded evaluation, clinicians emphasised the enhanced visibility of intricate anatomical structures,
intrastructural changes, such as the IMAT, muscle boundaries, tissue-tissue interface, brain structural delineation, and improved bias
field correction, which were previously not accentuated. Their feedback solidified the clinical importance of our enhancement,
particularly in
discerning smaller regions earlier concealed by noise and reduced contrast.
Conclusion: Aikenist enhancement doesn't merely elevate MRI image aesthetics but offers a robust solution to ensuring diagnostic
accuracy without extending scan times. As MRI scans become more integral to healthcare, innovations like this pave the way for a
more patient-centred and efficient imaging process.
Limitations: Results, though promising, stem from specific anatomies and scanner types. Efficacy may fluctuate with varied MRI
parameters.
Ethics committee - additional information: The IRB approved this study.
Frameworks for artificial intelligence research in medical image analyses: a systematic review (7 min)
Manjunath Kanabagatte Nanjundappa; Manipal / India
448
Author Block: M. Kanabagatte Nanjundappa1, V. Kulkarni2, A. Kulkarni3, Y. M4, C. Maram5; 1Manipal/IN, 2Leesburg, VA/US,
3
Bengaluru/IN, 4Karlsruhe/DE, 5Hyderabad/IN
Purpose: Artificial intelligence (AI) has a strong footprint in radiology workflow, from image acquisition to reporting findings. This
review attempts to overview such AI frameworks in medical image analyses (in diagnostics and therapeutics) from a biomedical
engineering perspective.
Methods or Background: Several AI, machine learning (ML), and deep learning (DL) frameworks have been developed by academic
research institutes and healthcare companies that are available as open-source software frameworks. Commercially available and
community-based DL frameworks are reviewed. The frameworks were compared according to various parameters such as the
technology used, CPU/GPU-based implementation, feature learning time, performance evaluation, whether they are desktop
installations or cloud-based applications to work with and deployment type (commercial grade with production code or research
prototype) and clinical validation.
Results or Findings: More than a hundred open-source DL frameworks are available. A few have done exceptionally well in
computer-aided diagnosis systems, such as Microsoft InnerEye, NVidia CLARA, pyRadiomics, and MONAI. Regulatory body approvals
and clinical validations are pending in many reviewed products.
Conclusion: This review paper helps the researchers, radiology residents, and radiologists to gain insight into these frameworks and
libraries and select the right one for fast prototype development for image analysis in radiology applications.
Limitations: We could not evaluate all the AI frameworks as they have vast applications in many imaging modalities for diagnosis
and therapy, and also, most of them are clinically not fully validated to accept them as clinical solution.
Ethics committee - additional information: This is a review and hence ethical issues did not arise.
Energy-optimised scheduling of CT examinations through mathematical modelling (7 min)

Martin Segeroth; Basel / Switzerland
Author Block: M. Segeroth1, A. Nurkanović2, J. Vosshenrich1, M. Diehl2, T. Heye1; 1Basel/CH, 2Freiburg/DE

Purpose: Radiology departments and medical imaging devices in particular are major energy consumers within a hospital. The aim of
this study was to calculate possible energy savings by optimally scheduling CT examinations.
Methods or Background: Data of all CT examinations performed in our tertiary care radiology department in 2015 on three CT
scanners was retrospectively included. Data consisted of examination timestamps and power consumption in kilowatts. The optimal
scheduling problem was formulated using linear constraints, a linear objective function and only binary decision variables as an
integer linear programming (ILP) problem. This formulation allowed rigorous modelling of a nonconvex and nondifferentiable objective
function and offers the possibility to compute the optimal solution even for very large models.
Results or Findings: In total 261 workdays were analysed, with 15’072 CT examinations scheduled on the three CT scanners. The
duration to solve the ILP for every workday was 10.14 s (9.46-10.80 s). The model yielded a 34.9% reduction in the scanners'
combined daily energy consumption through optimal examination scheduling. In absolute values, daily energy consumption could be
decreased by 42.2 kWh from 121.0 kWh (120.6-121.4 kWh) to 78.8 kWh (77.8-79.8 kWh; P <.001). Energy savings are primarily
attributed to examination shifting, allowing for increased system off times. Overall, 10,930.6 kWh in energy, $2,864 in cost, and
1,399.1 kgCO2eq in carbon emissions could theoretically be saved in our setting.
Conclusion: Optimised CT examination scheduling through automatic modelling has substantial sustainability and cost benefits for
radiology departments. Feasibility of model implementation in clinical routine needs further investigation.
Limitations: Although it was ensured that the approach was optimised, this study requires extensive clinical testing.
Ethics committee - additional information: This study was approved by the ethics committee of northwest and central
Switzerland, project ID 2022-01016.
medBERT.de: a German BERT model tailored for the medical domain: insights into the results of radiological text
classification and entity recognition (7 min)
449
Author Block: F. Busch1, J-M. Papaioannou1, P. Grundmann1, F. Borchert2, L. C. Adams3, L. Xu1, M. Makowski3, A. Löser1, K. K.
Bressem1; 1Berlin/DE, 2Potsdam/DE, 3Munich/DE
Purpose: We developed medBERT.de, a German BERT (Bidirectional Encoder Representations from Transformers) model for the
medical domain trained on 4.7 million German medical documents. Here, we present the results of our custom pretrained BERT
models for classification and entity recognition tasks from radiology reports.
Methods or Background: medBERT.de is built on the standard BERT architecture, featuring 12 layers with 768 hidden units each, 8
attention heads, and a 512-token input limit. Three distinct radiological benchmarks based on 2,000 radiology reports, respectively,
obtained from a level 1 hospital in Germany, were developed to span various report lengths and tasks: a short text classification from
chest x-rays, a longer report classification from chest CT examinations, and a named entity recognition (NER) task from medium-sized
CT/x-ray reports of the wrist. Reports were manually labelled by radiologists and medical students for various pathologies and
therapeutic devices. The model and benchmarks were made publicly available (https://huggingface.co/GerMedBERT/medbert-512).
Results or Findings: medBERT.de displayed superior performance for the chest x-ray (AUROC: 84.65) and CT classification (AUROC:
96.69) tasks compared to previously published German BERT models. For the NER task, the model trained with deduplicated data
achieved the highest AUROC of 83.28. Notably, medBERT.de's performance on longer texts from CT reports (258 ± 100 words) was
especially pronounced compared to x-ray (98 ± 27 words) or NER (108 ± 41 words) tasks.
Conclusion: The study underscores the potential of domain-specific BERT models in efficiently processing radiology reports. Their
ability to handle varying report lengths with remarkable accuracy makes them promising tools for radiological applications.
Limitations: medBERT.de is primarily based on data from radiology reports. The origin of the data from a single university hospital
could lead to bias.
Ethics committee - additional information: This study was approved by an ethics committee; IRB-approval number: EA2/078/22.
Image fusion using pixelwise gradient model for image fusion (PGMIF) (7 min)
Ka-Hei Cheng; Hong Kong / Hong Kong SAR China
Author Block: K-H. Cheng, J. Cai; Hong Kong/HK

Purpose: Magnetic resonance imaging (MRI) plays a pivotal role in the accurate delineation of tumours for radiotherapy. However,
conventional MRI sequences often show inconsistencies in tumour contrast across patients. This study aimed to assess the potential
of a novel multimodal image fusion method, the pixelwise gradient Model for Image Fusion (PGMIF), to improve MRI tumour contrast
and its consistency across patients.
Methods or Background: We utilised T1-w and T2-w MR images from a cohort of 80 patients. The proposed PGMIF was based on a
pixelwise gradient to capture the shape of the input images and a Generative Adversarial Network (GAN) term for capturing image
contrast. It was compared with other fusion algorithms: gradient model with maximum comparison among images (GMMCI), deep
learning model with weighted loss (DLMWL), pixelwise weighted average (PWA), and maximum of images (MoI). Two metrics were
used to test the fusion methods' performance: tumour contrast-to-noise ratio (CNR) and a refined Sobel operator analysis to measure
the edge sharpness.
Results or Findings: PGMIF surpassed in both metrics, registering a CNR of 1.237 ± 0.100. This marked a significant enhancement
compared to T1-w (0.976 ± 0.052) and T2-w MR images (1.077 ± 0.087). PGMIF also outperformed other models including GMMCI,
DLMWL, PWA, and MoI. In the Sobel operator analysis, PGMIF again showed the highest Sigmoid of Sobel Metric values for T1-w and
T2-w MR images comparisons, demonstrating the contrast amplification and edge acuity.
Conclusion: The novel PGMIF method shows its potential to enhance MRI tumour contrast while retaining the anatomical structures
from the source images. Its implementation could be useful in NPC tumour delineation.
Limitations: The ripples of the input images may be amplified in the fused images.
Funding for this study: This research was partly supported by research grants of Mainland-Hong Kong Joint Funding Scheme
(MHKJFS) (MHP/005/20), Project of Strategic Importance Fund (P0035421) and Projects of RISA (P0043001) from The Hong Kong
Polytechnic University of The Hong Kong Polytechnic University, Shenzhen Basic Research Program (JCYJ20210324130209023) of
Shenzhen Science and Technology Innovation Committee, and Health and Medical Research Fund (HMRF 09200576), the Health
Bureau, The Government of the Hong Kong Special Administrative Region.
Ethics committee - additional information: The data set used was approved by the Research Ethics Committee in Hong Kong
(Kowloon Central/Kowloon East, reference number: KC/KE-19-0085/ER-1).
Recurrence-free survival prediction in head and neck cancers using deep learning: a multicentre, multimodal approach
harnessing uncertainty estimation and counterfactual explainability (7 min)
Zohaib Salahuddin; Maastricht / Netherlands
450
Author Block: Z. Salahuddin, H. C. Woodruff, Y. Chen, X. Zhong, P. Lambin; Maastricht/NL
Purpose: This study aims to develop an end-to-end trustworthy deep learning model for predicting recurrence-free survival (RFS) in
head and neck cancers, utilising FDG-PET and CT images and automated delineations, with a focus on increasing confidence and
explainability through uncertainty predictions and counterfactual image generation.
Methods or Background: Given the prevalence and severity of head and neck cancers worldwide, an algorithm capable of
accurately predicting RFS could significantly enhance therapeutic planning and patient management. The developed adaptive 3D
resnet-50 deep learning model was trained on multimodal data (clinical data, FDG-PET, and CT images) using a multi-task logistic
regression framework. Fivefold cross-validation was performed on 378 patients from 5 different centres, and 111 patients from 2
different centers were used as an external test set. Automated delineations of tumour and lymph nodes were obtained via a modified
nnUNet. The model utilised a multi-head multi-loss function to estimate prediction uncertainty and employed a VAE-GAN for latent
space traversal, generating counterfactual images to explore and visualise hypothetical scenarios and enhance explainability.
Results or Findings: The model demonstrated a competitive c-index of 0.681 [95% CI: 0.663 - 0.694] in fivefold cross-validation and
0.671 on two external test sets. Predictions with lower uncertainty are correlated with superior performance, evidenced by a c-index
of 0.683. Kaplan-Meier curve demonstrated a significant split between low and high-risk groups. Counterfactuals revealed that both
shape and texture features from FDG-PET and CT images are important for predicting survival.
Conclusion: The developed model exhibits promising potential in providing trustworthy and interpretable RFS predictions for H&N
cancer patients, leveraging multicentre multimodal data, uncertainty estimates, and counterfactual explainability.
Limitations: The model necessitates prospective validation, and conducting an in-silico trial is imperative to assess the clinical
efficacy of the counterfactuals and uncertainty predictions.
Funding for this study: Funding for thist study was received from EuCanImage n° 952103.
Ethics committee - additional information: Institutional Review Boards of all participating PROVIDER institutions permitted use of
images and clinical data, either fully anonymised or coded, from all cases for research purposes only. Retrospective analyses were
performed in accordance with the relevant guidelines and regulations as approved by the respective institutional ethical committees
with protocol numbers: MM-JGH-CR15-50 (HGJ, CHUS, HMR, CHUM) and CER-VD 2018-01513 (CHUV). For CHUP, institutional review
board approval was waived as all patients signed informed consent for use of their data for research purposes at diagnosis. For MDA,
ethics approval was obtained from the University of Texas MD Anderson Cancer Center Institutional Review Board with protocol
number: RCR03-0800. For USZ, ethics approval was related to the clinical trial NCT01435252 entitled "A phase II study in patients
with advanced head and neck cancer of standard chemoradiation and add-on Cetuximab". For CHB, the fully anonymised data
originates from patients who consent to the use of their data for research purposes. List of PROVIDERS: HGJ: Hôpital Général Juif,
Montréal, CA; CHUS: Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, CA; HMR: Hôpital Maisonneuve-Rosemont, Montréal,
CA; CHUM: Centre Hospitalier de l’Université de Montréal, Montréal, CA; CHUV: Centre Hospitalier Universitaire Vaudois, CH; CHUP:
Centre Hospitalier Universitaire de Poitiers, FR; MDA: MD Anderson Cancer Center, Houston, Texas, USA; USZ: UniversitätsSpital
Zürich, CH; CHB: Centre Henri Becquerel, Rouen, FR.
A unified transformer-based model for characterisation and diagnosis of focal liver lesions on multiparametric MRI
images (7 min)
Zhehan Shen; Shanghai / China
Author Block: Z. Shen, F. Yan; Shanghai/CN

Purpose: The aim of this study was to develop and evaluate an AI model that incorporates liver imaging reporting and data system
(LI-RADS) criteria and other relevant radiological features for the diagnosis of FLLs using AMRI data. Apart from that, we aimed to
compare the performance of the AI model and radiologists with different levels of experience in the internal and external test set.
Methods or Background: We retrospectively collected MRI data from 1,024 patients with 1,147 FLLs who underwent contrast
enhanced abdominal MRI. The FLLs were classified into six categories: hepatocellular carcinoma (HCC), intrahepatic
cholangiocarcinoma (ICC), metastasis, cyst, haemangioma and focal nodular hyperplasia (FNH). We trained the AI model on MRI
images of 560 FLLs from January 2020 to July 2022 in the training set. We evaluated the performance of the AI model using the
internal test set (243 FLLs from July 2022 to August 2023) and the external test set (344 FLLs from public data set). We used the
DeLong method and the McNemar test to compare the performance of the AI model and two radiologists with different levels of
experience.
Results or Findings: The AI model achieved an overall accuracy of 0.87, a sensitivity of 0.85, a specificity of 0.91, and an area
under the receiver operating characteristic curve (AUC) of 0.92 for FLL diagnosis on the internal test set. The AI model outperformed
junior radiologists in terms of accuracy, sensitivity, specificity and AUC except one senior radiologist had similar accuracy and
specificity but lower sensitivity. The AI model also showed good generalisation ability across different centres, with an AUC of 0.90 for
external public data set.
Conclusion: The proposed AI model based on radiological characteristics can effectively diagnose FLLs using MRI data and can assist
radiologists in improving their diagnostic performance and efficiency.
451
Establishing robust ground truth labels to create machine learning response assessment models using an innovative
fusion technique of rectal MRI and whole mount histopathologic specimens (7 min)
Josip Ninčević; New York / United States
Author Block: N. Horvat, J. M. Santos, J. Ninčević, C. Firat, J. Heiselman, J. Chakraborty, J. Shia, J. Garcia-Aguilar, M. J. Gollub; New
York, NY/US
Purpose: MRI-based radiomics is a promising objective tool for predicting rectal cancer (RC) treatment response but lacks
generalisability. Whole mount histology (WMH) is considered the gold standard reference method for point-by-point comparison, and
radiomic models have yet to be trained using WMH to assess a rigid point-based registration method to evaluate the fusion of the
rectum between WMH and MRI accuracy.
Methods or Background: The study included 18 consecutive rectal cancer patients on neoadjuvant therapy and total mesorectal
excision from 2018 to 2021. A multimodal radiology-pathology image registration workflow was developed. First, a radiologist and
pathologist delineated the tumour bed, internal and external rectal borders, and eight corresponding MR and WMH image landmarks.
Second, automated rescaling computed point-based registration of images via delineated landmarks was performed. Third, initial rigid
alignment of MR and WMH images accounted for rectal distension differences using biomechanically constrained plane strain elastic
deformable registration. Fourth, a combination of in-house rigid registration, active contours, and finite element software performed
image registration. Fifth, a 3D slicer rendered outputs from the multimodal image fusion system to obtain accurate and precise
visualisation.
Results or Findings: Dice overlap and modified Hausdorff distance of the delineated MR and pathology images showed a
significantly good correlation between external and internal border segmentations (P-values ˂.05, comparing in each case mean
values averaged across the mean values from each of the three levels per case).
Conclusion: Deformable registration significantly improves the internal and external contour agreement over rigid point-based
registration. Establishing such a method will allow the generation of ground truth labels to predict complete response and improve
patient care by safely avoiding surgery.
Limitations: The limitations of the study are its retrospective design and small sample.
Funding for this study: This project was partly supported by the National Cancer Institute Cancer Center Core Grant P30 CA008748
and the Society of MSK (PI: Natally Horvat). The RSNA Research & Education Foundation supported the project described through
grant number RSD2302 (PI: Natally Horvat). The content is solely the authors' responsibility and does not necessarily represent the
official views of the RSNA R&E Foundation.
Ethics committee - additional information: This study was approved by the institutional review board, with a waiver for written
informed consent, and was compliant with the Health and Insurance Portability and Accountability Act.
External validation of Alzheimer's disease machine-learning models: generalisability and clinical features (7 min)
Helena Rico Pereira; Lisbon / Portugal
Author Block: H. R. Pereira1, V. Sá Diogo1, D. Prata2, H. Alexandre Ferreira1; 1Lisbon/PT, 2London/UK

Purpose: Recent studies have shown the potential of machine-learning models, based on magnetic resonance imaging (MRI)
features, in aiding the diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, these models usually lack
generalisability: most use data from the same public data sets, are trained with curated patient and healthy subject data and are not
validated with independent "real-world" data.
Methods or Background: We aimed to validate in clinical practice our previously developed models (derived also from public data
sets): model 1 – cognitively normal (CN) vs AD, showing a balanced accuracy (BAC) of 90.6%, sensitivity of 91.5%, and specificity of
89.7%; and model 2 - CN vs MCI vs AD, showing a BAC of 62.1% in the multiclassification. Additionally, we explored the features of
the misclassified cases. MRI T1-weighted MPRAGE morphometric data (computed with freesurfer 7.1.1) were used from Portuguese
hospital patients, comprising 8 AD, 9 MCI and 21 CN (19 headache and 2 depression) patients.
Results or Findings: Our model 1 showed a BAC of 97.6%, sensitivity of 100.0%, and specificity of 95.2%, whilst misclassifying one
CN as an AD patient. Model 2, on the other hand, showed a BAC of 65.8% (7 CN misclassified as 5 MCI and 2 AD; 5 MCI as 3 CN and 2
AD; 1 AD as MCI). Misclassified MCI patients showed volume changes in brain regions similar to those found in AD (amygdalar and
temporal atrophy) or CN (hippocampal sparing), and the opposite was observed for the misclassified AD as MCI patient (entorhinal
atrophy only).
Conclusion: The results suggest that our models may be of clinical use, provided that physicians frame the classification output
within anamneses and clinical findings.
Limitations: The study's main limitation was the small data set tested.
Funding for this study: This work was financially supported by Fundação para a Ciência e Tecnologia (FCT) under the projects
UIDB/00645/2020, SAICTPAC/0010/2015 and DSAIPA/DS/0065/2018. FCT has further supported HRP through the individual PhD grant
2021.08306.BD, and DP through 2022.00586.CEECIND.
Ethics committee - additional information: All the data used in this study was approved by the ethics committee of each
Portuguese hospital.
452
Towards safer imaging: a comparative study of deep learning-based denoising and iterative reconstruction in
intraindividual low-dose CT scans using an in-vivo large animal model (7 min)
Jonas Mück; Tübingen / Germany
Author Block: J. Mück, B. Stenzl, J. Hofmann, S. Afat, A. S. Brendlin; Tübingen/DE

Purpose: Computed tomography (CT) scans are a significant source of medically induced radiation exposure. Novel deep learning-
based denoising (DLD) algorithms have been shown to enable diagnostic image quality at lower radiation doses than iterative
reconstruction (IR) methods. However, most comparative studies employ low-dose simulations due to ethical constraints. We used
real intraindividual animal scans to investigate the dose reduction capabilities of a DLD algorithm in comparison to IR.
Methods or Background: Fourteen sedated pigs underwent two 100% CT scans on the same third generation dual-source scanner,
with a two-month interval between each scan. Both times, we additionally reduced the mAs to 50%, 25%, 10%, and 5%. All scans
were reconstructed using ADMIRE level 2 (IR2) and the DLD algorithm, resulting in a total of 280 data sets. Objective image quality
measures (CT number stability, noise, and contrast-to-noise ratio) were assessed. Three radiologists independently evaluated
subjective image quality, and interrater agreement was analysed using Spearman's correlation coefficient. Adequately corrected
mixed-effects modeling analysed objective and subjective image quality.
Results or Findings: Neither dose reduction nor reconstruction method negatively affected CT number stability (p>0.999). In terms
of objective image quality, the DLD algorithm achieved a 25% radiation dose while maintaining noise and contrast-to-noise ratio
comparable to 100% IR2. Interrater agreement for subjective image quality ratings was strong (r≥ 0.69, mean 0.93±0.05, 95% CI
0.92-0.94; each p <0.001). Subjective assessments indicated that DLD at 25% radiation dose was comparable to 100% IR2 in terms of
image quality, sharpness, and contrast (p≥ 0.281).
Conclusion: The DLD algorithm can achieve image quality comparable to the standard IR method but with a significant dose
reduction of up to 75%. This suggests a promising avenue for lowering patient radiation exposure without sacrificing diagnostic
quality.
Limitations: This was a single centre study with a specific hardware and software set-up.
Ethics committee - additional information: This large animal study was approved by the Regional Council (C 01/19 G) and
conducted following EU Directive No 2010/63/EU.
453
HW 11Cc - Cardiac inflammation: imaging insights and reporting strategies
Categories: Cardiac, Imaging Methods

CME Credits: 1.5
Moderator:
Iacopo Carbone; Roma / Italy


Davide Farina; Brescia / Italy
Ausami Abbas; Sherbourne st.john / United Kingdom
Luigia D'Errico; Cambridge / United Kingdom
1. To become familiar with typical and atypical imaging findings of inflammatory diseases of the myocardium and pericardium.
2. To become familiar with imaging criteria and other supporting diagnostic modalities.
3. To discuss the limits and technical drawbacks of cardiac MRI and CT.
4. To learn how to report cardiac MRI and CT using specific templates.
454
HW 11Sb - Deciphering wake-up stroke: MRI and CT insights for diagnosis and treatment

CME Credits: 1

1. To explore strokes that occur during sleep and their diagnostic and therapeutic implications using MRI.
2. To learn how to analyse strokes that occur during sleep and their impact on therapeutic decisions using CT.

455
RPS 1107 - Female genitourinary imaging

CME Credits: 1.5
Moderator:
Ivana Blazic; Belgrade / Serbia
Early regression index (ERI) for the prediction of local control in locally advanced cervical cancer (7 min)
Luca Russo; Roma / Italy
Author Block: S. Bottazzi1, L. Russo1, D. Cusumano2, R. Autorino1, L. Boldrini1, A. Amerighi1, A. Rame1, E. Sala1, B. Gui1; 1Rome/IT,
2
Olbia/IT
Purpose: Early Regression Index (ERI) is an image-based biomarker which combines gross tumour volume (GTV) measured at
staging (Vpre) and at mid-therapy (Vmid) to quantify early tumour shrinkage. It showed promising results in predicting pathological
complete response (pCR) in rectal and cervical cancer treated with neoadjuvant chemoradiotherapy. This study aims to evaluate the
feasibility of ERI in predicting local control (LC) after cisplatinum-based chemotherapy and external beam radiotherapy (CRT) followed
by brachytherapy in locally advanced cervical cancer (LACC), starting from T2-WI and apparent diffusion coefficient (ADC) maps.
Methods or Background: 91 patients with LACC (FIGO IB3-IVA) underwent MRI for staging and after CRT. GTV was delineated on
the axial oblique T2-WI and ADC map and ERI was calculated on both sequences. Response to CRT and brachytherapy was evaluated
six months after the end of treatment using MRI and PET/CT exams. LC was considered in case of the absence of residual disease.
The ERI performance was quantified by calculating the area (AUC) under the Receiver Operating Characteristic (ROC) curve and
measuring sensitivity and specificity at the best threshold value.
Results or Findings: The performance of ERI-T2 (AUC=0.84; 95% CI 0.76-0.95) was superior to that reported by ERI-ADC
(AUC=0.72; 95% CI 0.63-0.81). At the best cut-off threshold, ERI-T2 showed excellent specificity (100.0%) with limited sensitivity
(67.4%), while ERI-ADC showed high specificity (87.2%) and low sensitivity (59.1%). At subgroup analysis, both ERI-T2 and ERI-ADC
showed lower accuracy in adenocarcinoma (72.3% and 70.4%) compared to squamous subgroup (87.5% and 81.3%).
Conclusion: ERI is a promising biomarker in LACC treated with concurrent CRT. Combining ERI-T2 and ERI-ADC, it is possible to
identify poor responders after CRT thus modulating the brachytherapy boost consequently.
The performance of the node reporting and data system 1.0 (Node- RADS) in staging patients with cervical carcinoma
according to the new FIGO classification (2018) (7 min)
Roberta Valerieva Ninkova; Rome / Italy
456
Author Block: R. V. Ninkova, A. Calabrese, F. Curti, S. Riccardi, M. Gennarini, V. Miceli, A. Cupertino, C. Catalano, L. Manganaro;
Rome/IT
Purpose: The aim of this study is to explore the diagnostic accuracy, specificity, sensitivity, positive and negative predictive value
(PPV and NPV) of Node-RADS in assessing the risk of metastatic lymph node (LN) involvement of cervical cancer (CC) patients using
magnetic resonance imaging (MRI). We also evaluated the applicability and feasibility of the score among three different readers with
different years of experience.
Methods or Background: From December 2014 to June 2023, 140 patients were treated at our department. 68 patients underwent
staging MRI, neoadjuvant chemotherapy and radical surgery, and were included in our study. MRI images were retrospectively
reviewed and Node-RADS scores were assigned for the main bilateral pelvic LN stations. LNs were evaluated in a scoring category of 1
to 5 to assess risk of metastatic involvement: 1 (very low), 2 (low), 3 (equivocal), 4 (high), 5 (very high).
The adequacy of the score with respect to the actual neoplastic involvement was then evaluated by comparing it with the results of
histological examination.
Results or Findings: The resulting sensitivities, specificities, PPVs, NPVs, and accuracies were 100%, 45%, 56%, 100%, 68% for >1,
92.8%, 72.5%, 70.3%, 93.5%, 80.9% for >2, 71.4%, 100%, 100%, 83.3%, 88.2% for >3, 42.9%, 100%, 100%, 71.4%, 76.4% for >4.
The inter-observer agreement between the Node-RADS scores assigned by the senior reader compared with the scores assigned by
junior reader 1 and the scores assigned by the senior reader compared with junior reader 2 was 0.888 and 0.738, respectively.
Conclusion: The Node-RADS score is a valid system for correct and standardised evaluation of LN stations, increasing diagnostic
accuracy particularly for patients with CC. It represents an easily applicable standardisation system that can be useful in clinical
practice, improving the proper management of patients with CC.
Limitations: The main limitation is the patient sample and the retrospective nature of the study.
Ethics committee - additional information: This study was conducted in accordance with the Declaration of Helsinki.
CT‑derived radiomics analysis helps to identify ovarian clear cell carcinoma subtype in epithelial ovarian cancer: a
multicentre study (7 min)
Jing Ren; Beijing / China
Author Block: J. Ren, Y-L. He, L. Mao, Y. Li, Z. Yuan, Z. Y. Jin, X. Li, H-D. Xue; Beijing/CN
Purpose: The study aimed to assess the value of CT-derived radiomics in preoperatively identifying ovarian clear cell carcinoma
(CCC) subtype in epithelial ovarian cancer (EOC) using multicenter datasets.
Methods or Background: A total of 457 patients with EOCs were retrospectively recruited from three medical centres: Centre A
(n=349), divided into training (n=280) and internal testing (n=69) sets based on surgery date, while Centre B (n=74) and Centre C
(n=34) were external testing sets. Patients were grouped into CCC or non-CCC subtype based on postoperative pathology. We
documented six clinical characteristics and assessed ten radiological characteristics of each lesion by two radiologists. Three models
to identify CCC subtype in EOC were built: a radiomic signature using selected radiomic features, a traditional model with selected
clinical and radiological characteristics, and an integrated model combining radiomic signature, clinical characteristics, and
radiological characteristics. These models were constructed using JMIM feature selection and logistic regression in 10-fold cross-
validation. Diagnostic performance was assessed in internal and external test sets.
Results or Findings: The mean age of the 457 patients was 53.7 ± 10.4 years. CCC accounted for 21.0% (96/457) of all EOC cases.
The integrated model, which utilized the radiomic signature plus one clinical characteristic and three radiological characteristics,
demonstrated better performance over the traditional model and the radiomic signature. The integrated model achieved AUCs of
0.890 in the internal testing set, 0.885 in external testing set 1, and 0.781 in external testing set 2, surpassing the traditional model
with AUCs of 0.840, 0.884, and 0.735, and the radiomic signature with AUCs of 0.811, 0.836, and 0.778.
Conclusion: CT-derived radiomics analysis proved to be valuable in the identification of the CCC subtype in EOC, suggesting its
potential to enhance subtype-specific therapeutic approaches.
Funding for this study: Funding was provided by Natural Science Foundation of China [grant No. 81901829]; and National High
Level Hospital Clinical Research Funding [grant No. 2022-PUMCH-A-004 & 2022-PUMCH-A-109].
Ethics committee - additional information: This retrospective study was approved by the local institutional review board, which
waived the requirement for written informed consent.
Diagnostic performance and accuracy of the O-RADS MRI scoring system in adnexal masses (7 min)
Sena Bozer; Ankara / Turkey
457
Author Block: S. Bozer, D. Kuru Öz, G. A. Erden; Ankara/TR
Purpose: The ovarian-adnexal reporting and data system-magnetic resonance imaging (O-RADS MRI) guideline determines the
malignancy risk of adnexal lesions and provides a common language between radiologists and clinicians. The aim of this study was to
determine the diagnostic performance of the O-RADS MRI scoring system.
Methods or Background: Patients who underwent pelvic MRI with any clinical prediagnosis between January 2011-April 2023 were
screened. Exclusion criteria were as fallows, history of malignancy or gynaecological surgery for non-malignant reasons, O-RADS 1
category lesions and contraindication for contrast administration. MR images were evaluated by two radiologists who were unaware of
clinical and laboratory findings. Lesions with different scores were reviewed by consensus. Histopathologic results were accepted as
the reference standard. Lesions that did not have a histopathologic diagnosis but regressed on follow-up imaging or remained stable
for at least six months were considered benign. Lesions in the O-RADS 4 category and above were considered malignant. For
statistical analysis, borderline tumours were included in the malignant category.
Results or Findings: A total of 209 adnexal lesions, of which 184 were benign, three were borderline, and 22 were malignant, were
included in the study. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of the O-RADS MRI
score were 100%, 97.2%, 83.3%, 100%, and 97.6%. Interobserver agreement was very high (k=0.95).
Conclusion: The sensitivity, specificity and overall accuracy of the O-RADS MRI score in this study were similar to those reported for
the original score. Our findings support the use of the O-RADS MRI score in the evaluation of adnexal masses with high interobserver
reliability without loss of diagnostic accuracy.
Limitations: The limitations of the study are small sample size, relatively low number of malignant lesions and retrospective nature.
Ethics committee - additional information: This study was approved by the Ethics committee.
Role of shear wave elastography in prediction of preterm labour (7 min)

Manish Jha; Dehradun / India
Author Block: M. Jha; Dehradun/IN

Purpose: The objective of this study was to examine the stiffness of uterine cervix by means of shear wave elastography in low-risk
pregnant women. It shall provide us with quantitative as well as qualitative data regarding the cervical stiffness which shall be used
as a quantitative measure of the cervical ripening process.
Methods or Background: 150 patients were enrolled in the study prospectively. Patient with normal conception without any history
of preterm delivery were enrolled in the study. All the patients underwent shear wave elastography by trans vaginal USG once at
16-22 weeks period of gestation and later at 28-32 weeks period of gestation. The colour maps for shear wave elastography were
placed at five locations- anterior wall, posterior wall, internal os, external os and cervical canal. Patients were divided into term and
preterm delivery groups based on period of gestation at delivery.
Results or Findings: The shear wave velocity (SWV) of uterine cervix showed a steady decrease with advancing gestational age in
all the regions of cervix by as much as 3-14% in 3rd trimester as compared to second trimester indicating a negative correlation.
Statistically significant negative correlation was seen with values obtained in anterior wall, internal os and external os. The most
significant fall was seen in anterior part of cervix with a mean reduction of 10.8% which was statistically significant. The change in
anterior wall SWV, internal os SWV and external os shear wave velocity showed a diagnostic specificity of 94.1%, 68%, 73.5% and
diagnostic accuracy of 68.7%, 47.5% and 66.7% respectively.
Conclusion: Change in shear wave velocity has a good potential to predict preterm labour in low-risk pregnancies with fair diagnostic
accuracy and good interobserver agreement.
Limitations: The limitations of this study are the limited patient sample size, and the exclusion of high-risk pregnancies
Ethics committee - additional information: This study was approved by the Ethical committee.
Thrombohaematoma-associated placentomegaly: opening the “black box” of pregnancy (7 min)

Elizaveta Aleksandrovna Kirillova; St. Petersburg / Russia
458
Author Block: E. A. Kirillova1, I. Mashchenko1, P. Kozlova1, L. Kashapova2, T. Mamiseishvili2, E. Shelepova1, I. Zazerskaya1, G.
Trufanov1; 1St. Petersburg/RU, 2Moscow/RU
Purpose: Placentomegaly may be associated with severe foetal and maternal complications. The aim of this study was to provide
useful diagnostic information on the differential diagnosis of placentomegaly and to describe the specific features of
thrombohaematoma-associated cases.
Methods or Background: Placental thrombohaematomas (PTHs) represent rare cases of placentomegaly that are associated with
poor pregnancy outcomes and can be classified as subamniotic, subchorionic, retroplacental or marginal haemorrhages.
Results or Findings: We encountered six cases of PTHs in pregnant women aged 29.50 ± 5.96 years. The diagnosis was unclear on
a routine ultrasound exam and required additional confirmation using MRI of the placenta performed at 23.33 ± 2.58 weeks.
Unfavourable pregnancy outcome (foetal demise; n=3) included (1) acute massive retroplacental haematomas with a fluid-fluid level
as a result of a sedimentation effect, (2) in two cases, all four types of PTHs were observed, (3) the mean gestational age at the time
of delivery was 22 weeks.
Unfavourable pregnancy outcome (live births; n=3) involved (1) all three patients having subamniotic PTHs (massive [n=2] and local
[n=1] PTHs) and placental villous oedema, (2) two cases of preterm birth at 27 GW (in both cases of massive PTHs), and one case of
full-term birth at 37 GW, (3) the mean gestational age at the time of delivery was 31 weeks.
Some newborns demonstrated congenital anomalies (n=2), neurologic deficits and gastrointestinal complications (n=3 each).
Respiratory support was required in three cases.
Conclusion: MRI of the placenta is a valuable tool for the differential diagnosis of placentomegaly. Early diagnosis is important for
risk stratification and prenatal outcome prediction. Special caution should be exercised in case of thrombohaematoma-associated
placentomegaly as this finding may be associated with a higher risk of preterm birth and poor neonatal outcomes.
Funding for this study: No funding was provided for this study
Magnetic resonance protocol without contrast to characterise adnexal masses indeterminate at ultrasonography: a
Camilla Panico; Rome / Italy
Author Block: C. Panico, G. Avesani, V. Simeon, M. Mangialardi, P. Sciaccotta, B. Gui, E. Sala; Rome/IT
Purpose: Appropriate management of patients with adnexal masses needs an accurate characterisation. Up to 20% of adnexal
masses remains uncharacterised on US and MRI is needed for full characterisation. The O–RADS MRI score relies on intravenous
gadolinium contrast. The purpose of this study was to show interim data of a multicentre prospective external evaluation of the non-
contrast MRI adnex Score (NCMS) in characterising the adnexal masses.
Methods or Background: 73 patients from six Italian centres were evaluated. NCMS was used to predict the malignancy of the
ovarian lesions. The score is based on five categories (four and five indicate malignancy). One senior and one junior radiologist per
centre read the images, unaware of the patients' clinical information except for age. Presence of malignant tumours was verified
during histopathology after surgery or during the radiological follow-up after 12 months.
Diagnostic accuracy in terms of the percentage of correctly classified diagnoses was measured. Agreement between senior and junior
radiologist has been evaluated performing kappa statistics.
Results or Findings: 26 patients were classified as malignant. Malignant lesions were correctly classified in 78.08% of cases (Conf.
Int. 68.38% - 86.92%). The agreement between senior and junior radiologist was 93.15% (under an expected agreement of 53.74%),
with a kappa statistic of 0.79 (Conf. Int. 0.59-0.985).
Conclusion: The results on the accuracy and reproducibility of the diagnosis are in line with the potential use of the NCMS in clinical
practice when contrast medium cannot be used for logistic or patient factors.
Limitations: The limitation of this study is the number of observations being low (only 20% of the number of observations needed
according to the sample size calculation); the expansion of the number of cases in this multicentric study will overcome this problem.
Ethics committee - additional information: The proposal to the Ethical Committee of Fondazione Policlinico Gemelli (coordinating
center) was presented on June 21st, 2021 and approved. Each centre had their participation approved by their own ethical
committee.
Diagnostic value of magnetic resonance imaging with diffusion-weighted imaging for differentiating low-grade
endometrial stromal sarcoma from benign atypical leiomyoma (7 min)
Soomin Park; Seoul / Korea, Republic of
459
Author Block: S. Park, S. E. Rha, H. Kim, Y. R. Shin; Seoul/KR
Purpose: The purpose of this study was to evaluate the diagnostic value of known qualitative magnetic resonance imaging (MRI)
features and diffusion-weighted imaging (DWI)-based quantitative parameters in the differentiation of uterine low-grade endometrial
stromal sarcoma (LGESS) from benign atypical leiomyoma (ALM).
Methods or Background: This retrospective study includes 113 women (ALM= 93, LGESS= 20) imaged with pelvic MRI with DWI
prior to surgery. Two readers independently evaluated each lesion for conventional qualitative features as well as quantitative
parameters including the mean ADC value and the relative contrast ratio (RCR) on high b-value DWI. The diagnostic performance of
both qualitative and quantitative parameters in discriminating LGESS and ALM was assessed.
Results or Findings: The qualitative MRI features that significantly differed between LGESS and ALM were ill-defined or infiltrative
margin (p=0.018), irregular or nodular shape (p<0.001), intratumoral low-SI band on T2WI (p<0.001), intramyometrial nodular
extension (p< 0.001), nodule-in-nodule appearance (p= 0.017), and cystic changes (p= 0.010). The mean ADC value of LGESS (0.99
± 0.29 x 10−3mm2/s) was significantly lower than that of ALM (1.25 ± 0.31 x 10−3mm2/s) (p< 0.001) and the mean RCR of LGESS
was 8.3 ± 5.5, which was significantly higher than that of ALM, 4.2 ± 2.7 (p< 0.001).
Conclusion: Although there are substantial overlaps in conventional imaging features between LGESS and ALM, applying quantitative
parameters based on DWI and ADC map in addition to qualitative imaging features may help distinguish LGESS from ALM.
Limitations: The study is retrospective. There are heterogeneous MRI acquisition parameters, but no major differences were
detected in the protocols among different equipment. In addition, the study includes a relatively small number of patients with low-
grade ESS. Nonetheless, it's important to highlight that low-grade ESS is an exceedingly rare tumor, and our research encompasses a
more substantial number of cases compared to previous reports.
Ethics committee - additional information: The study was approved by the catholic university of Korea, IRB
“Cysts margination” sign in distinguishing ovarian thecoma-fibroma groups from other ovarian cystic and solid
tumours (7 min)
Chang Li; Guangzhou / China
Author Block: C. Li, M. Sun, R. Xu, J. Guan; Guangzhou/CN

Purpose: The purpose of this study was to recognise and analyse the cause of “cysts margination” sign in ovarian thecoma-fibroma
groups (OTFG). And to investigate the value of this specific sign in the differential diagnosis between OTFG and other ovarian cystic
and solid tumours.
Methods or Background: A total of 112 patients with surgically proven OTFG were screened consecutively from January 2016 to
December 2022. Preoperative CT/MR examinations were reviewed by one radiologist with 20 years of experience in
genital/reproductive imaging. The “cysts margination” sign was clearly recognised in 25 patients (group 1), while absent in the
remaining 87 patients (group 2). Clinical data were recorded and compared between the two groups. Preoperative CT/MR imaging of
25 OTFG patients with “cysts margination” sign and 50 matched patients with surgically proven other ovarian cystic and solid
tumours were retrospectively reviewed by two independent radiologists (five and nine years of experience) before and after learning
the sign. The diagnostic accuracy of OTFG and other ovarian tumours was calculated.
Results or Findings: 22.3% (25/112) of patients with OTFG presented with “cysts margination” sign. Histopathology confirmed
these peripheral cysts to be ovarian cysts. The age (56±15 vs 43±16, p= 0.02), menopausal state (p< 0.001), and presence of
endometrium thickening (p< 0.001) were significantly different between the two groups. With the addition of the sign into radiological
evaluation, the diagnostic accuracy of OTFG by two independent radiologists improved from 10% and 35% to 90% and 90%,
respectively.
Conclusion: “Cysts margination” sign can be seen in around 20% of OTFG patients, mostly in peri- or post-menopausal state, and
may be associated with the elevated oestrogen production of the tumour. Familiarity with this specific sign can significantly increase
the diagnostic efficacy of OTFG in menopausal women, thereby avoiding unnecessary further examination or treatment.
Ethics committee - additional information: The study was approved by the First Affiliated Hospital, Sun Yat-Sen University
Hybrıd F18-FDG PET/MRI in endometrial cancer: staging accuracy, quantitative analysis, and imaging characteristics of
histological subtypes (7 min)
Diğdem Kuru Öz; Ankara / Turkey
460
Author Block: D. Kuru Öz, M. Araz, S. N. Yılmazer Zorlu, B. Varlı, C. Soydal, S. Taşkın, E. Ozkan, G. A. Erden, O. Kücük; Ankara/TR
Purpose: The aim of this study is to evaluate the diagnostic performance of F18-FDG PET/MRI in endometrial cancer (EC) staging,
determine the individual contributions of PET and MRI, compare PET/CT and PET/MRI findings, quantitatively assess the primary tumor
(PT) and metastatic lymph nodes (LN), and investigate the differences in quantitative measurements among histological subtypes.
Methods or Background: Between February 2020 and March 2022, a total of 44 patients who were diagnosed with EC via biopsy
were included in the study. Following whole-body PET/CT, pelvic PET/MRI was performed. For quantitative analysis, the apparent
diffusion coefficient (ADC) values for PT and LN were measured with MRI, while the maximum standard uptake value(SUVmax),
metabolic tumor volume(MTV), and total lesion glycolysis(TLG) values were measured with PET/CT and PET/MRI. For statistical
analysis Spearman correlation and Mann-Whitney U tests were performed.
Results or Findings: Out of the 44 patients, 39 (88.6%) had concordance between PET/MRI staging and pathological staging, while 5
(11.3%) were discordant.
There were nine patients (20.4%) with pelvic and/or para-aortic LN metastasis. In three patients (6.8%) with suspected LN
involvement on MRI, PET accurately predicted LN involvement with pathological activity.
In LN evaluation, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for PET/CT were 66.6%,
93.9%, 66.6%, 91.1%, and 88%, respectively. For PET/MRI, the corresponding values were 88.8%, 88.2%, 66.6%, 96.7%, and 88.3%,
respectively.
For both PT and pathological LN, no strong correlation was found between ADC and SUVmax values (r> -0.009, p>
0.095,respectively).
Conclusion: PET/MRI, can be used in preoperative staging for the accurate assessment of both the primary tumor and nodal
involvement. It offers high negative predictive value and has the potential to prevent unnecessary lymph node dissections while
providing a high diagnostic performance as a hybrid imaging method.
Limitations: The limitation of this study is the small sample size.
Ethics committee - additional information: The Institutional Ethics Committee approved this retrospective study protocol and
waived informed consent
Imaging of peritoneal metastases in ovarian cancer using MDCT, MRI and FDG PET/CT: a systematic review and meta-
analysis (7 min)
Martha Tzoumpa; Ioannina / Greece
Author Block: M. Tzoumpa1, T. Siempis2, G. Alexiou1, M. Argyropoulou1, A. C. Tsili1; 1Ioannina/GR, 2Belfast/UK

Purpose: The purpose of this study was to compare the diagnostic performance of multidetector CT (MDCT), MRI including diffusion-
weighted imaging and FDG PET/CT in the detection of peritoneal metastases (PMs) in ovarian cancer (OC) patients.
Methods or Background: A comprehensive search was performed for articles published from January 2000 to February 2023, in
PubMed database. Inclusion criteria were the following: diagnosis/suspicion of PMs in patients with ovarian/fallopian/primary
peritoneal cancer; initial staging or suspicion of recurrence (primary outcome); MDCT, MRI and/or FDG PET/CT performed for the
detection of PMs; population of at least 10 patients; at least three data sets available for each imaging modality; surgical results,
histopathologic analysis and/or radiologic follow-up, used as reference standard; per-patient and per-region data; and data for
calculating sensitivity and specificity reported.
Results or Findings: Thirty-three studies were assessed, including 487 women with OC and PMs. On a per-patient basis, MRI (P =
0.03) and FDG PET/CT (P < 0.01) had highest sensitivity compared to MDCT. MRI and PET/CT had comparable sensitivities (P = 0.84).
On a per-lesion analysis, no differences in sensitivity estimates were noted between MDCT and MRI (P = 0.25), MDCT and FDG PET/CT
(P = 0.68), MRI and FDG PET/CT (P = 0.35).
Conclusion: Based on our results, FDG PET/CT and MRI had a higher diagnostic performance in the detection of PMs compared to
MDCT, on a per-patient analysis. No differences were found on a per-lesion basis.
FDG PET/CT and MRI can be considered equivalent alternatives for the detection of PMs in OC. MDCT can be used as an alternative.
Limitations: The limitations of this study are due to selection bias, and the heterogeneity of the studies.
The pattern of radiological first relapse in ovarian cancer differs between different maintenance therapies: a
retrospective analysis of a tertiary referral centre (7 min)
Giacomo Avesani; Rome / Italy
461
Author Block: G. Avesani, L. D'Erme, A. Iacono, C. Panico, S. Bottazzi, F. Capomacchia, C. Marchetti, B. Gui, E. Sala; Rome/IT
Purpose: The aim of this study was to evaluate the radiological distribution of disease at the first recurrence in patients with
advanced ovarian cancer treated with different maintenance therapies (PARP-inhibitors versus Bevacizumab).
Methods or Background: We retrospectively collected data on patients with recurrent advanced ovarian cancer from January 2017
to December 2022, who had III-IV FIGO stage at the diagnosis, a complete debulking (primary or interval surgery) followed by
adjuvant chemotherapy and maintenance therapy. We included those with an available contrast-enhanced CT at the recurrence time
(clinical, serological or radiological).
We evaluated whether the presence of recurrence was visible (visible/non-visible solid tissue), the type (macronodular/micronodular)
and the number of localisations (oligometastatic (<4 sites)/multi-metastatic) and if the relapse was only in lymph nodes.
Results or Findings: We evaluated 80 patients; 40 had Bevacizumab and 40 had PARP-i.
The incidence of recurrence without any clearly visible soft tissue was significantly higher during Bevacizumab (20% versus 10%; p=
0.04). Micronodular diffuse relapse was more frequent in patients treated with Bevacizumab (58% versus 30%; p= 0.03).
Oligometastatic recurrence was more frequent in PARP-i (45% versus 24%; p= 0.03). No difference was found in only nodal relapse
between the two groups (18% versus 15%, p> 0.05).
Conclusion: The recurrence pattern in ovarian cancer differs based on the administered maintenance therapy. Secondary
cytoreduction, which can be performed in oligometastatic disease, is more frequent during PARP-i maintenance. Radiologists should
know this to make the correct diagnosis, especially avoiding delayed diagnosis of recurrence in patients treated with Bevacizumab.
Limitations: This is a retrospective study with a small sample size.
Ethics committee - additional information: This study was approved with the approval code: 6031
462
RPS 1103 - Evolving cardiac MRI techniques: development and clinical applications

CME Credits: 1.5
Moderator:
Dietrich Beitzke; Vienna / Austria
Blood native T1 time for estimating synthetic haematocrit and extracellular volume: derivation of a conversion formula
at 3T (7 min)
Johannes Schmid; Graz / Austria
Author Block: M. Puseljic, C. Reiter, M. Fuchsjäger, J. Schmid; Graz/AT

Purpose: Cardiac magnetic resonance imaging (CMR) has emerged as a vital tool for characterising cardiac disorders, with T1
mapping offering quantitative insight into myocardial tissue. The aim of the study was to correlate haematocrit (Hct) levels from blood
sampling with native T1 times in the left ventricular (LV) blood pool to derive a formula for estimating synthetic Hct levels (Hctsyn)
and for calculation of synthetic extracellular volume fraction (ECVsyn).
Methods or Background: In this retrospective analysis, native T1 times in the LV blood pool (T1blood) were correlated with Hct
levels from blood sampling within 24 hours (Hct24h) in 250 CMR scans (3T, MOLLI5(3)3), divided into a derivation and validation
cohort. A linear regression equation was derived to calculate Hctsyn and ECVsyn, which was then externally validated.
Results or Findings: In the derivation cohort (n=167), Hct24h exhibited a strong association with T1blood (r = -0.711, p < 0.001).
The resultant regression equation, Hctsyn = 1/T1blood * 1355.52 - 0.310, enabled calculation of Hctsyn. In the validation cohort
(n=83), Hctsyn displayed a good correlation with Hct24h (r = 0.726, p < 0.001). Additionally, ECVsyn, calculated using Hctsyn,
demonstrated an excellent correlation with ECV24h (r = 0.940, p < 0.001). Bland-Altman plots confirmed minimal bias in ECVsyn
estimation (0.28%).
Conclusion: A formula for estimating Hctsyn from T1blood was derived, enabling the calculation of ECVsyn. This approach offers a
non-invasive alternative to blood sampling and demonstrates good agreement with established methods, enhancing the clinical utility
of CMR.
Limitations: Applicability of the results to different scanner setups may be limited by variations in scanner and sequence
characteristics.
Ethics committee - additional information: This study was approved by an ethics committee; internal reference number: 35-324
ex: 22/23.
A novel approach to measuring synthetic extracellular volume without invasive blood sampling: the sheva-3t cmr study
(7 min)
Cesare Mantini; Chieti / Italy
463
Author Block: C. Mantini, A. Sorella, V. Di Mascio, G. Bisaccia, M. Foglietta, D. Calvo Garcia, L. D'Angelo, S. Gallina, F. Ricci; Chieti/IT
Purpose: Existing CMR methods for extracellular volume (ECV) measurement necessitate haematocrit level measurements, which
can be impractical in clinical settings. We aimed to derive and validate a multiparametric model for synthetic ECV assessment in
clinical 3T CMR.
Methods or Background: We recruited 505 consecutive patients undergoing clinical 3T CMR exams with 48-hour haematocrit
sampling. Participants were randomly split into derivation (n=405) and validation (n=100) cohorts. Native T1 was measured in both
left ventricular (LV) and right ventricular (RV) blood pools. We derived and validated a multiparametric model for synthetic
haematocrit estimation, including covariates selected by multivariate linear regression analysis. The conventional ECV was calculated
using a standard blood haematocrit value. Synthetic ECVs were obtained from LV and RV T1 values using Fent's equation and from
the 4-factor synthetic haematocrit. We assessed the correlation, agreement, accuracy of classification, and trueness between
synthetic and conventional ECVs.
Results or Findings: In the derivation cohort, sex, heart rate, and LV and RV native T1 values were selected as independent
predictors of haematocrit and built into a 4-factor model. The 4-factor synthetic haematocrit showed better correlation with blood
sampling than the LV and RV synthetic haematocrits (R2:0.380; R2:0.341; R2:0.316, respectively). The 4-factor ECV model showed
good correlation with conventional ECV, similar to LV and RV ECVs (R2:0.834, R2:0.823, and R2:0.815, respectively), yet yielded the
lowest bias (4-factor ECV:-0.024; RV-ECV:-0.162; LV-ECV:-1.067). These findings were confirmed in the validation cohort (4-factor ECV:
R2:0.835; bias:-0.26; LV-ECV: R2:0.807; bias:-0.38; RV-ECV: R2:0.777; bias:-1.22).The 4-factor model exhibited substantial agreement
(Cohen's kappa: 0.64) and trueness compared with conventional ECV.
Conclusion: The novel 4-factor synthetic model improves precision and trueness for haematocrit and ECV estimation. Our findings
support broader utilisation of synthetic ECV in 3T settings, obviating the need for invasive blood sampling while ensuring clinical
accuracy and reliability.
Limitations: This was a retrospective singlecentre study.
Ethics committee - additional information: No information was provdided by the submitter.
T1 mapping and extracellular volume for predicting pre-heart failure in children with Duchenne Muscular Dystrophy (7
min)
Xinyuan Zhang; Chengdu / China
Author Block: X. Zhang, H. Xu, H. Fu, Y-K. Guo; Chengdu/CN

Purpose: Pre-heart failure (pre-HF) is being recognised as an essential stage known to progress to symptomatic HF. However, early
detection of HF in Duchenne Muscular Dystrophy (DMD) is difficult due to progressive muscle necrosis, physical inactivity, and lack of
awareness of their deteriorating cardiac function. Therefore, our study aimed to investigate the prognostic value of cardiac magnetic
resonance (CMR) for pre-HF in children with DMD.‍
Methods or Background: A total of 113 patients with DMD (age 8.0 [7.0, 9.0] years; 111 boys, 98.2%) were included from July 2018
to July 2022. The study outcome was pre-HF. According to the AHA/ACC/HFSA guideline for the management of HF, pre-HF was
defined as patients without current or prior symptoms of HF but with evidence of abnormal cardiac biomarkers, functional or
structural heart disease. Survival estimates were calculated by Kaplan-Meier curves with the log-rank test.
Results or Findings: During a mean follow-up of 24.2 ± 11.6 months, a total of 34 patients reached pre-HF. Univariate Cox
regression analyses and Multivariate stepwise analyses showed that native T1 (hazard ratio [HR]: 1.014, 95% CI: 1.005-1.022; p
=0.002) in model 1 and ECV (HR: 1.366, 95% CI: 1.219-1.531; p <0.001) in model 2 had significant prognostic associations with pre-
HF. The cut-off value of ECV was 30.1% (AUC: 0.782; sensitivity: 79.4%; specificity: 75.9%). A native T1 cut-off value of 1294.83 ms
had a sensitivity of 55.9%, a specificity of 96.2, and an AUC of 0.790. The cut-off value of ECV and native T1 significantly differed
between patients with and without pre-HF in both the LGE-positive and -negative groups (all P < 0.001).
Conclusion: T1 mapping and ECV had prognostic value for pre-HF in DMD patients, which provided optimal risk stratification for early
cardiac involvement in DMD patients.‍
Limitations: This was a retrospective single centre analysis.
Ethics committee - additional information: The Institutional Review Boards of our hospital approved this retrospective study.
Joint bright- and black-blood late gadolinium enhancement and T1-rho mapping for robust myocardial scar imaging (7
min)
Victor De Villedon De Naide; Pessac / France
464
Author Block: V. De Villedon De Naide1, M. Stuber2, J. H. Zhang2, K. Narceau1, P. Gut2, V. Nogues1, M. Villegas-Martinez1, H. Cochet1,
A. Bustin1; 1Pessac/FR, 2Lausanne/CH
Purpose: Bright-blood sequences are used to retrieve heart anatomy information, while black-blood late gadolinium enhancement
has shown potential for scar detection. Contrast-agent-free T1-rho mapping is a promising technology for quantifying
cardiomyopathies. The aim of this study is to provide a single sequence that combines the strengths of black-blood scar detection,
bright-blood scar localisation, and T1-rho scar quantification.
Methods or Background: The proposed 2D whole-heart SPOT1-rho acquisition is a single-shot breath-held sequence gathering
black- and bright-blood images that are averaged for optimal detection and localisation of scarred tissue. For the bright-blood shots,
five shots with increasing T1-rho preparation times were acquired, to generate a T1-rho map, for scar quantification.
A phantom (T1MES) experiment was first conducted with PSIR, SPOT, T1-rho mapping and the proposed SPOT1-rho acquisition. The
obtained T1-rho values were compared, along with mean signal intensities from bright- and black-blood images. The same sequences
were then prospectively tested in two patients with myocardial infarction and one healthy subject using a 1.5T Siemens Aera scanner.
3-slice short-axis images were acquired 15min post-injection of Gadolinium. T1-rho values were measured in both remote and injured
myocardial regions.
Results or Findings: In phantom, an excellent correlation was observed between the reference T1-rho values and those obtained
using the SPOT1-rho sequence, along with consistent signal intensities. In patients, elevated myocardial T1-rho values were measured
in injured areas, closely matching values obtained from the reference T1-rho mapping sequence.
Conclusion: The proposed SPOT1-rho combines bright- and black-blood imaging and T1-rho mapping to enhance scar detection,
localisation, and quantification, offering a promising and versatile tool for myocardial assessment in both research and clinical
settings.
Limitations: Validation of the SPOT1-rho sequence requires a greater patient cohort, while clinical application of T1-rho mapping is
still at an early stage.
Funding for this study: This project was supported by funding from the French National Research Agency under grant agreements
Equipex MUSIC ANR-11-EQPX-0030, ANR-22-CPJ2-0009-01, ANR-21-CE17-0034-01, Programme d'Investissements d'Avenir ANR-10-
IAHU04-LIRYC. This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020
research and innovation programme (grant agreement No101076351).
Cardiac magnetic resonance in infarct-like myocarditis: transmural late gadolinium enhancement is associated with
long-term outcome (7 min)
Alexander Isaak; Bonn / Germany
Author Block: A. Isaak, J. Wirtz, D. Kravchenko, N. Mesropyan, U. I. Attenberger, C. Öztürk, S. Zimmer, D. Kütting, J. A. Luetkens;
Bonn/DE
Purpose: The aim of this study was to evaluate the prognostic value of cardiac magnetic resonance (CMR) imaging parameters in
terms of the occurrence of MACE in patients with infarct-like myocarditis.
Methods or Background: Patients with acute myocarditis confirmed by CMR between 2007 and 2020 were retrospectively
identified. Only patients with infarct-like presentation (chest pain and ST-segment elevation on electrocardiogram and/or troponin
elevation) were included into analysis. Functional and morphological imaging analyses were performed. Late gadolinium
enhancement (LGE) was qualitatively and quantitively assessed. The association between different parameters and the occurrence of
major adverse cardiac events (MACE; including cardiovascular death, new onset of acute symptoms or heart failure symptoms,
implantation of pacemaker or defibrillators) within 5 years after discharge was tested using a univariable and multivariable Cox
regression and Kaplan-Meier analysis.
Results or Findings: 130/345 patients (38%) had infarct-like presentation (mean age, 40±19 years; 97 men, 75%). LGE lesions
involved mostly the subepicardium (111/130 patients [85%]; midwall: 45/130 patients [35%]; both subepicardium and midwall:
27/130 patients [21%]). Septal segments were involved in 42/130 patients (32%). Transmural LGE extension was present in 15/130
patients (12%). The median extent of LGE was 7% (IQR, 4-10). Median duration of follow-up was 19.3 months (IQR, 4.5-53), and MACE
occurred in 18/130 patients (14%). Univariable Cox regression analyses revealed an association between MACE and both, LGE extent
and transmural LGE extension. In multivariable Cox regression analysis, transmural extension of LGE was an independent predictor
for MACE (Hazard ratio, 6.34; 95% CI: 2.29, 17.49; P <0.001). Patients with transmural extension of LGE had a shorter event-free time
on Kaplan-Meier analysis (mean [95%CI], 136 weeks [70, 198] vs 236 weeks [221, 252]; log rank P <0.001).
Conclusion: Transmural LGE on CMR seems to be associated with long-term occurrence of MACE in infarct-like myocarditis.
Limitations: The retrospective nature of the study was identified as a limitation.
Funding for this study: A.I. was funded by the BONFOR Research Commission of the Medical Faculty Bonn (BONFOR-
Forschungskommission der Medizinischen Fakultät Bonn) and by the German Research Foundation (Deutsche
Forschungsgemeinschaft, DFG) under Germany's Excellence Strategy—EXC2151—390873048.
Left ventricular diastolic dysfunction grading from a single MR 4D flow measurement (7 min)
465
Clemens Reiter; Graz / Austria
Author Block: C. Reiter, U. Reiter, E. Kolesnik, C. Kräuter, A. Schmidt, D. Scherr, M. Fuchsjäger, G. Reiter; Graz/AT
Purpose: The aim of this study was to evaluate the feasibility of assessing left ventricular (LV) diastolic dysfunction from a single MR
4D flow measurement using the established echocardiographic algorithm.
Methods or Background: 94 prospectively recruited patients underwent echocardiography and nearterm MR whole-heart 4D flow
imaging. LV ejection fraction (EF), left atrial volume index (LAVI), early- (E) and late (A) diastolic transmitral velocities, early diastolic
myocardial tissue velocity (e'), and tricuspid regurgitation velocity (TR) were determined from echocardiography and used for grading
of diastolic dysfunction according to the 2016 ASE/EACVI algorithm. Comparable variables were derived from MR 4D flow: LVEF and
LAVI were evaluated from multiplanar reformatted magnitude images. E, A and e' were analysed from 4D flow velocity fields, and
mean pulmonary arterial pressure (mPAP) was assessed from duration of vortex in the main pulmonary artery. LV diastolic
dysfunction was graded according to the same algorithm as in echocardiography. Relationships between echocardiographic and 4D
flow parameters were analysed by correlation analysis, the agreement for grading between modalities for LV diastolic dysfunction was
investigated by contingency table analysis.
Results or Findings: LV diastolic dysfunction of grade 0, indeterminate, grade I, grade II and grade III were found in 51, 9, 13, 13
and 8 subjects by echocardiography. All volumetric and velocity parameters from MR and echocardiography correlated strongly
(r=0.75-0.92). In cases where TR was assessable with echocardiography, a strong correlation to MR-derived mPAP was found
(r=0.81). Using cut-offs of LAVI >50 ml/m2 and mPAP >25mmHg, there was excellent agreement between 4D flow and
echocardiographic grading of diastolic dysfunction with a weighted kappa of 0.85.
Conclusion: The assessment of left ventricular diastolic function from a single 4D flow measurement is possible with excellent
agreement to echocardiography.
Funding for this study: Funding was received from the OeNB Anniversary Fund 17934.
Ethics committee - additional information: This study was approved by an ethics committee; ClinicalTrials.gov, NCT01728597 &
ClinicalTrials.gov, NCT03253835.
Non-invasive diagnosis of pulmonary hypertension: a comparison between MR 4D flow and echocardiography (7 min)
Gert Reiter; Graz / Austria
Author Block: G. Reiter, C. Reiter, C. Kräuter, E. Kolesnik, D. Scherr, A. Schmidt, M. Fuchsjäger, U. Reiter; Graz/AT
Purpose: MR time-resolved 3-directional phase-contrast (4D flow) imaging allows estimation of mean pulmonary arterial pressure
(mPAP) and diagnosis of pulmonary hypertension (PH) from the duration (t_vortex) of vortical blood flow in the main pulmonary
artery. The purpose of the study was to investigate the relationship between 4D flow and standard echocardiographic measures for
pressure estimation and diagnosis of PH.
Methods or Background: 94 patients were prospectively investigated by transthoracic echocardiography and near-term 3T 4D flow
imaging. Echocardiographic evaluation included the measurement of the tricuspid regurgitant jet velocity (TR) and calculation of the
tricuspid regurgitant pressure gradient (TRPG). 4D flow data were employed to derive t_vortex and calculate mPAP. The relationship
between TRPG and t_vortex was analysed by regression analysis, the agreement on the presence of PH by contingency table analysis.
Results or Findings: A tricuspid jet was identified in 69 patients (73%). For these subjects the relationship between t_vortex and
TRPG was well described by a linear model of t_vortex on TRPG (R=0.88). The definitions of PH by mPAP≥25 mmHg and by TR>2.8
m/s for echocardiography were consistent with the linear model and optimised the agreement of 4D flow and echocardiography for
diagnosis of PH. The resulting kappa-values were 0.94 for subjects with visible tricuspid jet and 0.90 for all subjects. Using the
threshold of mPAP>20 mmHg for diagnosis of PH resulted in the cut-off TR≥2.5 m/s, yielding kappa-values of 0.79 for subjects with
visible tricuspid jet and 0.67 for the entire population.
Conclusion: There is a strong relationship between 4D flow and echocardiographic measures for diagnosis of PH. While the old PH-
cut-off mPAP≥25 mmHg shows high agreement between 4D flow-based and echocardiographic PH diagnosis, the recently introduced
PH-cut-off mPAP>20 mmHg reduces this agreement substantially.
Limitations: There were no invasive measurements.
Funding for this study: Funding for this study was received from the OeNB Anniversary Fund 17934.
Ethics committee - additional information: This study was approved by the ethics committee of the Medical University of Graz,
Austria; ClinicalTrials.gov: NCT01728597, NCT03253835.
Coronary microvascular dysfunction associates with left ventricular outflow tract and aorta diameter ratio in non-
obstructive hypertrophic cardiomyopathy: insights from CMR first-pass perfusion imaging (7 min)
Wei Gao; Kunming / China
466
Author Block: W. Gao, T. Qi, Z. Li, J. Deng, W. Chen; Kunming/CN
Purpose: Coronary microvascular dysfunction (CMD) may mediate the development of adverse cardiovascular outcomes in patients
with non-obstructive hypertrophic cardiomyopathy (NOHCM). This study aims to assess CMD in NOHCM and to analyse its risk factors.
Methods or Background: The study cohort consisted of 47 NOHCM and 18 obstructive hypertrophic cardiomyopathy (OHCM), and
28 healthy controls (HCs). Left ventricular outflow tract (LVOT) and aorta (AO) diameter ratio, maximal wall thickness, LGE, and
perfusion parameters, including time to peak (Tpeak), peak signal intensity (SIpeak), and upslope were calculated globally and
segmentally. Univariable and multivariable linear regression analyses were performed to assess the potential risk factors for CMD.
Results or Findings: Globally, OHCM group had significantly increased Tpeak and reduced upslope (both P <0.001) compared with
NOHCM group; NOHCM group had significantly prolonged Tpeak and decreased upslope compared with HCs (both P <0 .001).
Myocardial segments with no hypertrophy and LGE (n = 217) in NOHCM patients had a longer Tpeak and lower upslope compared
with segments of HCs (n = 448) (Tpeak: 29.98 ± 3.64 vs. 25.11 ± 4.11s; upslope: 18.99 ± 4.88 vs. 23.84 ± 6.99, both P < 0.001). In
NOHCM patients, LVOT/AO diameter ratio was independently associated with Tpeak (adjusted β: -0.501, P <0.001) and upslope
(adjusted β: -0.499, P =0.002).
Conclusion: CMD can occur in NOHCM patients at rest, even in seemingly normal myocardial segments. The LVOT/AO diameter ratio
was an independent risk factor of microvascular dysfunction, which implies it has potential potential predictive value in CMD in
NOHCM patients.
Ethics committee - additional information: This study was approved by the ethics committee of Kunming Medical University.
Quantitative perfusion CMR with adenosine for the assessment of coronary microvascular disease in heart failure with
preserved ejection fraction patients (7 min)
Ana Ezponda Casajus; Pamplona / Spain
Author Block: A. Ezponda Casajus1, C. Mbongo1, P. Kellman2, A. González Miqueo1, G. Bastarrika Alemañ1; 1Pamplona/ES, 2Bethseda,
MD/US
Purpose: Coronary microvascular disease (CMD) is a main mechanism in the development of diastolic dysfunction, a condition
frequently observed in heart failure with preserved ejection fraction (HFpEF) patients. Quantitative stress perfusion cardiovascular
magnetic resonance (CMR) is a useful non-invasive technique for the assessment of CMD. The aim of our study was to compare
myocardial perfusion reserve (MPR) and myocardial blood flow (MBF) values by using an adenosine stress/rest quantitative perfusion
CMR protocol in stage-B (asymptomatic with known risk factors and diastolic dysfunction) and stage-C (with signs and symptoms)
HFpEF patients.
Methods or Background: Between December 2021 and November 2022, 39 HFpEF patients (23 in stage-B and 16 in C) underwent
quantitative CMR with adenosine for the assessment of CMD. None of the patients had a prior history of severe CAD. Global MBF
during rest, stress and MPR indices were calculated using automated pixelwise quantitative myocardial perfusion mapping.
Demographics and cardiovascular disorders and risk factors were recorded for the different groups of the HFpEF cohort.
Results or Findings: Stage-B and stage-C patients do not present statistically significant differences in cardiovascular risk factors
nor in prior medical history. Stage-C patients were significantly older than stage B patients (76.8n ±5.2 vs 67.1 ±8.4 years old, p
<0.001). Stage-C HFpEF patients presented a significantly lower median of global MPR compared to stage B HFpEF patients (2.31, IQR
1.72-2.74 vs 3.20, IQR 2.80-3.55; p =0,004). Regarding stress perfusion values, global median endocardial BF was also significantly
lower in patients at stage C (1.60, IQR 1.22-2.05 vs 2.23, IQR 1.66-2.55; p =0,044). There were no significant differences in rest MBF
values between groups.
Conclusion: Stage-C HFpEF patients present significant lower values of non-invasive biomarkers of MCD evaluated with quantitative
perfusion CMR, than asymptomatic HFpEF patients.
Funding for this study: CRUCIAL (H2020): this project has received funding from the SESAR Joint Undertaking under the European
Union's Horizon 2020 research and innovation programme under grant agreement No 848109.
Ethics committee - additional information: This study was approved by the ethics committee of the Clínica Universidad de
Navarra.
Residual myocardial hyperaemia in regadenoson stress/rest quantitative perfusion CMR (7 min)

Ana Ezponda Casajus; Pamplona / Spain
467
Author Block: A. Ezponda Casajus, C. Mbongo, M. B. Barrio Piqueras, M. R. López de la Torre Carretero, P. Kellman, M. Vidorreta Díaz
de Cerio, G. Bastarrika Alemañ; Pamplona/ES
Purpose: Regadenoson is a recently introduced vasodilator for stress CMR that possesses a relatively long half-life. As this fact may
impact myocardial blood flow (MBF) and myocardial perfusion reserve (MPR) estimated when quantitative stress/rest CMR perfusion is
performed, this study sought to investigate the presence of residual myocardial hyperaemia on the recovery phase in patients
undergoing stress CMR.
Methods or Background: Fifty patients with clinical indication for stress CMR underwent quantitative perfusion imaging in resting
conditions, after regadenoson-induced hyperaemia (400 mcg, 5 mL), and 10 minutes after recovery with aminophylline. A total dose
of 0.15 mmol/kg of Gadobutrol was administered. Studies showing late gadolinium enhancement (LGE) were excluded. Global
myocardial blood flow during rest (MBF-rest), stress (MBF-stress) and recovery (MBF-recov) and MPR indices (MPR-rest and MPR-recov)
were calculated using automated pixel-wise quantitative myocardial perfusion mapping.
Results or Findings: A total of 33 patients (25 male, mean age of 61.4 ±2.2 years) were included in the analysis. Seventeen studies
showing LGE (15 transmural, 2 subendocardial) were excluded. Global MBF-rest and MBF-stress were 0.81 ±0.26 and 2.11 ±0.73,
respectively. After recovery with aminophylline, myocardial perfusion did not return to the resting values (MBF-recov of 0.91 ±0.33)
and statistically differed from MBF-rest (P <0.01), suggesting residual myocardial hyperaemia. This resulted in an abnormally low
MPR-recov (2.51 ±0.85) with respect to MPR-rest (2.68 ±0.93) (P <0.01).
Conclusion: Despite the use of aminophylline to reverse the vasodilator effect, MBF does not return to resting values and MBF-recov
cannot be used as a substitute for MBF-rest when regadenoson is used. Consequently, a rest/stress protocol is advised for
quantitative CMR perfusion to obtain accurate MBF and MPR parameters.
Limitations: This study has included a reduced number of participants.
Funding for this study: No funding has been received for this study.
spCMR as a predictor validity for MACE (7 min)

Pierpaolo Palumbo; L'Aquila / Italy
Author Block: P. Palumbo, D. Boccetti, S. Lamja, A. Barile, E. Di Cesare; L'Aquila/IT

Purpose: The aim of our study is to analyse the impact of stress perfusion CMR on the management of patients with known or
suspected angina CAD.
Methods or Background: Coronary Artery disease (CAD) is a much-discussed topic because of its burden on healthcare systems
and patients' quality of life. However, there is still debate about the best strategy to control angina symptoms and to reduce future
events.
This was a historical, prospective study. 164 patients with known or suspected CAD who underwent 3.0-T stress CMR were enrolled
(22 F, 142 M; mean age 65.9 years CI 95% 64.67-67.11).
We recorded all clinical information including angina symptoms and all major cardiac events (MACE) occurred during the follow-up,
including cardiovascular death, arrhythmias and acute myocardial infarction. All therapeutic strategies adopted from clinicians were
collected.
Results or Findings: Mean follow-up was 2.8 years (IQ range 0.9-3.9). During the follow up 52 patients reported MACE (32%), while
70 patients showed modification of angina symptoms (43 patients reported improvement of symtoms).
In spite of the way angina symptoms developed, none of the therapeutic strategies analysed were advantageous in angina
improvement.
Ischaemia shows a high stratification validity for clinical evolution in IHD patients (OR 2.18, 95%CI 1.12-4.26; P-value 0.022). among
therapeutic strategies, only revascularisation showed a significant impaction on MACE occurrence (OR 0.28 P-value 0.016).
Conclusion: IHD is a dynamic disease. CMR-derived ischaemia revealed a good prediction validity for MACE occurrence and adequate
therapeutic strategies should be oriented on spCMR stratification.
Limitations: The number of patients, time of follow-up as well as the retrospective nature of the study were identified as limitations.
468

Moderators:
Interview (30 min)
469
MD 4 - Ovarian cancer: new developments in imaging and treatment - recommendations for clinical
Categories: Genitourinary, Multidisciplinary, Oncologic Imaging, Research

CME Credits: 1
Moderator:

1. To highlight new developments in the treatment of ovarian cancer.

2. To critically review recent evidence in imaging and put this in perspective of new treatment developments.
The gyno-oncological surgeon's perspective (8 min)

Stacey Bryan; London / United Kingdom
The medical oncologist's perspectives (8 min)

Stacey Bryan; London / United Kingdom

470

Date: March 1, 2024 | 07:50 - 08:00 CET
Moderators:
Interview (30 min)

471

ETC Level: LEVEL II
Date: March 1, 2024 | 08:00 - 08:30 CET
CME Credits: 0.5
Moderators:


3. To gain deep knowledge of abdominal radiology and have the opportunity to grow


472
RPS 1203 - Spectral CT: new developments and clinical applications (part 1)

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Nils Robrecht Planken; Amsterdam / Netherlands
Iodine quantification with a CdZnTe clinical prototype photon-counting scanner at reduced radiation dose: initial
cardiac phantom results and reproducibility analysis (7 min)
Author Block: T. W. Holmes1, S. Sharma2, S. Ross2, T. Labno2, R. Zhang2, S. Wu2, Z. Yu2, R. Thompson2, A. Pourmorteza1; 1Atlanta,
GA/US, 2Vernon Hills, IL/US
Purpose: This study aimed to assess effectiveness of a clinical prototype CT scanner with CdZnTe-based photon-counting detector
(PCD) CT technology in spectral mode, specifically regarding the measurement of iodinated contrast agents at both standard and
reduced radiation dosage levels.
Methods or Background: We 3D-printed a series of phantoms mimicking coronary arteries with lumen diameter of 3.5 mm with
stents, soft plaque, and hard plaques placed in them. The phantoms were placed inside two water tanks with 12 cm diameters. The
water tanks were scanned on the prototype scanner at 120 kVp and 200, 100, and 50 mAs exposures in 6-bin spectral mode. The 50
mAs scan was repeated twice to assess reproducibility. We measured accuracy, stability with respect to radiation dose, and
reproducibility of iodine through Bland-Altman analysis of values measured in circular regions of interest (ROIs) ranging from 2.2 mm
to 10 mm in diameter.
Results or Findings: We measured very small bias of -0.15 and good 95% confidence interval (CI) [-1.52 1.21] for concentrations of
iodine ranging from 0 to 35 compared to calibrated values (all values are reported as mg I/mL). At 50% radiation dose reduction we
observed bias of 0.35 with CI=[-0.32 1.04] compared to the high-dose (200 mAs) measurements. At 75% dose reduction the bias and
CI were 0.45 and [-0.49 1.44]. The 75% dose-reduced iodine maps showed good reproducibility with bias of 0.01 mg I/mL and
confidence interval of [-0.66 0.69].
Conclusion: Accurate quantification of iodine is an important tool in characterisation of coronary artery plaques as well as for making
detailed measurements of myocardial perfusion. PCD-CT shows potential for making accurate iodine measurements in dose-reduced
settings.
Limitations: Limitations of this study were: this was a limited pilot study with 2 phantoms; more experiments mimicking different
patient sizes are warranted.
Funding for this study: Funding was received through a sponsored research agreement with Canon Medical Research USA.
Impact of photon-counting detector CT-based calcium removal algorithm on coronary stenosis grading and CT-FFR
values (7 min)
Zsofia Jokkel; Budapest / Hungary
473
Author Block: Z. Jokkel1, B. Vattay1, M. Boussoussou1, M. Vecsey-Nagy1, M. Kolossvary1, M. Kiss2, B. Sipos1, P. Maurovich-Horvat1, B.
Szilveszter1; 1Budapest/HU, 2Forcheim/DE
Purpose: CT-based fractional flow reserve (CT-FFR) provides non-invasive evaluation of lesion specific ischaemia. However, blooming
artefacts from heavily calcified plaques could hinder the assessment of luminal stenosis and alter CT-FFR values. The objective of this
study was to investigate the effect of a calcium removal algorithm on the grading of coronary stenoses and CT-FFR values in
comparison to measurements based on the standard image reconstruction used for clinical evaluation.
Methods or Background: We enrolled consecutive patients with calcified plaques and intermediate coronary artery stenosis
(30-90%) who underwent a dual-source photon counting CT (NAEOTOM Alpha, Siemens Healthineers) for suspected coronary artery
disease. A total of 54 lesions were analysed and two reconstructions were used (standard: 0.4 mm slice thickness, Bv40, 70keV;
PureLumen calcium removal algorithm: 0.8mm slice thickness, Qr40, 70keV). Anatomical landmarks were used as fiducial markers to
ensure that the same location was used for CT-FFR assessment. Area-stenosis and CT-FFR values distal to the plaques were measured
using syngo.via Frontier CT-cFFR software (Siemens Healthineers). Comparison of stenosis and CT-FFR values was performed using
Wilcoxon signed rank test.
Results or Findings: A total of 25 patients (mean age 64.5±8.1years, 24% female) were included in the study. Median calcium
score and Q1-Q3: 258 (154.35-401.75). Mean area stenosis of the 54 lesions was 50.5±22.5% on standard reconstruction compared
to 43.5±20.9% on PureLumen images (p<0.001). CT-FFR values were higher for PureLumen reconstructed lesions compared to
standard reconstruction: 0.87±0.11 vs 0.82±0.15 (p<0.001).
Conclusion: PureLumen calcium removal technique might allow for more accurate assessment of the degree of stenosis by removing
calcium blooming artefacts. In patients with intermediate stenosis, this resulted in lower stenosis grades and higher CT-FFR values,
which may affect further therapeutic decisions.
Limitations: Correlation with invasive coronarography is warranted and will be available in a subpopulation.
Funding for this study: Funding was provided by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences.
Project number RRF-2.3.1-21-2022-00003 has been implemented with the support provided by the European Union.
Ethics committee - additional information: This study was approved by the Scientific and Research Ethics Committee of the
Hungarian Medical Research Council (IV/665-3/2022/EKU) and was carried out in accordance with the tenets of the Declaration of
Helsinki.
Stenosis quantification on virtual non-calcium images from photon-counting detector CT (7 min)

1 2 2 2 3 1 1 1 1
Author Block: V. Mergen , S. Rusek , F. Civaia , P. Rossi , R. Rajagopal , R. Manka , A. Candreva , M. Eberhard , H. Alkadhi ;
1 2 3
Zurich/CH, Monaco/MC, Jodhpur/IN
Purpose: This study aimed to evaluate, in patients with known or suspected coronary artery disease, the feasibility and accuracy of
quantification of calcified coronary stenoses using virtual non-calcium (VNCa) images in CCTA with photon-counting detector (PCD) CT
compared with quantitative coronary angiography (QCA).
Methods or Background: This retrospective, institutional review board approved study included consecutive patients with calcified
coronary artery plaques undergoing CCTA with PCD-CT and invasive coronary angiography between July and December 2022. Virtual
monoenergetic images (VMI) and VNCa images were reconstructed. Diameter stenoses were quantified on VMI and VNCa images by
two readers. Stenosis measurements from 3D-QCA served as the standard of reference. Measurements were compared using Bland-
Altman analyses, Wilcoxon tests, intraclass correlation coefficients (ICC), and weighted-Kappa analysis.
Results or Findings: Thirty patients (mean age, 64 years ± 8 [standard deviation]; 26 men) with 81 coronary stenoses from
calcified plaques were included. Ten of the 81 stenoses (12%) had to be excluded because of erroneous plaque subtraction on VNCa
images. Median diameter stenosis determined on 3D-QCA was 22% (interquartile range, 11-35%; total range, 4-88%). As compared
with 3D-QCA, VMI overestimated diameter stenoses (mean differences -7%, p<.001, ICC:.84 and -10%, p<.001, ICC:.87 for reader 1
and 2, respectively), whereas VNCa images showed similar diameter stenoses (mean differences 1%, p=.07, ICC:.93 and 0%, p=.68,
ICC:.94 for reader 1 and 2, respectively).
Conclusion: This preliminary experience suggests that virtual calcium removal in CCTA with PCD-CT is feasible in a high proportion of
calcified, minimal to moderate stenoses and has the potential to improve the quantification of coronary stenoses.
Limitations: First, this single centre retrospective study included only a limited number of patients and stenoses. Second, only a
limited number of moderate and severe stenoses were present.
Improvement of the visualisation of coronary stents by ultra-high-resolution photon-counting detector CT (7 min)

Le Qin; Shanghai / China
474
Author Block: L. Qin, S. Zhou, F. Yan, W. Yang; Shanghai/CN
Purpose: This study aimed to investigate the quantitative and qualitative characteristics of coronary stents of ultra-high-resolution
coronary CT angiography (CCTA) with different reconstruction kernels on photon counting detector-CT (PCD-CT) in patients after
percutaneous coronary intervention (PCI).
Methods or Background: From July 2023 to September 2023, 68 patients (132 stents) after PCI were enrolled to undergo CCTA on a
dual-source PCD-CT system. Standard resolution images with thickness of 0.6 mm, increment of 0.4 mm and kernel of Bv48, and UHR
images with thickness and increment of 0.2 mm and kernel of Bv48, Bv56, Bv60, Bv64, Bv72 and Bv76 were reconstructed. CT
attenuation and noise were measured in the aorta root and within the stents. Stent and in-stent diameters were also assessed.
Subjective image quality was evaluated by a Likert-5 point scale.
Results or Findings: Image noise significantly increased with the reduction of image thickness and elevation of kernels
(2.6±5.7～136.3±18.5 HU, P<0.001). UHR images with Bv72 and Bv76 had the smallest differences between aorta and in-stent CT
values (14.8±68.3 HU, 6.2±73.9 HU, P<0.001), and differences between stent and in-stent diameters (1.2±0.2 mm, 1.2±0.2 mm,
P<0.001). UHR images with Bv72 had the largest in-stent lumen diameter (2.3±0.5 mm, P<0.001) and the smallest differences
between stent diameter and nominal diameter (0.4±0.3 mm, P<0.001) compared to other images. Subjective analysis showed that
images with Bv72 had the most superior effect of blooming artefact reduction (5 [5, 5], P<0.001), and in-stent lumen and stent
demonstration (5 [5, 5], P<0.001). Bv72 also had the highest diagnostic confidence (5 [5, 5], P<0.001).
Conclusion: UHR CCTA on PCD-CT results in the significantly improved visualisation of coronary stents and Bv72 is the optimal
reconstruction kernel to show the stent struts and in-stent lumen.
Limitations: An identified limitation was that the in-stent diameter and diagnostic accuracy for in-stent re-stenosis were not
compared with the quantitative invasive coronary angiography.
Ethics committee - additional information: This study was approved by Ruijin hospital ethics committee, Shanghai Jiaotong
University School of Medicine.
Prognostic value of deep learning based RCA PCAT and plaque volume beyond CT-FFR in patients with stent
implantation (7 min)
Zengfa Huang; Wuhan / China
Author Block: Z. Huang, X. Du, R. Tang, Y. Ding, W. Wang, Z. Li, X. Wang, X. Wang; Wuhan/CN
Purpose: The study aimed to investigate the prognostic value of deep learning based pericoronary adipose tissue attenuation
computed tomography (PCAT) and plaque volume beyond coronary computed tomography angiography (CTA) -derived fractional flow
reserve (CT-FFR) in patients with percutaneous coronary intervention (PCI).
Methods or Background: A total of 183 patients with PCI who underwent coronary CTA were included in this retrospectively study.
Imaging assessment included PCAT, plaque volume and CT-FFR which were performed using an artificial intelligence (AI) assisted
workstation. Kaplan-Meier and multivariate Cox regression were used to estimate major adverse cardiovascular events (MACE)
including non-fatal myocardial infraction (MI), stroke and mortality.
Results or Findings: In total, 22 (12%) MACE occurred during the median follow-up of 38.0 months (interquartile range 34.6-54.6
months). Kaplan-Meier survival curves indicated that right coronary artery (RCA) PCAT (p=0.007) and plaque volume (p=0.008) were
significantly associated with increasing MACE. Multivariable Cox regression analysis showed that RCA PCAT [hazard ratios (HR): 2.94,
95% CI: 1.15-7.50, p=0.025] and plaque volume (HR: 3.91, 95% CI: 1.20-12.75, p=0.024) were independent predictors of MACE after
adjusting for clinical risk factors. However, CT-FFR was not independently associated with MACE in multivariable Cox regression
(p=0.271).
Conclusion: Deep learning based RCA PCAT and plaque volume derived from coronary CTA was found to be more strongly associated
with MACE than CT-FFR in patients with PCI.
Limitations: An identified limitation was that this study used a composited end point of nonfatal MI, stroke and death rather than the
cause of death, as this was unavailable.
Ethics committee - additional information: The study was approved by the ethics committee of the Central Hospital of Wuhan,
Tongji Medical College, Huazhong University of Science and Technology.
Photon-counting CT angiography in pre-TAVR aortic annulus sizing: high-pitch spectral vs low-pitch ultra-high-
resolution CT-angiography (7 min)
Muhammad Taha Hagar; Freiburg im Breisgau / Germany
475
Author Block: M. T. Hagar1, T. Jennerjahn1, D. Westermann1, S. Faby2, C. Von Zur Mühlen1, C. L. Schlett1, F. Bamberg1, T. Krauß1, M.
Soschynski1; 1Freiburg im Breisgau/DE, 2Forchheim/DE
Purpose: The study aimed to assess the diagnostic efficacy of low-pitch, retrospective ultra-high-resolution CT-angiography (UHR-
CTA) and high-pitch, prospective spectral CTA (HPS-CTA) utilising a first-generation, dual-source photon-counting detector CT scanner
in preprocedural planning for transcatheter aortic valve replacement (TAVR).
Methods or Background: Consecutive patients with severe aortic valve stenosis were referred clinically, underwent UHR-CTA
(collimation: 120 x 0.2 mm) and HPS-CTA (144 x 0.4 mm with full spectral information) for TAVR planning and were retrospectively
enrolled. Radiation doses were extracted from CT reports. Two radiologists independently evaluated UHR-CTA and HPS-CTA datasets,
analysing aortic annulus image quality (on a 4-point scale) and measuring the aortic annulus area (AAA) and perimeter (AAP). These
measurements then informed hypothetical valve prosthesis sizing according to vendor recommendations.
Results or Findings: Among a total of 64 patients (mean age 81, ±7 SD; 28 women), HPS-CTA was associated with lower radiation
exposure (4.1 mSv) than UHR-CTA (12.6 mSv, p<0.001, while UHR-CTA exhibited superior image quality (median score: 1, IQR: 1–2
vs. 2, IQR: 2–3; p<0.001). The AAA and AAP assessments from both CTA datasets were strongly correlated (Pearson r2 = 0.857),
leading to consistent valve prosthesis sizing in 89% of patients. However, those with lower image quality on HPS-CTA (score ≥3) more
frequently encountered varied sizing recommendations in hypothetical aortic valve prosthesis selection.
Conclusion: UHR-CTA and HPS-CTA by photon-counting CT technology provide reliable aortic annular evaluations for TAVR planning.
While UHR-CTA provides enhanced image quality, HPS-CTA features lower radiation exposure. However, impaired image quality in
HPS-CTA may necessitate additional UHR-CTA to avoid valve sizing discrepancies.
Limitations: Identified limitations were: (1) the study's generalisability may be limited by its focus on primarily older patients, (2)
this was a single-centre study with a retrospective design, and (3) hypothetical valve prosthesis sizing based on aortic annular
measurements might overlook other patient-specific factors in real-world decision-making.
Funding for this study: Funding was received from the Ministerium für Wirtschaft, Arbeit und Wohnungsbau Baden-Württemberg
(35-4223.10/20).
Ethics committee - additional information: Informed written consent was obtained from all patients. This study was approved
with the approval code: 21–2469; approval date: 21-09-2021.
First clinical experience with a novel reconstruction algorithm to reduce stair-step artefacts in step-and-shoot cardiac
photon-counting detector CT (7 min)
Lukas Jakob Moser; Zürich / Switzerland
1 1 2 1 1 2
Author Block: L. J. Moser , V. Mergen , M. Eberhard , H. Alkadhi ; Zurich/CH, Unterseen/CH
Purpose: One of the most common cardiac CT scan modes is prospective ECG-triggering (or step-and-shoot (SAS)). Due to the stack-
wise acquisition over usually several heart cycles, SAS may be associated with "stair-step" artefacts in the transitional areas, where
stacks are misaligned or overlapping due to irregular heart rates. This study aimed to evaluate a novel cardiac imaging reconstruction
algorithm designed for photon-counting detector CT to reduce the occurrence and severity of such artefacts.
Methods or Background: In this clinical study, we included 50 consecutive patients who underwent cardiac CT on a clinical dual-
source photon-counting detector CT scanner in the SAS mode. Each scan was reconstructed without and with the ZeeFree algorithm,
which uses overlapping information to perform a non-rigid registration between the borders of two adjacent sub-volumes respectively,
to minimise potentially occurring stair-step artefacts. The presence or absence and, if present, the extent of stair-step artefacts were
rated on a visual analogue four-point scale.
Results or Findings: Forty of the total 800 coronary segments (5%) had stair-step artefacts, from which 12 (30%) led to a non-
diagnostic image quality (1.5% of all segments). The novel reconstruction algorithm significantly reduced the number and extent of
these artefacts compared to standard reconstructions (median score ZeeFree: 1; median score standard reconstruction: 3; p<0.001).
From the initially rated 12 non-diagnostic segments in standard reconstruction, 9 (75%) improved to a diagnostic image quality using
the algorithm.
Conclusion: Our results demonstrate the feasibility and effectiveness of a novel reconstruction algorithm, which significantly reduces
stair-step artefacts in coronary CT angiography with dual-source photon-counting detector CT acquired in the SAS mode.
Limitations: Identified limitations were (1) that this was a single centre experience and (2) the effect on diagnostic accuracy was not
evaluated.
Ethics committee - additional information: This was a retrospective study design and a waiver of written informed consent was
obtained.
Myocardial extracellular volume fraction from late iodine enhancement with spectral detector computed tomography
for risk stratification in non-ischaemic heart failure (7 min)
Jie Deng; Kunming / China
476
Author Block: J. Deng, X. Fan, W. Chen; Kunming/CN
Purpose: Our previous study demonstrated that CT-ECV derived from late iodine enhancement (LIE) can serve as an excellent
alternative to CMR-ECV in non-invasively quantifying diffused myocardial fibrosis. The study aimed to further explore using CT-ECV
from late iodine enhancement (LIE) via dual-layer spectral detector CT (SDCT) to stratify disease risk in non-ischaemic heart failure
(NIHF) patients.
Methods or Background: Eighty-two NIHF patients (52±13 years, 21 female) underwent SDCT. CT-ECV was calculated based on LIE
images per AHA's 16-segmentation. Clinical data of NIHF patients were reviewed and patients were followed up for major adverse
cardiovascular event (MACE): hospital admission for heart failure and all-cause mortality. Receiver operating curve (ROC curve) and
area under curve (AUC) were used to evaluate the prediction model. ROC curve and Yoden index determined the optimal cut-off value
of CT-ECV. Kaplan-Meier curve and log-rank test were used to analyse the relationship between MACE and CT-ECV of patients with
NIHF.
Results or Findings: Clinical outcome data were collected from 82 NIHF patients after a 10-month median follow-up (interquartile
range: 5 to 13). Final status check was performed during March 2023 and 7 patients were lost to follow-up. Of 75 patients, 31 (41.3%)
had MACE, including 28 (37.3%) who were hospitalised for heart failure. All-cause mortality occurred in 3 (4%) patients. The ROC
curves demonstrated CT-ECV ≥ 31.28 % to be the optimal cut-off point for MACE with 83.9% sensitivity, 75% specificity and the area
under the ROC curve = 0.863 (95% CI 0.782 to 0.944). Kaplan–Meier survival curves and Log-rank test demonstrate that NIHF
patients with CT-ECV ≥ 31.28% had a higher probability of MACE than NIHF patients with CT-ECV < 31.28%.
Conclusion: CT-ECV derived from LIE can assist in risk stratification for non-ischaemic heart failure patients.
Limitations: An identified limitation was that CT-ECV involves exposure to additional radiation.
Ethics committee - additional information: This study was approved by the First Affiliated Hospital of Kunming Medical
University.
477
RPS 1209 - New insights in musculoskeletal interventional radiology
Categories: Artificial Intelligence & Machine Learning, Interventional Oncologic Radiology, Interventional Radiology, Musculoskeletal,
Vascular
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Evangelos Petsatodis; THESSALONIKI / Greece
Feasibility and safety of percutaneous CT-guided bone biopsies in cancer patients using a patient-mounted robotic
system: a retrospective analysis of 40 consecutive biopsies (7 min)
Agnieszka Witkowska; New York / United States
Author Block: A. Witkowska, E. Nadia, A. M. A. H. Moussa, D. Sarkar, E. Lis, F. H. Cornelis; New York, NY/US
Purpose: The objective of this study was to evaluate the feasibility and safety of percutaneous computed tomography (CT)-guided
bone biopsies performed by a patient-mounted robotic system with steering capabilities in patients with cancer.
Methods or Background: This was a retrospective observational study of 40 consecutive biopsies in 39 outpatients (17 female, 22
male; median age: 65.5, interquartile range: 54.8–71). Median body mass index was 28.7 kg/m2 (24.2–31.7). Biopsies were performed
in the pelvis (n=19), spine (n=8), ribs (n=5), shoulder (n=5), femur (n=2), and sternum (n=1). Median size of lesions was 26 mm
(17–32). The lesions biopsied were lytic (14/40, 35%), mixed (16/40, 40%), or sclerotic (10/40, 25%). The robot advanced the needle
on demand, allowing operators to not wear lead. For mixed and sclerotic lesions, needles were manually exchanged over a k-wire
prior to drilling for lesion access. The samples were manually collected.
Results or Findings: Technical success was 100%. Median trajectory length was 55.5 mm (47–73). Intermediary checkpoints were
utilised in eight biopsies. Median time of needle insertion from skin to target was 19 seconds (15–31). The median time from first to
final scan was 21 minutes (17–37). The median procedure time was 30 minutes (23.5–36). The median dose length product and
effective dose were 536.6 mGy.cm (396.2–837.7) and 7.1 mSv (4.7–10.8 mSv), respectively. No adverse events were reported. The
diagnostic yield was 72.5% (29/40) for cancer, but 100% (40/40) were considered diagnostic.
Conclusion: Percutaneous CT-guided bone biopsy performed by a patient-mounted robot allows for high technical success and
diagnostic yield with reduced complication rate and low procedure time.
Limitations: An identified limitation was that this was a retrospective, non-comparative study.
Ethics committee - additional information: This study was approved by the Memorial Sloan Kettering Cancer Center.
Usefulness of augmented reality with computed tomography fusion in chronic total occlusion percutaneous femoral
interventions (7 min)
Irene Nieri; Paris / France
478
Author Block: I. Nieri, J. M. Pernes, K. Najeh, S. Kalinger, C. Del Giudice; Paris/FR
Purpose: Femoral interventions could be challenging, particularly in the presence of CTO. The aim of the study was to evaluate the
use of augmented reality superpositioning of a CT fusion to treat femoral lesions in patients with claudication and critical limb
ischaemia in a retrospective registry
Methods or Background: From November 2022 to August 2023, a total of 47 patients (study group mean age 81 years old, 41
male), 32 of whom were suffering with claudication and 15 with critical limb ischaemia, with femoral CTO lesion were treated by
endovascular recanalisation supported by augmented reality (Vessel Assist, GE Healthcare) to guide the procedure. A preprocedural
CT scan was performed on all patients, that was used to fuse the 3D volume rendering with the fluoroscopy during the procedure in
an augmented reality pattern. These patients were compared to a historical cohort of 120 matched patients (control group).
The primary outcome of the study dose area product (DAP) and total cumulated air kerma (CAK) were evaluated. Technical success,
fluoroscopy time and amount of used contrast agent were also evaluated.
Results or Findings: Technical success was obtained in all procedures. No difference in terms of risk factor and lesion length was
observed between the study group and control group. DAP and CAK were significantly lower in the study group compared to the
control group (3611.5 versus 15408.0 mGy*cm², p<0.001 – CAK 19.8 versus 68.5 mGy, p<0.001). Fluoroscopy time was significantly
inferior compared to the study group (528 versus 1108s, p<0.001). Moreover, in the study group a significant inferior amount of
contrast agent was used (67.9 versus 156.0 ml, p<0.001).
Conclusion: The use of augmented reality CT fusion during femoral recanalisation may reduce x-ray exposure and fluoroscopy time
without any difference in technical success. A reduced amount of contrast media was used in the study group.
Limitations: The bias of a matched control study was an identified limitation.
Ethics committee - additional information: Ethics committee approval was not required for this retrospective study.
Study protocol: percutaneous treatment in patients with lumbar spinal stenosis. Rationale and design of a phase III,
multicentre, randomised, controlled trial (7 min)
Alessandro Napoli; Rome / Italy
Author Block: A. Napoli, G. Alfieri, M. Mattone, S. Perotti, C. Catalano; Rome/IT

Purpose: The study aimed to compare the efficacy and safety of percutaneous therapy and surgical treatment for the management
of lumbar spinal stenosis.
Methods or Background: This represents the protocol for a multicentre, prospective, randomised, controlled, phase III trial
comparing percutaneous therapy (PT) and surgical treatment (ST) for the management of lumbar spinal stenosis. Eligible patients are
between 65 and 85 years with radiologically proven lumbar spinal stenosis causing neurogenic claudication. Patients will be randomly
assigned to one of the two groups, PT and ST. Patients in the PT group will undergo treatment under local anaesthesia and mild
sedation in prone position using a minimally invasive percutaneous approach under fluoroscopic guidance. Primary endpoints will be
overall pain as measured by a numerical rating scale (range 0-10), functional disability as measured by the Zurich Claudication
Questionnaire and adverse events. Outcomes will be assessed at 3-, 6- and 12-months following treatment. Statistical analysis will be
performed on an intention-to-treat basis.
Results or Findings: We determined that a minimum sample size equal to 216 patients is needed to detect between-group
differences of at least 2 NRS points at 3-month follow-up, allowing for a standard deviation of 2.5. Assumptions for the study design
considered a two-tailed significance with alpha 5% and 90% statistical power.
Conclusion: A phase III trial will provide evidence on the actual efficacy of percutaneous treatment of patients with lumbar spinal
stenosis. The intended size of the study population is sufficiently large to detect differences between the two groups.
Limitations: Unedified limitations were (1) that this was an open-label trial and (2) the absence of a sham or placebo group.
Percutaneous computer tomography (CT): guided treatment of lumbar joint synovial cysts (7 min)
Sonia Triggiani; Milan / Italy
479
Author Block: S. Triggiani, G. Pellegrino, M. T. Contaldo, A. M. Ierardi, S. A. Angileri, P. Biondetti, P. Torcia, G. Carrafiello; Milan/IT
Purpose: The aim of the study was to describe our experience of percutaneous treatment of eight patients with lumbar facet
synovial cysts (LFCS).
Methods or Background: Eight patients referred to our centre for neurological symptoms such as low back pain and radiculopathy,
refracting to the oral unpainful therapy. At MRI scan all of them were diagnosed for LFCS: five patients on the right side, three on the
left side. All but two cysts (L5-S1) were at L4-L5 level.
The patients underwent CT-guided treatment in an aseptic environment. All of them were positioned prone and local anaesthesia was
administered. Under CT- guidance, a 21-gauge needle was inserted into the facet joint with a posteroanterior approach homolateral to
the cyst and 1-2 mL of non-ionic contrast agent was locally injected to localise the cyst. The rupture was performed through the
injection of dilute contrast in the cyst with a high-pressure syringe. The treatment was considered successfully completed when an
extravasation of contrast was observed outside the cyst’s wall and into the epidural space.
VAS-score was used to evaluate pain before and after the treatment.
Results or Findings: Before the treatment all patients referred a VAS-score of more than six, up to ten. 100% of procedures
achieved the successful rate, radiologically intended as contrast extravasation in the site of the cyst.
On telephone follow-up at six to ten months after the treatment, all patients referred a VAS-score lower than four. No complications
occurred.
Conclusion: The benefits of CT-guided cyst rupture technique can justify the future prevalence of surgical choice, as all patients
achieved an immediate improvement in pain and no complications were reported during the procedure and on telephone follow-up.
Limitations: Identified limitations were (1) that a small series was used and (2) that this was a single centre retrospective study.
Ethics committee - additional information: Not applicable as the study was retrospective.
Genicular artery embolisation for the treatment of symptomatic knee osteoarthritis (7 min)
Florian Nima Nima Fleckenstein; Berlin / Germany
Author Block: F. N. N. Fleckenstein, B. Gebauer, B. Hamm, F. Collettini; Berlin/DE

Purpose: Genicular artery embolisation (GAE) is an innovative minimally-invasive therapy for patients with symptomatic knee
osteoarthritis (OA) refractory to conservative treatments, aiming to reduce synovial arterial hypervascularity. This study aimed to
evaluate the safety and efficacy of GAE for the treatment of symptomatic knee OA.
Methods or Background: A retrospective, single-centre study was conducted at our institution. Patients enrolled in the study were
aged 40 to 90 years, had moderate to severe knee OA (Kellgren-Lawrence grade 2 to 4), and had previously experienced failure of
conservative therapy. Baseline pain (assessed using the visual analog scale [VAS]) and symptom scores (Knee Injury and
Osteoarthritis Outcome Score [KOOS]) were evaluated. After achieving femoral arterial access via a 4 Fr sheath, embolisation was
performed using Imipenem/Cilastatin. Target vessels were determined using digital subtraction angiography in correlation with the
patients' pain points. Adverse events and symptom scores were assessed at six weeks, three months, and six months after GAE.
Results or Findings: A total of 46 patients were enrolled, with a median age of 69 years (IQR, 61, 74). Knee OA severity was grade 2
in 12% of patients, grade 3 in 41%, and grade 4 in 47%. Technical success was achieved in 100% of procedures. Transient skin
discoloration and transient mild knee pain after the procedure were noted in 18% of all cases, as expected. No complications were
reported. The KOOS daily activity index and VAS improved by 87% and 71%, respectively, at six months from a median baseline of 57
(of 100) and 7 (of 10), respectively.
Conclusion: This retrospective study demonstrates that GAE is an effective and safe treatment for reducing OA-associated
symptoms that are refractory to conservative therapy.
Limitations: This was a retrospective single-centre study, without a placebo or control arm.
Patellar tendinopathy: a promising treatment approach with stromal vascular fraction (SVF) (7 min)
Nicolas Papalexis; Bologna / Italy
480
Author Block: G. Martinese, N. Papalexis, G. Facchini, M. Di Carlo, M. Casavola, G. Vara, G. Filardo, S. Zaffagnini, M. Miceli;
Bologna/IT
Purpose: Patellar tendinopathy is a common overload-related pathology in both athletes and the general population, which can
hinder sports activities and limit daily life functions. The use of stromal vascular fraction (SVF) is an innovative approach in the
treatment of cases resistant to conservative therapy. The aim of this study is to assess the effectiveness and safety of the
experimental SVF treatment at 12 months post-infiltrative procedure, along with clinical and radiological correlations. A secondary
objective is to evaluate interoperator agreement on certain radiological parameters to establish reproducibility.
Methods or Background: Thirteen patients (11 males and 2 females) were included, treated with ultrasound-guided SVF infiltration
obtained from subcutaneous abdominal adipose tissue. Each patient was clinically assessed at the baseline and 12-month follow-up
using the following questionnaires: VISA-P, VAS scale, EQ-5D, EQ-VAS, and Tegner score. Simultaneously, B-mode ultrasound
evaluation, colour and power Doppler, microvascular imaging (MVI), and compressive elastography were performed by two operators
with varying levels of experience.
Results or Findings: No adverse events were reported following the infiltrative procedure. A significant clinical improvement was
observed in all questionnaires (increase in VISA-P from 57.36±17.67 to 80.2±19.80; p=0.001), with a significant correlation with
reduced vascular signal (VAS-colour Doppler C.I.=-0.747; p=0.024), thickness reduction (EQ-VAS-thickness C.I.=0.68; p=0.04), and
increased elastographic modulus (Tegner score-E1 C.I.=0.65; p=0.04). Interoperator agreement proved to be excellent for various
ultrasound parameters (colour Doppler ICI=0.91; 95% CI=0.08-0.94).
Conclusion: Ultrasound-guided SVF infiltration is a safe and promising procedure for treating patellar tendinopathy, with significant
clinical and radiological improvements at a 1-year follow-up.
Limitations: Identified limitations were (1) that this was a non randomised study and (2) that there was a lack of control group.
Ethics committee - additional information: This study was approved by the Institutional Review Board and by the Ethics
Committee.
The performance of non-contrast magnetic resonance angiography using cardiac-gated 3D fast spin-echo for 3D fusion
guidance in transarterial embolisation for refractory shoulder pain (7 min)
Yi-Chih Hsu; Taipei / Taiwan, Chinese Taipei
Author Block: Y-C. Hsu, Y-C. Chen, C. Tung, L-H. Chiang; Taipei/TW
Purpose: Transarterial embolisation (TAE) is a widely used the treatment for aggravating musculoskeletal pain in the extremities,
that is refractory to conservative treatment. Three dimensional (3D) fusion guidance for fluoroscopic navigation has been used to
overcome difficulties in identifying overlapping vessels during TAE. As a pre-fusion imaging for fluoroscopic navigation, the main
advantage of using magnetic resonance angiography (MRA) is that it can be performed without the use of contrast media. However, a
feasible sequence for non-contrast MRA (NC-MRA) suitable for the upper extremity has not yet been established. The aim of this study
was to evaluate the performance of NC-MRA using cardiac-gated 3D fast spin-echo (FSE) in the detail of arterial morphology of the
shoulder.
Methods or Background: Seven consecutive patients who had refractory shoulder pain underwent NC-MRA and TAE. Two reviewers
assessed image quality for visualisation of shoulder arteries by using a qualitative 4-point scale (1=not assessable to 4=excellent).
The number and morphology of arterial branches of subclavian and axillary arteries that were adequate for TAE treatment of
refractory shoulder pain were also evaluated. The results were compared with those from digital subtraction angiography during TAE.
Interobserver agreement was evaluated with the κ statistic.
Results or Findings: NC-MRA using cardiac-gated 3D FSE is technically successful for arterial depiction of the shoulder. Image
quality was considered excellent (median=4) and the κ coefficient was 0.85. Additionally, NC-MRA correctly represented arterial
branches of subclavian and axillary arteries that were adequate for TAE treatment of refractory shoulder pain in all patients.
Conclusion: NC-MRA using cardiac-gated 3D FSE is technically and clinically feasible and represents a promising technique for 3D
fusion guidance in TAE for refractory shoulder pain.
Limitations: The small number of patients was an identified limitation.
Ethics committee - additional information: This study was approved by the research institution of Tri-Service General Hospital.
Intra-articular hyaluronic acid and platelet-rich plasma as monotherapy or combination therapy, what is the evidence?
(7 min)
Sheng-Fei Oon; Melbourne / Australia
481
Author Block: S-F. Oon, C. Nguyen; Melbourne/AU
Purpose: This study aimed to use a validated evidence-based practice tool to identify best current evidence for combined intra-
articular hyaluronic acid (HA) and platelet-rich plasma (PRP) versus monotherapy in treating knee osteoarthritis.
Methods or Background: Knee osteoarthritis is a degenerative joint disease characterised by progressive loss of articular cartilage
from wear and tear. Combined HA and PRP intra-articular injections gained recent interest and may relieve pain and reduce
progression of osteoarthritis through synergistic mechanisms. Current literature causes confusion due to mixed conclusions, even
among meta-analyses. We used the McMaster University and National Health Service five-step systematic approach to conduct a
bottom-up search through the primary and secondary literature. We limited our search to review and meta-analysis articles in the
English language only, within the last three years.
Results or Findings: A total of 69 articles were retrieved and evaluated for validity and strength. Five final articles were selected for
individual review and the results tabulated. For combined HA+PRP versus HA alone, HA+PRP demonstrated significant improvement
in stiffness and physical function symptoms at 3, 6 and 12 months (p <0.001). There was no statistically significant difference
between PRP+HA and HA alone in pain control (p=0.195). For HA+PRP versus PRP alone, pain reduction was seen at one month with
PRP only. After this time, no statistically significant difference was found between the two groups with regards pain, stiffness or
physical function.
Conclusion: Combination PRP with HA demonstrates superior symptom control when compared to HA monotherapy, but combination
PRP and HA does not provide further long term statistically significant symptom control over PRP monotherapy.
Limitations: The limitation of articles to the English language and within the last 3 years could have reduced the robustness of
search. Meta-analyses were of low quality due to extensive heterogeneity of the included studies.
Ethics committee - additional information: Ethical approval waived by Peter MacCallum Cancer Centre Ethics and Governance
Committee: "Literature reviews of publically available information do not require ethics committee review."
482
BS 12 - Caring for and maximising the profile of the radiography profession
Categories: Education, Management/Leadership, Multidisciplinary, Professional Issues, Radiographers

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Putting care under pressure: the need to care for our profession (18 min)
Jonathan McNulty; Dublin / Ireland
1. To explore the overlapping challenges faced by the workforce linked to care and compassion and self-care and self-compassion.
2. To stimulate discussion about the significance of these challenges in medical imaging.
3. To identify simple approaches for managers and individuals to address such workforce challenges.
Maximising the potential of the modern radiographer workforce (18 min)

Julie Michelle Nightingale; Sheffield / United Kingdom
1. To reflect upon a wide range of career development opportunities within the radiography profession.
2. To compare and contrast the potential scope of enhanced, advanced and consultant practice.
3. To identify opportunities for embedding elements of research, leadership, education and service improvement within the
radiographer role.
Building relationships and becoming a valued team member (18 min)

1. To identify and apply best practices for effective communication.

2. To describe team-building strategies that strengthen relationships and build networking.
3. To demonstrate skills for sharing vision and goals as well as enhance emotional intelligence.
483
RC 1204 - Chameleons in thoracic radiology
Categories: Chest, Oncologic Imaging

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Nicholas J. Screaton; Cambridge / United Kingdom

Nicholas J. Screaton; Cambridge / United Kingdom
Atypical manifestations of lung cancer (15 min)

Danielle Byrne; Dublin / Ireland
1. To know when to suspect atypical imaging presentations of lung cancer.

2. To understand which further imaging to suggest for clarification.
3. To confidently differentiate between lung cancer and other causes.
Lung metastases: not always as it seems (15 min)

Cornelia M. Schaefer-Prokop; Amersfoort / Netherlands
1. To recognise when lung nodules may be metastatic.

2. To understand which further imaging to suggest for clarification.
3. To confidently differentiate between lung metastases and other causes.
Benign diseases mimicking a thoracic malignancy (15 min)

Lucio Calandriello; Roma / Italy
1. To understand which infections mimic malignancies.

2. To understand which granulomatous diseases mimic malignancies.
3. To understand when to biopsy and when to suggest "wait and see".
Panel discussion: How to approach uncertainties in thoracic oncology (10 min)
484
RPS 1205 - Recent development in AI for lung nodule detection
Categories: Artificial Intelligence & Machine Learning, Chest, Imaging Methods

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Giuseppe Cicchetti; Roma / Italy
External validation of the Sybil risk model as a tool to identify low-risk individuals eligible for biennial lung cancer
screening (7 min)
Fennie Van der Graaf; Nijmegen / Netherlands
Author Block: F. Van der Graaf1, N. Antonissen1, Z. Saghir2, C. Jacobs1, M. Prokop1; 1Nijmegen/NL, 2Hellerup/DK
Purpose: Lung cancer screening protocols for follow up intervals should minimise harm, maximise cost-effectiveness, and avoid
diagnostic delays. ILST suggests biennial follow-up for low-risk participants. The study aimed to retrospectively evaluate Sybil, a deep
learning algorithm predicting lung cancer risk for 6 years from one LDCT, comparing it to PanCan2b for identifying biennial screening
eligibility.
Methods or Background: DLCST baseline scans included 1870 non-cancer and 25 screen-detected cancer cases, diagnosed within
2 years. Sybil (scan level) and PanCan2b (per nodule) predicted risk of developing cancer within 2 years. For cases with no screen-
annotated nodules, the PanCan2b risk score for participants was set as 0%. For both models, we used a nodule-risk cut-off of <1.5%
to identify low-risk participants for biennial follow-up, based on ILST. For PanCan2b, the risk dominant nodule per scan was
considered.
Results or Findings: The Sybil and PanCan2B models identified 1616 and 1697 individuals, respectively, meeting the criteria for
biennial screening. This would result in a reduction of 87% and 94% of CT scans in the second screening round, respectively. The
group referred for biennial screening included 8 and 9 cancers for Sybil and PanCan2B, respectively.
Conclusion: Both Sybil and PanCan2B selected a large group of low-risk participants for biennial screening when a <1.5% risk
threshold was used at baseline CT. The difference between Sybil and the PanCan2b model is small. More research is needed to study
the type of cancers with delayed diagnosis and whether such delay leads to diagnostic stage shift. In addition, more external
validation of the Sybil model on other datasets is necessary to further assess its applicability in lung cancer screening, and to
evaluate its performance on follow-up imaging.
Limitations: This study is a baseline, retrospective analysis on data from one screening trial.
Funding for this study: Funding for this study is supplied by a research grant that is funded by the Dutch Science Foundation and
Mevis Medical Solutions, Bremen, Germany
Ethics committee - additional information: The study included data collected from the Danish Lung Cancer Screening Trial
(DLCST). For DLCST, the Ethics Committee of Copenhagen county approved the study, and informed consent was obtained from all
participants.
Artificial intelligence: the key to significant reduction in baseline LDCT lung cancer screening associated workload
when used as a first-read filter (7 min)
Harriet Louise Lancaster; Groningen / Netherlands
485
Author Block: H. L. Lancaster1, B. Jiang1, M. Silva2, J. W. Gratema3, D. Han1, J. Field4, G. De Bock1, M. A. Heuvelmans1, M. Oudkerk1;
1
Groningen/NL, 2Parma/IT, 3Apeldoorn/NL, 4Liverpool/UK
Purpose: Artificial intelligence (AI) is not a new concept in the field of lung cancer screening. To date, AI has predominantly been
used to predict lung nodule malignancy risk (rule-in principle). However, to have an impact on reducing radiologist workload, a new
rule-out approach is needed. This study aimed to evaluate if AI can be used as a first-read filter to rule-out negative cases (nodules
<100 mm3), so that radiologists would only need to assess indeterminate-positive nodules, therefore significantly reducing workload.
Methods or Background: External validation of AI (AVIEW_LCS, v1.1.39.14) was performed in a UKLS dataset containing 1254
LDCT-baseline scans. Scans were assessed independently by four manual readers and AI. Discrepancies between reads (manual/AI)
were reviewed by a consensus panel of two experienced thoracic radiologists, blinded from the original results. Final classification was
based on the consensus reference read. Cases were ultimately classified as
correct, positive-misclassifications (PMs) (nodules classified by the reader/AI as ≥100 mm3, but at consensus <100 mm3) and
negative-misclassifications (NMs) (nodules classified by the reader/AI as <100 mm3, but at consensus ≥100mm3).
Results or Findings: Based on consensus reference read, 816 (65%) cases were negative and 438 (35%) indeterminate-positive. AI
had fewer NMs 68 (5%) than all manual readers [reader 1; NMs 205 (16%), reader 2; NMs 200 (16%), reader 3; NMs 236 (19%), reader
4; NMs 220 (18%)], which was reflected in an AI negative predictive value (NPV) of 91.7% (89.8-91.4%) [reader1; 79.0% (77.5-82.0%),
reader 2; 80.1% (81.3-85.4%), reader 3; 77.5% (76.0-79.0%) and reader 4; 78.5% (77.0-80.0%). Workload reduction using AI was
calculated at 65% [(total scans;1254-(correct positives; 370 + positive misclassifications; 59))/total scans;1254].
Conclusion: AI negative predictive performance is better than all manual readers. If used as a first-read filter, radiologists would only
need to assess 35% of cases with indeterminate-positive nodules, meaning significant workload reduction.
Limitations: An identified limitation of this study was that true positives and negatives based on histological outcome were not
reported, analyses will begin shortly.
Ethics committee - additional information: The UKLS study was approved by an ethics committee; this substudy was covered by
previous approval as it only included de-identified data.
Can an AI-driven decision aid reduce the time between chest x-rays and treatment for lung cancer patients? (7 min)
Andrew Keen; Aberdeen / United Kingdom
Author Block: A. Keen, S. Wilkie, B. E. Morrissey, S. Prior, L. Cameron; Aberdeen/UK

Purpose: The primary aim of this study was to evaluate whether an AI product with the ability to identify chest x-ray (CXR) images of
highest risk of lung cancer can reduce the time between imaging and treatment in those patients subsequently diagnosed with lung
cancer.
Methods or Background: The NHS Grampian Innovation, Radiology and Cancer Teams collaborated with the Centre for Sustainable
Delivery and the Scottish Health Technology Group to design an evaluation of the real-world impact of using an AI product designed
to risk stratify CXR images. Full pathway mapping was carried out and baseline time delays between all key points (CXR, CXR
reporting, CT, CT reporting, MDT diagnosis and treatment) were established. CXR images flagged as highest risk of lung cancer were
expedited for CXR reporting, CT and CT reporting. NHS Grampian radiologists collaborated with the company to calibrate the product
in ways that maximised identification of lung cancer whilst not overwhelming CT capacity.
Results or Findings: Several months into the project the time between CXR and CT report has dropped from 22 to 10.3 days
(N=132). Radiologists identified 28 images not flagged by the product about which they were concerned about cancer. Thus far, none
of these patients have been diagnosed with cancer. Under the current calibration conditions, using radiologists’ judgements, the
product performs at 84.4 sensitivity and 90.5 specificity (N=24071).
Conclusion: Early results suggest AI risk stratification of CXR images may help healthcare organisations reduce the time taken to
treat people diagnosed with lung cancer. This could be especially important for people who are diagnosed following CXR imaging for
non-cancer reasons. In our region, this is about two thirds of people diagnosed with lung cancer.
Limitations: A limitation of this study is that these are early results from a 12 month evaluation.
Funding for this study: Funding was received from the Scottish Government.
Ethics committee - additional information: We were advised by our Research and Development Office that permissions were not
necessary on this occasion. Local and national clinical governance and executive-level sponsorship is in place.
Comparison of the impact of reconstruction kernels on pulmonary nodule volumetry in low-dose CT with iterative vs
deep learning image reconstruction (7 min)
Louise D'hondt; Ghent / Belgium
486
Author Block: L. D'hondt1, C. Franck2, P-J. Kellens1, F. Zanca3, D. Buytaert4, K. Bacher1, A. Snoeckx2; 1Ghent/BE, 2Antwerp/BE,
3
Leuven/BE, 4Aalst/BE
Purpose: The objective of the study was to investigate the impact of different reconstruction kernels and its interaction with other
imaging parameters on nodule volumetry, since scan protocols, screening guidelines and vendor specifications typically define the
soft kernel as the standard, thereby disregarding its potential influence.
Methods or Background: We scanned the Lungman phantom containing 3D-printed lung nodules, encompassing six diameters (4
to 9 mm) and three morphology classes (lobular, spiculated, smooth), using a 256-slice CT scanner at various radiation doses (CTDIvol
6.04, 3.03, 1.54, 0.77, 0.41, 0.20 mGy) and reconstructed using different combinations of either soft or hard reconstruction kernels
and iterative reconstruction (IR) or deep learning image reconstruction (DLIR) at varying strengths. The impact of these imaging
parameters on semi-automatic volumetry measurements was analysed through multiple linear regression.
Results or Findings: We found that reconstruction kernel significantly impacts volumetric accuracy, both as primary factor and in
interaction with the reconstruction algorithm and radiation dose. Overall, volumetric errors are lower with the soft kernel compared to
the hard kernel. Additionally, we observed that the soft kernel exhibited reduced errors with increasing radiation dose, while this
remained relatively constant across all doses for the lung kernel. Combination of the lung kernel with DLIR resulted in relative
reduction in volumetric error up to 50% as opposed to IR, at all doses. Furthermore, this effect became more pronounced as the DLIR
strength increased. Across all nodule morphologies and diameters using the lung kernel, DLIR consistently outperformed IR, with
relative reductions between 20 and 90% in error.
Conclusion: Compared to other combinations of reconstruction algorithms and kernels, application of DLIR in combination with a
hard kernel overall returns the highest volumetric accuracy for all pulmonary nodules, also at (ultra)low radiation doses.
Limitations: An identified limitation is that this is a phantom study.
Funding for this study: Funding was provided by the FWO “Kom op tegen Kanker” project for lung cancer screening research in
Belgium (Project number: G0B1922N).
Ethics committee - additional information: No ethics committee approval was needed, since this study used phantom images.
Improving image quality of sparse-view lung cancer CT images using convolutional neural networks (7 min)
Tina Dorosti; Neuried / Germany
1 2 1 1 1 2 1 2 1
Author Block: T. Dorosti , A. Ries , J. B. Thalhammer , A. Sauter , F. Meurer , T. Lasser , F. Pfeiffer , F. Schaff , D. Pfeiffer ;
1 2
Munich/DE, Garching/DE
Purpose: This study aimed to improve the image quality of sparse-view computed tomography (CT) images with a U-Net for lung
cancer detection and to determine the best trade-off between number of views, image quality, and diagnostic confidence.
Methods or Background: CT images from 41 subjects (34 with lung cancer, seven healthy) were retrospectively selected
(01.2016-12.2018) and forward projected onto 2048-view sinograms. Six corresponding sparse-view CT data subsets at varying levels
of undersampling were reconstructed from sinograms using filtered back projection with 16, 32, 64, 128, 256, and 512 views,
respectively. A dual-frame U-Net was trained and evaluated for each subsampling level on 8,658 images from 22 diseased subjects. A
representative image per scan was selected from 19 subjects (12 diseased, seven healthy) for a single-blinded reader study. The
selected slices, for all levels of subsampling, with and without post-processing by the U-Net model, were presented to three readers.
Image quality and diagnostic confidence were ranked using pre-defined scales. Subjective nodule segmentation was evaluated
utilising sensitivity (Se) and Dice Similarity Coefficient (DSC) with 95% confidence intervals (CI).
Results or Findings: The 64-projection sparse-view images resulted in Se=0.89 and DSC=0.81 [0.75, 0.86], while their
counterparts, post-processed with the U-Net, had improved metrics (Se=0.94, DSC=0.85 [0.82, 0.87]). Fewer views lead to
insufficient quality for diagnostic purposes. For increased views, no substantial discrepancies were noted between the sparse-view
and post-processed images.
Conclusion: Projection views can be reduced from 2048 to 64 while maintaining image quality and the confidence of the radiologists
on a satisfactory level.
Limitations: The sparse-view data generated for this study was obtained using simplified conditions not reflective of the complex
reconstruction processes in clinical settings. Therefore, an exact measure of dose reduction is hence unachievable.
Funding for this study: Funding was received from the Federal Ministry of Education and Research (BMBF) and the Free State of
Bavaria under the Excellence Strategy of the Federal Government and the Länder, the German Research Foundation (GRK2274), as
well as by the Technical University of Munich - Institute for Advanced Study.
Ethics committee - additional information: The study was approved by the ethical review committee and was conducted in
accordance with the regulations of our institution. All data was analysed retrospectively and anonymously.
Head-to-head validation of AI software for the detection of lung nodules in chest radiographs: Project AIR (7 min)
Steven S Schalekamp; Nijmegen / Netherlands
487
Author Block: K. G. van Leeuwen1, S. S. Schalekamp1, M. J. Rutten1, M. Huisman1, C. M. Schaefer-Prokop2, M. De Rooij1, B. Van
Ginneken1; 1Nijmegen/NL, 2Amersfoort/NL
Purpose: Multiple commercial artificial intelligence (AI) products exist for the detection of lung nodules on chest radiographs,
however, comparative performance data of the algorithms is limited. The purpose of the study was to perform independent stand-
alone comparison of commercially available AI products for lung nodule detection on chest radiographs, benchmarked against human
readers.
Methods or Background: This retrospective, multicentre (n=7 Dutch hospitals) study was carried out as part of Project AIR, which is
a Dutch initiative for independent, repeatable, multicentre validation of AI products in radiology. Seven out of 14 eligible AI products
for the detection of lung nodules on chest radiographs were validated on a dataset of 386 chest radiographs. The reference was chest
CT within 3 months of the chest radiograph. Performance was measured using area under the receiver operating characteristic curve
(AUROC). Random subsets of chest radiographs (n=140) were read by 17 human readers, with varying levels of experience.
Results or Findings: Seven lung nodule detection products were validated on chest radiographs (January 2012 to May 2022) of 386
patients (mean age, 64 years ± 11 [SD]; 223 males). Compared to human readers (mean AUROC, 0.81 [95% CI: 0.77, 0.85]), four
products performed better (AUROC range, 0.86-0.93 [95% CI: 0.82, 0.96]; P range, <.001-.04). No significant difference was found
between the remaining three products and human readers (AUROC 0.79 [0.74,0.84] P=.33, 0.80 [0.75, 0.85] P=.60, 0.84 [0.80, 0.88]
P=.26).
Conclusion: Compared to human readers, four AI products for detecting lung nodules on chest radiographs showed superior
performance whereas three other products showed no evidence of difference in performance for the detection of lung nodules.
Limitations: The added value of these AI products in clinical practice has yet to be determined.
Ethics committee - additional information: The analysis was on anonymised retrospective data.
Autonomous algorithmic monitoring of a deep-learning chest radiograph AI model using temporal divergences, in a
clinical real-world setting (7 min)
Jin Yee Charlene JY Liew; Singapore / Singapore
Author Block: J. Y. C. J. Liew1, A. Gupta2, A. M. Surve2, V. K. Venugopal2; 1Singapore/SG, 2New Delhi/IN

Purpose: This study aimed to evaluate autonomous monitoring of AI models in a clinical environment by measuring temporal
divergence of mathematical probability distributions.
Methods or Background: Daily prediction scores and overall abnormality prediction score of a chest radiograph classification
solution (Lunit Insight CXR) were used. There were a total of 11,572 chest radiograph studies analysed over 57 days continuously on
an AI platform solution in a real-world clinical setting. Of these, 7,005 studies were classified as abnormal by the AI model. The
probability distributions for the abnormal predictions and probability scores for 10 abnormal diagnoses were plotted on a daily basis.
Jensen-Shannon divergence was used to measure the similarity between the probability distributions of current day and the previous
day in a continuous moving fashion. Daily divergence between the probability distributions against the distribution of a fixed
reference was also measured. A threshold of 0.2 for acceptable divergence was used. The studies in days where the threshold was
breached were reviewed for any potential errors or misclassification.
Results or Findings: On Day 55, there was a system technical downtime, resulting in fewer cases being processed. This day's
divergences were particularly prominent, with Pneumothorax recording 0.723. Excluding day 55, divergence ranged from 0.009-0.329
across findings. Divergence values for such findings were recalibrated against the moving averages of the previous three days.
Conclusion: We introduced an innovative algorithmic system to monitor deep learning AI solution performance using divergence
scores. Divergences were detected on days where there were technical downtimes in the AI system. This emphasises the importance
of continuous monitoring of AI in clinical applications, to detect various failures of AI models, which may be due to catastrophic
algorithmic failure, data bias, model drift or population data drift.
Ethics committee - additional information: This study was approved by SingHealth Central Institutional Review Board:
2023/2280.
A deep learning module for automated detection and reporting of clinically significant lung nodules on low-dose chest
CT scans (7 min)
Veljko Popov; Westford / United States
488
Author Block: V. Popov1, J. Afnan1, U. Kalabic2, Z. Li3, D. Chen3, D. Hassan2, D. Radisic2; 1Burlington, MA/US, 2Wenham, MA/US, 3East
Lansing, MI/US
Purpose: Lung cancer remains the leading cause of cancer death worldwide. Multicentre trials (NLST, NELSON) have proven the
efficacy of lung cancer screening in high-risk patients using low-dose, non-contrast chest CT scans. A novel Artificial intelligence (AI)
module for automated nodule detection and output to the structured report is proposed, to assist with increasing screening rates
while maintaining high levels of diagnostic accuracy.
Methods or Background: The nnDetection framework was applied to train a one-stage detector to segment nodules. Predictions
from the nodule detector were fed through an efficient mechanism for reducing overlapping bounding boxes and a separate 3D deep
convolutional neural network was trained for false positive reduction (FPR).
The model was then trained on the LUNA16 database (800+ LDCT studies). The model was tested on a holdout subset of LUNA16 (89
studies) and the Cornell ELCAP database (40 studies), for nodules 6 mm or greater.
Results or Findings: LUNA16 dataset: The performance of the nnDetection framework results in a recall of 100%, a precision of
77%, and a false negative rate of 0%. Adding the FPR model, the recall remains at 100%, the precision increases to 84%, and the
false negative rate is 0%.
ELCAP dataset: For nodules 6 mm or larger, nnDetection with FPR results in a recall of 100%, a precision of 58%, and a false negative
rate of 4%.
Conclusion: nnDetection + FPR performs very well in detecting clinically relevant nodules on the LUNA16 dataset. In addition, the
model shows the ability to scale across LDCT datasets without fine tuning when applied to the ELCAP Cornell dataset, detecting all
nodules 6 mm or greater.
Limitations: An identified limitation was the small datasets.
Funding for this study: Private funding was obtained for this study.
Ethics committee - additional information: Ethics committee approval was not needed as the study used publicly available
datasets.
489
E³ 22C - The traumatic ankle

ETC Level: LEVEL I
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Tendon and ligament injuries (25 min)

1. To learn about basic ligament anatomy.

2. To classify ligament and tendon injuries.
3. To recognise typical hints for tendon injuries in the ankle.
Osteochondral lesions: pre- and postoperative findings (25 min)

Iris Melanie Noebauer-Huhmann; Vienna / Austria
1. To learn about typical imaging findings of osteochondral lesions.

2. To learn about different surgical techniques for osteochondral repair and their normal imaging findings.
3. To recognise postoperative complications, and how to evaluate the outcome in the follow-up by a semiquantitative MR scoring
system.
490
PA 12 - Next-generation radiology: life well spent with good communication is long for radiologists
and patients
Categories: Artificial Intelligence & Machine Learning, Breast, Education, Oncologic Imaging, Professional Issues
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

The view of the President of the European Society of Radiology on next-generation communication (10 min)
1. To learn about the demands of next-generation radiology.

2. To appreciate the impact of support of societies.
3. To understand the important key points of communication in the changing world of radiology.
360-degree approach radiology: communication best practice example (10 min)

1. To learn how Be accepted is used in practice and a 360-degree approach to radiology is reached.
2. To appreciate the 360-degree approach's benefits on a multidisciplinary level.
3. To understand the importance of implementing tools for communication.
Responsibility and opportunity for next generation of radiologists and patients: tools in use, Be accepted (10 min)
1. To learn how Be accepted covers all needs.

2. To appreciate the implementation of AI tools via Be accepted.
3. To understand the responsibility and opportunities of all stakeholders.
Next-generation radiology: future communication in progress (10 min)

1. To learn how the shifting role of radiologists demands different communication.

2. To appreciate AI implementations to focus on the benefits.
3. To understand the young generation of radiologists' view on communication.
Preparing and educating the patient for shared decision-making (10 min)
1. To learn how important evidence-based information is and how to protect patients from false information.
2. To appreciate being in a core role for patients along their pathway.
3. To understand the importance of educating patients about their treatment process to make shared decision-making possible.
4. To address how to handle the patients accessing their reports before their clinician has given them their diagnosis.
491
492
E³ 24C - Infection and inflammation
Categories: Hybrid Imaging, Molecular Imaging, Nuclear Medicine

ETC Level: LEVEL I
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Hybrid imaging of infection and inflammation in paediatric patients (15 min)

Ora Israel; Haifa / Israel
1. To define the clinical indications for hybrid imaging in children with known or suspected infectious and inflammatory processes.
2. To present the performance indices of hybrid imaging in children with known or suspected infection and inflammation.
3. To define the clinical and technical challenges as well as recommended imaging protocols for hybrid imaging in the paediatric
patient population.
Inflammation of orthopaedic prosthesis (15 min)

1. To understand the role of hybrid imaging in infection and inflammation.

2. To learn the limitations and pitfalls of hybrid imaging, mainly SPECT/CT and PET/CT, to diagnose infections and inflammation.
3. To learn about the different radiopharmaceuticals that may be used in the inflammation process.
FDG PET/CT in patients with signs and symptoms of GCA and PMR (15 min)
Riemer Slart; Groningen / Netherlands
1. To understand the patient’s journey in GCA and PMR.

2. To get informed about the indications, value, procedure, and reading of FDG PET/CT in the diagnosis and therapy-monitoring of
GCA and PMR.
3. To get informed about novel developments of PET/CT in GCA and PMR.
Panel discussion: Is hybrid imaging troubleshooter or standard of care in infection and inflammation? (10 min)
493
RC 1215 - Diagnosis and management of portal venous thrombosis
Categories: GI Tract, Interventional Radiology, Multidisciplinary, Vascular

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Pedro Filipe Goncalves Teixeira Sousa; Porto / Portugal

Pedro Filipe Goncalves Teixeira Sousa; Porto / Portugal
Clinics of portal venous thrombosis (15 min)

Pál Ákos Deák; Budapest / Hungary
1. To understand the terminology and classification of portal venous thrombosis.

2. To understand the prevalence and risk factors of portal venous thrombosis in patients with and without cirrhosis.
3. To understand the portal venous thrombosis treatment options.
Diagnostic imaging of portal venous thrombosis (15 min)

Pedro Melo-Freitas; Aveiro / Portugal
1. To understand the indications for non-invasive portal venous thrombosis diagnostic imaging.
2. To understand which imaging technique should be applied to different clinical scenarios.
3. To understand the essential protocol for angio-CT in the context of portal venous thrombosis.
4. To recognise the imaging signs of portal venous thrombosis and how to report them.
Management of portal venous thrombosis: the role of interventional radiology (15 min)
Charlotte Ebeling Barbier; Uppsala / Sweden
1. To name the relevant materials and infrastructure needed for the IR management portal venous thrombosis.
2. To summarise and compare IR treatments' results.
3. To analyse the role of IR in the management of portal venous thrombosis in comparison with non-invasive treatment.
Panel discussion: The radiologists in the multidisciplinary management of portomesenteric venous thrombosis: are we
sufficiently involved in the decision process? (10 min)
494
RC 1207 - Female pelvic oncology: an update

ETC Level: LEVEL II
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Milagros Otero Garcia; Vigo / Spain

Endometrial cancer (15 min)

Stephanie Nougaret; St Clement de Riviere / France
1. To learn about the clinical indications and imaging modalities for endometrial cancer staging.
2. To learn how to appropriately acquire and report MRI images according to new guidelines.
Cervical cancer (15 min)

1. To learn about the clinical indications and imaging modalities for cervical cancer staging.
Vulvar cancer (15 min)

Olivera Nikolic; Novi Sad / Serbia
1. To learn about the clinical indications and imaging modalities for vulvar cancer staging.
Panel discussion: How does imaging influence the management of patients? (10 min)
495
RC 1212 - Musculoskeletal and soft tissue tumours in children
Categories: Musculoskeletal, Oncologic Imaging, Paediatric

ETC Level: LEVEL I
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Sandra Diaz; Solna / Sweden

Sandra Diaz; Solna / Sweden
Typical paediatric bone tumours (15 min)

Pier Luigi Di Paolo; Rome / Italy
1. To understand the typical imaging features of paediatric bone tumours.

2. To learn how best to use imaging modalities in differential diagnosis.
3. To learn how to combine imaging and clinical features to suggest a specific diagnosis.
Typical paediatric soft tissue tumours (15 min)

Emilio Jose Inarejos Clemente; Esplugues de Llobregat, Barcelona / Spain
1. To review the use of imaging techniques for the evaluation of soft tissue tumours, including radiography, ultrasound, conventional
MRI, and multiparametric MRI.
2. To describe the radiological management of soft tissue tumours.
3. To illustrate the most frequent soft tissue tumours with their corresponding key imaging features.
Tumour mimics (15 min)

Janina Maria Patsch; Vienna / Austria
1. To learn about bone entities in paediatric MSK imaging and soft tissue tumour mimics as important.
2. To understand the underlying pathophysiology leading to tumour-like appearance (of bone and soft tissue tumour mimics).
3. To appreciate the importance of bone and soft tissue tumour mimics as essential differential diagnoses for paediatric
musculoskeletal masses.
Panel discussion: If imaging cannot always make a definite diagnosis, when should you opt for a biopsy? (10 min)
496
RPS 1202 - Artificial intelligence (AI) in breast imaging
Categories: Artificial Intelligence & Machine Learning, Breast, Imaging Methods

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Henrik Wethe Koch; Stavanger / Norway
Adding artificial intelligence (AI) case malignancy scoring in a breast screening programme to reduce screen-reading
workload: a retrospective study (7 min)
Andrea Nitrosi; Reggio Emilia / Italy
Author Block: A. Nitrosi, P. Giorgi Rossi, L. Verzellesi, N. Cucurachi, R. Vacondio, C. Campari, M. Bertolini, P. Pattacini, M. Iori; Reggio
Emilia/IT
Purpose: The purpose of this study was to evaluate a strategy of integrating AI mammography case malignancy score (AI-CMS) to
reduce breast screen-reading workload avoiding human second reading for mammograms with low AI-CMS.
Methods or Background: We retrospectively analysed 31,747 consecutively collected screening exams from Reggio Emilia breast
screening program (BSP), including 92 proven tumours and 5 pending diagnoses, to assess decision to recall (RD), recall rate (RR) and
tumour detection of two simulated integrated AI and human reading protocols (ProFound AI 2D system iCAD Inc.). iCAD AI-CMS is a
relative score representing the AI algorithm’s confidence that a case is malignant in a 0% to 100% scale.
To estimate the potential reduction in the numbers of human readings, iCAD acts as a reader C1 recalling women with AI-CMS greater
than a predefined threshold (10%/15%/20%). If the radiologist – reader RH1 - disagrees with iCAD, the case undergoes to another
radiologist RH2 and to a third radiologist arbitration RH3 in case of human disagreement (standard screening protocol).
Results or Findings: Assuming respectively 10%/ 15%/ 20% AI-CMS threshold, RD for C1 was 49.4%/ 37.7%/ 29.7%, for RH1 4.6%/
4.6%/ 4.6%, for RH2 8.7%/ 10.6%/ 12.5% and for RH3 69.9%/ 69.7%/ 69.1%. The final RR was 3.70%/ 3.62%/ 3.58% versus actual RR
of 3.86%.
This corresponds to 48,975/ 45,180/ 42,785 versus 65,097 total human readings (corresponding to human workload reduction of
24.8%/ 30.6%/ 34.3%). There’s no increase in false negative with 10% and 15% thresholds, whereas using the 20% threshold results
in an additional false negative.
Conclusion: Adding AI-CMS support to a standard screening scenario could result in a substantial lower screen reading workload, a
modest decrease in RR without any additional false negatives.
Limitations: The study only examines data from the Reggio Emilia BSP.
Funding for this study: This study was partially supported by the Italian Ministry of Health—Ricerca Corrente
Ethics committee - additional information: Compliance with Ethical Standards Institutional Review Board approval was not
required because it is a Clinical Audit about a technical development. This study was conducted in accordance with the routine quality
assurance procedures established by the Local Health Authority for its screening programs. The Reggio Emilia Cancer Registry, which
routinely collects the screening history of each case of breast cancer, has been approved by the Provincial Ethic Committee.
An artificial intelligence tool to empower junior radiologists in breast cancer screening; AI as a second pair of eyes in
mammography reading (7 min)
Mehran Arab Ahmadi; Tehran / Iran
497
Author Block: N. Ahmadinezhad, N. Sadighi, R. Ghavami Modegh, M. Arab Ahmadi, M. Rahmani, A. Arian, H. Dashti, H. R. Rabiee, M.
Gity; Tehran/IR
Purpose: The aim of this study was to assess whether an artificial intelligence (AI) application can assist non-expert radiologists in
improving their performance in detecting the possibility of cancer using digital mammography in an adjunctive workflow.
Methods or Background: A retrospective study was conducted using 2060 digital mammography (DM) images of 515 women from
2018-2022, including 120 positive and 910 negative breasts, with four junior radiologists participating in the study. Radiologists
independently reviewed and interpreted each case without AI assistance. Immediately after submitting their initial interpretations,
they were provided with AI-generated interpretations to support their analysis and evaluation of the cases. Armed with this additional
information, the radiologists had the opportunity to revise and resubmit their interpretations based on their expertise and insights
from the AI system. Radiologists' performance before and after receiving AI assistance was compared using AUC, sensitivity, and
specificity metrics.
Results or Findings: According to our findings, the integration of AI technology resulted in a notable enhancement in the
performance of junior radiologists. The area under the curve (AUC) improved significantly from 0.812 to 0.837, sensitivity increased
from 75.0% to 92.9%, and balanced accuracy enhanced by 3.63%. Additionally, AI proved to be highly beneficial for radiologists in
identifying previously missed lesions across various types, including mass, calcification, distortion, and asymmetry.
Conclusion: AI can improve the diagnostic capabilities and detection rates of radiologists with less than five years of experience,
enhancing their medical imaging performance.
Limitations: One of the limitation of this study is that we did not assess the effect of AI on final recall rates, which would require a
live large-scale survey with a normal distribution to yield reliable results.
Ethics committee - additional information: The clinical study was conducted under the Research Ethic Certificate No. IR.TUMS.
IKHC.REC.1399.490 issued by Tehran University of Medical Sciences on Feb. 24, 2021.
Is it worth reading low-risk breast cancer screening mammograms as determined by an artificial intelligence (AI)
system? A prospective, population-based study for DM and DBT (AITIC trial) (7 min)
Esperanza Elías Cabot; Cordoba / Spain
Author Block: E. Elías Cabot1, S. Romero Martin1, J. L. Raya Povedano1, A. Rodriguez Ruiz2, M. Álvarez Benito1; 1Cordoba/ES,
2
Nijmegen/NL
Purpose: The purpose of this study was to prospectively evaluate AI for safe workload reduction by excluding low risk cases for
human reading and applying double reading to the rest in screening with digital mammography (DM) and digital breast tomosynthesis
(DBT).
Methods or Background: Participants in a breast cancer screening programme in Córdoba, Spain, (women, aged 50-71) are
included in this prospective study and imaged with either DM or DBT. Two reading strategies are independently applied to each exam:
Double blind and non-consensual reading of all exams (control arm) and an AI-based triaging (intervention arm), where an AI system
(Transpara, ScreenPoint Medical) evaluates the cancer risk of all exams. Cases identified by AI as low risk (operating point pre-defined
to yield approximately 70% of exams in this category) are automatically assessed as negative, while cases with intermediate and
elevated risk are double read with AI-support. Readers are randomly assigned to each reading and blinded to other reading outcomes.
We hypothesise that an AI-based screening workflow allows for substantial workload reduction and non-inferior cancer detection rate
(CDR) and recall positive predictive value (PPV).
Results or Findings: Between March 2022 and June 2023, 19243 women participated. AI-based triaging, reading only 6583 exams
(the 34% of the total scored by AI as intermediate and elevated risk), achieved superior CDR compared to double-reading of all cases
(CDR 6.7/1000, 130 cancers vs. 5.9/1000 (114 cancers), p= 0.017), non-inferior recall PPV (12.6% [10.6-14.8%], 130/1032, versus
12.0% [10.0-14.3%], 114/947, p= 0.699), and increased RR (5.4% [5.0-5.7%], versus 4.9% [4.6-5.2%], p= 0.016).
Conclusion: AI-based triaging, excluding low risk mammograms from human reading, leads to a substantial reduction in reading
workload in breast cancer screening without negatively affecting performance.
Limitations: The results include 70% of the target population (27000 women). We expect to complete the study in February 2024.
Funding for this study: Funding was received from the SEDIM foundation grant, to the value of 20.000 euros.
Ethics committee - additional information: Approval granted on 30.03.2021.
Correlating breast lesions in tomosynthesis CC and MLO views using artificial intelligence (AI) (7 min)
Sarah Maier Friedewald; Chicago / United States
498
Author Block: S. M. Friedewald1, A. Dsouza2, C. Parghi3, A. Kshirsagar2; 1Chicago, IL/US, 2Marlborough, MA/US, 3Houston, TX/US
Purpose: The purpose of this study was to evaluate a deep learning (DL) model for matching regions of interest (ROI) corresponding
to the same lesion on tomosynthesis craniocaudal (CC) and mediolateral oblique (MLO) views
Methods or Background: A CC-MLO lesion correlation system (CMCS) was developed to automatically match ROIs flagged by an AI
breast lesion detection algorithm (Genius AI Detection, Hologic) in both CC and MLO views. The system combines geometric
information with similarity between pairs of ROIs to assign a lesion correlation score. ROI pairs above a pre-defined threshold are
presented to the reader for potential workflow enhancement.
864 consecutive subjects with biopsy proven malignant cancers were collected retrospectively under an IRB approved protocol from
two large multi-centre breast imaging networks and one breast imaging facility at an independent cancer centre. Ground truth was
determined by an expert using image available data. The pairs of ROIs flagged by the AI algorithm on malignant lesions were
analysed by the CMCS system and compared with ground truth to estimate accuracy.
Results or Findings: Out of 864 patients with biopsy-proven malignancies, 614 lesion ROI pairs identified by experts were detected
by the AI algorithm on both views. Analysis of these by CMSC resulted in 555 correctly matched pairs, resulting in an overall accuracy
for all findings of 90.4% (95% CI: 88.1, 92.7) for biopsy-proven cancer cases.
Conclusion: The CC-MLO lesion correlation system was able to correctly match pairs of ROIs in CC and MLO views over 90% of the
time for biopsy proven malignant lesions that were correctly flagged by the AI algorithm. This matching algorithm can be used to
assist radiologists in triangulating one-view findings in the orthogonal view.
Limitations: This study is retrospective and not performed in a clinical setting.
Funding for this study: Funding was received from Hologic Inc.
Evaluating performance of an artificial intelligence (AI) detection system on prior screening tomosynthesis studies of
breast cancer patients (7 min)
Sarah Maier Friedewald; Chicago / United States
Author Block: S. M. Friedewald1, B. Shi2, C. Parghi3, A. Kshirsagar2; 1Chicago, IL/US, 2Marlborough, MA/US, 3Houston, TX/US
Purpose: The purpose of this study was to evaluate an AI detection system in identifying breast cancer in up to two prior screening
tomosynthesis studies in patients with biopsy-proven cancer detected during their most recent screening examinations.
Methods or Background: Tomosynthesis screening and diagnostic studies with one or two associated prior screening examinations
acquired between 2014 and 2021, were consecutively collected from 814 biopsy-proven cancer patients. At least one prior screening
study (prior1) was available for 814 patients while two prior studies (prior1 and prior2) were available for 272. In the index exam
where the cancer was mammographically detected, the cancer was annotated by an expert using all available data. The annotator
also retrospectively reviewed prior1 and prior2 examinations and annotated corresponding lesions if visible irrespective of
actionability. The AI algorithm (Genuis AI Detection 2.0, Hologic) independently analysed tomosynthesis examinations and marked
potentially malignant findings with a score corresponding to the overall level of suspicion. Study-level sensitivity was calculated by
comparing the location of any AI marks with ground truth.
Results or Findings: Sensitivity for cancer in 814 cases was 90.7% (738/814, 95% CI:88.6%-92.6%) in index studies. Sensitivity was
89.9% for studies with non-calcified malignancies and 92.6% for studies with malignant calcifications. Sensitivity for retrospectively
visible findings amongst prior1 studies was 68.5% (341/498, 95%CI:63.1%-74.2%) and amongst the prior2 studies was 48.8% (79/162,
95%CI:39.9%-57.6%). For the 272 patients having 2 prior studies, average AI case score was 34.7 (SD:28.7) for prior2 studies, 40.9
(SD: 26.8) for prior1 study, and 63.8 (SD 24.7) for index studies diagnosed with biopsy-proven cancers.
Conclusion: This AI system can assist in identifying cancer on prior mammograms interpreted as normal. The temporal increase in
case score for each study potentially correlates with cancer progression.
Limitations: This is a retrospective study.
Funding for this study: Funding was received from Hologic, Inc.
AI detection of interval cancers: does size, grade, and time since screening affect sensitivity? (7 min)
Muzna Nanaa; Cambridge / United Kingdom
499
Author Block: M. Nanaa, T. van Nijnatten, N. Stranz, S. Carriero, N. Payne, I. Allajbeu, E. Giannotti, R. Manavaki, F. Gilbert;
Cambridge/UK
Purpose: The objective of this study was to evaluate AI detection of interval cancers (IC) on screening mammograms by tumour size,
grade, and time since screening.
Methods or Background: Two radiologists (8 and 3–13 years’ experience) classified 488 ICs (2011–2018) as visible or non-visible on
screening mammography. Tumour volume doubling time (TVDT) was calculated for visible cancers [TVDT=ln (2). Δt/3. (ln d1-ln d2)],
with the median TVDT for grade and receptor status of visible cancers used as a surrogate to estimate cancer size for non-visible
cancers [T(SS)=T(SD)×e-(ln(2)/TVDT)×Δt], T(SS): tumour size screening, T(SD): tumour size diagnosis. The sensitivity of a commercial
AI algorithm was analysed by tumour size, grade, receptor status, and time from screen to diagnosis at its default threshold for cancer
detection (score 10).
Results or Findings: Median screening size was 12 mm (IQR 9–18) for visible cancers (280/488), with median estimated size 2.65
mm (IQR 1.26–5) for non-visible cancers (208/488).
AI detected 58.2% (163/280) of visible and 30.7% (64/208) of non-visible cancers, p<0.001. AI localised 58.4% (31/53) of grade 1,
46.3% (103/222) of grade 2, 43.2% (87/201) of grade 3 cancers, p=0.14, 49.3% (195/395) of ER-positive cancers and 31.7% (27/85)
of ER-negative cancers, p=0.003. The median time to interval was 666 days (IQR 405–895) for localised cancers and 708 days (IQR
480–929) for non-localised, p=0.057.
The median size was 13 mm (IQR 9–19) and 12 mm (IQR 8–17) for localised and non-localised visible cancers, p=0.027, and 3.15 mm
(IQR 1.93–5.51) and 2.27 mm (IQR 0.97–4.25) for non-visible cancers, p=0.002, respectively. Sensitivity for cancers <5mm, 5–9.9mm,
and >=10mm was 33.3% (1/3), 49.4% (43/87), 62.6% (119/190) for visible; 27% (42/155), 32.2% (10/30), 54.5% (12/22) for non-
visible.
Conclusion: AI is more likely to detect larger, ER-positive cancers, with a trend towards grade 1.
Limitations: The limitations of this study were the single site; tumour size was estimated in 42.6% of cases.
Funding for this study: This research was supported by the NIHR Cambridge Biomedical Research Centre (BRC-1215-20014) and
the CRUK early detection programme grant. The views expressed are those of the authors and not necessarily those of the NIHR or
the Department of Health and Social Care. Council for At-Risk Academics (Cara) funded the research fellowship for M.N. (award no.
210211). We would like to thank the company for taking part in this study.
Ethics committee - additional information: This retrospective study used anonymised mammograms from two National Health
Service Breast Screening Programme (NHSBSP) centres under ethical approval [Health Research Authority Research Ethics Committee
(HRA REC) 20/LO/0104, HRA Confidentially Advisory Group (CAG) 20/CAG/0009, and Public Health England (PHE) Research Advisory
Committee (RAC) BSPRAC_090].
How much has AI improved over the last five years? A benchmark evaluation of different versions of an AI
mammography interpretation system (7 min)
Alejandro Rodriguez Ruiz; Nijmegen / Netherlands
Author Block: A. Rodriguez Ruiz, A-K. Brehl, N. Karssemeijer, I. Sechopoulos, R. M. Mann; Nijmegen/NL
Purpose: The study aimed to retrospectively evaluate the breast cancer detection performance of different versions of the same
mammography AI system developed since 2018.
Methods or Background: Two enriched datasets (A: 60 exams, 24 cancers, read by 107 radiologists; B: 60 exams, 20 cancers, read
by 73 radiologists) and one consecutively collected double-read screening dataset (22,961 exams with 370 cancers, including 163
screen-detected, 48 interval, and 159 next-round screen-detected) were gathered. All exams and radiologists are part of the Dutch
breast cancer screening program.
Each exam was processed by four versions of the same AI system (v1.3, v1.5, v1.7, and Beta-2023, Transpara, ScreenPoint Medical),
developed between 2018-2023. All exams were independent from the AI development process. The sensitivity of AI was compared to
that of the radiologists using the average radiologist specificity on each dataset, using parametric T-tests.
Results or Findings: In dataset A, the average radiologist specificity and sensitivity was 92% (CI: 91%-94%) and 83% (CI:
81%-85%). At this specificity, AI system versions v1.3-v1.7 achieved sensitivities ranging from 62% to 79%, while Beta-2023 achieved
88% sensitivity (CI: 68%-97%, P=0.92). In dataset B, the average radiologist specificity and sensitivity was 80% (CI: 78%-83%) and
85% (CI: 83%-88%). AI v1.3-v1.7 versions achieved 75%-85% sensitivity, while Beta-2023 achieved 95% sensitivity (CI: 75%-99%,
P=0.73).
In the screening dataset, the average radiologist specificity was 97.7% (CI: 97.4%-97.8%). Due to interval and next-round cancers,
average radiologist sensitivity was only 40% (CI: 35%-45%). AI Beta-2023 achieved a sensitivity of 48% (CI: 43%-53%), statistically
higher sensitivity than average single reading (P=0.002).
Conclusion: AI systems are continuously improving performance. In the evaluated system, the breast cancer detection performance
has improved over time to surpass that of an average radiologist.
Limitations: An identified limitation was that the study includes data from a single country.
Ethics committee - additional information: Data was collected following IRB waiver at the institution.
500
RC 1201 - Clinical guidelines for the management of primary focal liver tumours: lost in translation?
Categories: Abdominal Viscera, Multidisciplinary

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Maria Carmen Ayuso Colella; Barcelona / Spain

Diagnosis and management of benign liver lesions (15 min)

Cäcilia Reiner; Zürich / Switzerland
1. To diagnose benign liver tumours in the non-cirrhotic liver.

2. To choose the appropriate diagnostic workup.
3. To apply current management guidelines for benign liver tumours.
Diagnosis and management of HCC and indeterminate lesions in the cirrhotic liver (15 min)
Maxime Ronot; Paris / France
1. To understand the process of hepatocarcinogenesis and how it translates into imaging.

2. To describe the concept of imaging-based noninvasive diagnosis of HCC.
3. To describe the most recent versions of the main diagnostic systems in cirrhotic patients.
4. To discuss the complementary role of biopsy and imaging.
5. To discuss the concept of prognostic imaging and how it can help with the management of patients.
Management of HCC: interventional or surgical treatment (EASL) (15 min)

1. To distinguish the disparities between interventional and surgical treatment modalities

for Hepatocellular Carcinoma (HCC) delineated in the EASL guidelines.
2. To recognise the pivotal clinical factors that impact the selection between interventional
and surgical interventions for the management of HCC.
3. To utilise the recommended strategies for both interventional and surgical approaches,
evaluating their suitability across diverse clinical scenarios.
Panel discussion: The benefit of clinical guidelines in the MDT discussion (10 min)
501
RC 1216 - Artificial intelligence implementation in clinical routine: are we ready?
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Oncologic Imaging
ETC Level: LEVEL II
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Image acquisition and reconstruction (15 min)

1. To understand AI's impact on CT images.

2. To understand AI's impact on MRI images.
3. To evaluate artificial intelligence in X-ray and US.
Clinical application (15 min)

Georg Langs; Vienna / Austria
1. To understand which solutions are already implemented in market radiology.

2. To learn about future steps and implementations.
3. To know the possible direction of AI in clinical routine.
Business solutions (15 min)

1. To understand how AI may impact business solutions.

2. To learn about which solutions are already available.
3. To investigate possible and desirable future steps.
Panel discussion: Advantages and barriers (10 min)
502
RW 12 - How to improve head and neck cancer staging reports
Categories: Head and Neck, Professional Issues

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Tips on improving head and neck CT reports and MRI reports (15 min)
1. To write clear reports addressing pertinent points for head and neck cancer staging.
Short cases review, interactive discussion and critiquing of reports (45 min)

2. To show how to make reports brief yet clinically pertinent.
503
OF 12R - Enhancing service and care for cases of suspected physical abuse
Categories: Forensic Imaging, Management/Leadership, Medico-legal, Paediatric, Professional Issues, Radiographers

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
This session spotlights the imperative of a comprehensive and meticulous approach to radiography in cases involving Suspected
Physical Abuse. This session comprises three distinctive talks, each addressing pivotal aspects of radiological practice in safeguarding
the welfare of vulnerable individuals. The session is therefore aimed to equip radiographers and health professionals with a better
understanding of best practices, guidelines, practical strategies, as well as potential pitfalls so as to ultimately enhance the service
and care for individuals facing suspected physical abuse. The session will also highlight the pivotal role of radiography in safeguarding
the well-being and justice of vulnerable populations.
Moderator:
Vassilis Georgios Syrgiamiotis; Goudi / Greece

Vassilis Georgios Syrgiamiotis; Goudi / Greece
A radiographer's guide for suspected physical abuse examinations (16 min)

Jessica Eaton; London / United Kingdom
Tips for radiographers to safeguard the chain of evidence (16 min)

Michaela Dawn Davis; Dublin / Ireland
The benefits of a specialised team approach to imaging cases of suspected physical child abuse (16 min)
Jannie Bøge Steinmeier Larsen; Aarhus V / Denmark
504
RC 1211 - Spine imaging: painful dilemmas
Categories: Interventional Radiology, Musculoskeletal, Neuro

Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Alexis Kelekis; Athens / Greece

Osteoporotic fracture or metastasis? (15 min)

mario muto; Naples / Italy
1. To clarify the role of diagnostic imaging in vertebral compression fracture.

2. To explain the differential diagnosis criteria in patients with vertebral compression fractures.
3. To identify the best diagnosis technique and follow-up on those patients.
Do these imaging findings explain this patient's pain? (15 min)

Johan Van Goethem; Antwerpen / Belgium
1. To learn the appropriate imaging technique in a patient with LBP or neck pain.
2. To learn how to discern coincidental from significant findings.
3. To understand how to determine the pain generator in a patient with LBP or neck pain.
Pain management spine treatment: what to expect from imaging (15 min)
Luigi Manfré; Catania / Italy
1. To identify the pain trigger in the spine before doing specific spinal interventions.
2. To describe the flowchart in case of vertebrogenic pain leading to basivertebral nerve ablation.
3. To list the less invasive interventional radiological treatments according to the pain trigger identification.
Panel discussion: How to improve reports of the spine: structured reports according to pathology? (10 min)
505
IF 12 - Changing the game: interventional radiology's growing impact on imaging
Categories: Artificial Intelligence & Machine Learning, General Radiology, Interventional Oncologic Radiology, Interventional
Radiology, Management/Leadership
Date: March 1, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Since the first applications of image-guided minimally-invasive treatment, interventional radiology has seen tremendous
developments. In many cases, it transformed how patients are treated by either providing additional therapeutic options or even
replacing surgery in some cases. This session will discuss interventional radiology from its humble beginnings to the newest
developments and future directions. New potential applications of image guided-therapy and new approaches to training are
highlighted to provide the best possible education to the next generation of interventional radiologists.
Moderator:

From past to present: the evolution of interventional radiology (15 min)

Gilles Soulez; Montreal / Canada
New technologies in interventional procedures: which surgeries will be replaced next? (15 min)
The future of training in interventional radiology: simulators and virtual reality (15 min)
Elif Can; Freiburg / Germany
Panel discussion: How will interventional radiology impact patients' health in the next 10 years? (10 min)
506
MD 5 - Cholangiocarcinoma: new developments in imaging and treatment - recommendations for


Date: March 1, 2024 | 08:15 - 09:15 CET
CME Credits: 1
Moderator:

1. To highlight new developments in the treatment of cholangiocarcinoma.

2. To critically review recent evidence in imaging and put this in perspective of new treatment developments.

Ailton Sepulveda; Clichy / France
The hepatologist's perspective (8 min)

Mohamed Bouattour; Clichy / France

507
CUBE 13 - Aortic soup

Date: March 1, 2024 | 09:00 - 09:30 CET
Central IR Day - Topic Coordinator: Dr. Mohammad Tariq Ali
in the audience.
Moderator:

Aortic soup (28 min)

Mohammad Tariq Ali; Norwich / United Kingdom
1. To discuss the various types of devices used in aortic aneurysmal disease.

2. To learn how to decide on device selection based on underlying anatomy/pathology.
508
RPS 1317 - Hot topics in emergency radiology
Categories: Abdominal Viscera, Artificial Intelligence & Machine Learning, Chest, Contrast Media, Emergency Imaging, Head and
Neck
Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Maria Antonietta Mazzei; Siena / Italy
Automated torso CT haemorrhage burden: a novel precision-medicine decision support tool for transfusion in trauma
patients (7 min)
Nathan Sarkar; Baltimore / United States
Author Block: N. Sarkar, L. Zhang, M. Unberath, U. Bodanapally, J. Hu, G. Li, O. Turan, A. Corkum, D. Dreizin; Baltimore, MD/US
Purpose: Torso haemorrhage is a leading cause of preventable death in traumatic injury and can be precisely quantified on whole-
body CT using deep learning segmentation methods. Prior studies have shown that quantitative assessment of haemorrhage burden
in individual body cavities is associated with transfusion requirement. The purpose of this study is to examine total torso CT
haemorrhage burden (tCTHB) in the thorax, abdomen, and pelvis as a predictor of massive transfusion (MT).
Methods or Background: A dataset of 5060 trauma CT scans was used to select patients with hemothorax, haemoperitoneum,
pelvic haematoma, or combinations thereof (n=593). A state-of-the-art deep learning method (nnU-net) was used to derive
segmentation masks and volumes. tCTHB was calculated as the sum total of segmented torso haemorrhage in mL. MT, defined as at
least 10 packed red blood cells (PRBCs) in 24 hours, served as the outcome. The area under curve (AUC) of shock index (SI) + tCTHB
was compared to that of SI alone. Optimal tCTHB cutoff was determined by Youden J index. Spearman correlations between PRBCs
and volumes were determined for tCTHB and individual intracavitary haemorrhage types.
Results or Findings: The AUC of SI + tCTHB (0.86 [95% CI: 0.75-0.97]) was higher (p=0.02) than the AUC of SI alone (0.67 [95% CI:
0.50-0.84]). 363 mL served as an optimal cut-off at peak Youden J index. The sensitivity and specificity of SI>1 and tCTHB>363 mL
was 90% and 81%, whereas SI>1 alone had sensitivity and specificity of 50% and 91%. Spearman’s r was moderate for tCTHB (0.42),
low for haemorrhage volume in individual compartments (0.12-0.23), and low for SI (0.28).
Conclusion: tCTHB is a promising precision medicine marker for predicting massive transfusion, significantly improving accuracy
over shock index alone.
Funding for this study: Funding was received from NIH K08 Grant EB027141-01A1.
Ethics committee - additional information: This study was reviewed and approved by the University of Maryland Baltimore
Institutional Review Board (UMB IRB). It was determined by the UMB IRB to be exempt from the need for informed consent due to no
more than minimal risk to patients.
Coronary calcification score in polytrauma CT of severely injured polytrauma patients as a prognostic factor for
hospital mortality and intensive care unit treatment (7 min)
Hans-Jonas Meyer; Leipzig / Germany
509
Author Block: H-J. Meyer, T. Dermendzhiev, T. Denecke, M. Struck; Leipzig/DE
Purpose: Coronary artery calcifications defined by computed tomography (CT) provide prognostic relevance for vascular disorders
and coronary heart disease, whereas their prognostic relevance in severely injured trauma patients remains unclear.
Methods or Background: All consecutive trauma patients requiring emergency tracheal intubation before initial computed
tomography at a level-1 trauma centre over a 12-year period (2008-2019) were reanalysed. The Weston score, a semiquantitative
method to quantify the coronary calcifications, was evaluated as prognostic variable based upon whole-body trauma CT analysis.
Results or Findings: 458 patients (75% male) with a median age of 49 years, median injury severity score (ISS) of 26, and in-
hospital mortality rate of 23.1% met the inclusion criteria and were analysed. Coronary artery calcification was present in 214
patients (46.7%). After adjustment with age, ASA classification ≥ III, ISS abbreviated injury scale head, and shock, the Cox
proportional hazard model identified the Weston score as independent predictor for 24- hour-, 30- day-, and in-hospital mortality (HR
1.2, 95% CI 1.08 – 1.34, p= 0.001; HR 1.09, 95% CI 1.02 – 1.18, p= 0.017; and HR 1.1, 95% CI 1.02 – 1.18, p= 0.012, respectively).
Conclusion: CT-derived coronary calcification was significantly associated with hospital mortality in intubated polytrauma patients. In
the subgroup of survivors, it was significantly associated with ICU LOS but not with mechanical ventilation duration.
Limitations: First, it is a single-centre retrospective study. Second, only patients requiring tracheal intubation and who underwent
whole-body CT were included in this analysis. Third, although the Weston score is a semiquantitative imaging analysis, we cannot
exclude investigator- related bias.
Ethics committee - additional information: The retrospective analysis was approved by the Ethical Committee at the Medical
Faculty, Leipzig University, Leipzig, Germany, (IRB00001750, project ID 441/15ek, September 14, 2020).
Prevalence and extent of blunt cervical vascular injury (BCVI): a longitudinal single-centre study over 19 years (7 min)
Bianca Mazini; Lausanne / Switzerland
Author Block: B. Mazini, S. Sanson, V. Dunet, A. Denys, S. Schmidt; Lausanne/CH

Purpose: To aim of this study was to assess the prevalence and extent of blunt cervical vascular injury (BCVI) in the context of high
energy trauma, and to analyse the associated radiological findings, immediate clinical management and patients’ outcome according
to the extent of BCVI.
Methods or Background: This single-center retrospective study included 89 consecutive patients with acute BCVI proven by
computed tomography angiography-(CTA), that were selected from 8053 polytrauma patients addressed to our emergency
department within 19 years. Two radiologists blinded to all clinical information jointly reviewed images regarding location and extent
of BCVI, associated radiological findings and cerebral complications (stroke). Epidemiological and clinical data, and outcomes were
retrieved from patients’ medical records. Univariate and multivariate statistical analysis were performed. A p-value <0.05 was
considered statistically significant.
Results or Findings: The prevalence of BCVI in our polytrauma population was 1.15% (26 females, mean age 46 years). In the 89
patients with BCVI 119 cervical arterial dissections were found. In 23 patients (25.8%) acute stroke was associated; overall mortality
was 19% (n = 17). Patients’ mortality was significantly predicted by clinical factors (Glasgow coma scale, and intubation at arrival),
and radiological features, such as a higher grade of vascular injury and exact length of left vertebral artery dissection. In 61 (68.5%)
patients with cervical fractures the vascular injury was adjacent. However, 28 (31.5%) arterial injuries were not associated with any
bone fracture.
Conclusion: BCVI is a rare but serious complication of high energy cervical trauma, with high rate of morbidity and mortality.
Systematic screening of the cervical arteries with angio-CT seems necessary in all polytrauma patients since >30% of cervical arterial
lesions are not associated with nearby bone fractures.
Artificial intelligence derived models may be used to predict the risk of first variceal haemorrhage: a novel technique
based on non-invasive computed tomographic portography images (7 min)
Shang Wan; Chengdu / China
510
Author Block: S. Wan, M. Liu, Y. Wei, M. Liu, B. Song; Chengdu/CN
Purpose: The acute first variceal haemorrhage (FVH) from oesophageal varices is a life-threatening complication in patients with
liver cirrhosis and is an indication of emergency admission, the invasive endoscopy is now the gold standard for FVH prediction.
Therefore, we aimed to develop and validate predictive models based on novel algorithm of artificial intelligence to noninvasively
predict FVH, using computed tomographic portography (CTP) images.
Methods or Background: This multicentre and retrospective study contains 186 patients collected from West China Hospital and
Chengdu Medical College Hospital from November 2016 to October 2022, patients were divided into a FVH group (n= 108) and a non-
FVH group (n= 78) according to the FVH history, and were divided into a training cohort (n= 126) and a validation cohort (n= 60).
With deep learning algorithm of artificial intelligence, two novel models developed with convolutional neural networks (CNN) and
vision-transformer blocks were proposed to predict the risk of FVH. Statistical analyses and the receiver operating characteristic
(ROC) curves were performed to determine the predictive performance, and the decision curve analysis (DCA) was used to evaluate
the clinical value of the models.
Results or Findings: For the CNN networks, the best area under the ROCs (AUCs) for FVH prediction were 1.0 (95%CI, 1.0-1.0,
training cohort) and 0.913 (95%CI, 0.887-0.939, validation cohort), respectively, with the accuracy, sensitivity, specificity of 0.9,
0.833, 0.967. As for the vision-transformer networks, the best AUCs were 0.997 (95% CI, 0.997-0.999, training cohort) and 0.940 (95%
CI, 0.919-0.963, validation cohort), respectively, with the accuracy, sensitivity, specificity of 0.9, 0.8, 1.0. The DCA suggest that all
proposed models can bring positive clinical benefits, of which the vision-transformer model brings the highest benefits.
Conclusion: The deep learning networks may be used to predict the risk of FVH non-invasively, of which the vision-transformer
model could yield highest performance.
Limitations: The sample size needs to be further expanded
Ethics committee - additional information: It has been approved by the medical centers.
Computed tomography imaging in earthquake trauma patients with crush syndrome and different levels of creatine
kinase: is there a correlation with renal enhancement? (7 min)
Sezer Nil Yılmazer Zorlu; Ankara / Turkey
Author Block: S. N. Yılmazer Zorlu, M. Kul, A. Gürsoy Çoruh; Ankara/TR

Purpose: The objective of this study was to investigate whether varying serum creatine kinase (CK) levels in patients with crush
syndrome have an impact on renal enhancement on computed tomography (CT) even in cases where renal injury is not clinically
evident.
Methods or Background: Portovenous phase abdominopelvic CT scans of earthquake (Turkey, February 2023) survivors and of
patients without involvement in the earthquake and with preserved renal function (control group) were retrospectively analysed.
Attenuation measurements were made from the cortex and medulla of the left kidney at the level of renal hilus, abdominal aorta
(level of renal artery origin) and main portal vein. The medulla-to-cortex ratio was normalised to the contrast phase using the formula:
nM/C-R= (Medulla Attenuation/Cortex Attenuation)/ (Portal Vein Attenuation/Aorta Attenuation). Renal cortex/aorta (C/A) attenuation
ratio was also calculated.
Earthquake survivors were categorised by serum CK levels into risk groups for acute kidney injury (AKI) (high-risk group:≥773 U/l,
low-risk group:<773 U/l).
CT findings were compared between groups and correlations with laboratory findings were made.
Results or Findings: In the earthquake survivor group 33 patients were in the low-risk and 12 in the high-risk category for AKI. The
control group consisted of 50 patients. None of the 95 patients had impaired renal function.
Though no significant age, gender, nC/M-R and creatinine value differences were found (p> 0,05), the high-risk group had significantly
lower renal C/A attenuation ratios than the low-risk and control groups (p= 0.043; p= 0.023, respectively).
There was a negative correlation between CK and C/A attenuation ratios and a positive correlation between calcium levels and C/A
attenuation ratios. The nC/M-R and serum levels for creatinine, potassium and CK did not significantly correlate.
Conclusion: Even in the absence of clinically evident AKI, cortical enhancement is diminished in patients with crush syndrome with
higher CK levels.
Limitations: Laboratory follow up of patients could not be made.
Ethics committee - additional information: This study was approved by the Research Ethics Committee of Ankara University
Faculty of Medicine.
Visual communication of traumatic injuries by automatic rendering of a graphic summary based on natural language
processing of CT reports (7 min)
Nitai Bar; Haifa / Israel
511
Author Block: N. Bar, A. Ilivitzki, E. Bercovich; Haifa/IL
Purpose: Initial medical response in traumatic injury scenarios requires a multidisciplinary approach involving radiologists as well as
trauma surgeons, emergency medicine physicians and lead administrators. In such settings immediate and efficient data flow is
essential, but remains a major challenge. This is particularly pronounced in mass casualty incidents, where triage, patient navigation
and team coordination become critical. We aim to develop a tool leveraging an advanced language model to extract and classify
findings from CT reports to achieve real-time rendering of a graphical summary of traumatic findings, supporting life-saving decisions
taken under time pressure in the trauma room.
Methods or Background: Our dataset comprised chest CTA trauma protocol scans performed between 1.1.2012–1.4.2022 at a
tertiary trauma centre in the north of Israel. Traumatic injuries were manually annotated to produce “gold-standard” labels. GPT
models were subsequently used to extract meaningful data from the reports and classify them into predefined trauma-related
categories, enabling a rule-based automatic rendering of a graphical summary using open-source graphical tools. Performance of the
language model was assessed with mean AUC, F1, and exact match scores as compared to the gold-standard labels.
Results or Findings: Preliminary findings suggest exact match scores >80% and F1 scores >90%, demonstrating the model's ability
to accurately predict and classify traumatic injury labels, including rare ones. The NLP model thus allows real-time generation of
structured data, which can be utilised to streamline communication between teams, promote notification of acute findings and ensure
adequate patient prioritisation.
Conclusion: We introduce a pipeline harnessing natural language models and graphic tools to allow real-time data analysis.
Graphical illustrations of medical data based on automatically generated classifiers from medical reports are a promising novel tool
contributing to optimise the trauma workflow.
Limitations: This was a single centre study, which solely focused on chest CTs.
Ethics committee - additional information: This study was approved with the approval code: RMB-02-0503 - Use of radiology
reports was permitted under the condition of de-identified data analysis, patient consent was waived.
Portal venous contrast enhancement ratio of the adrenal glands and spleen as prognostic marker of mortality in
patients with acute mesenteric ischaemia (7 min)
Felix Alexander Pfister; Leipzig / Germany
Author Block: F. A. Pfister, M. Mehdorn, C. Schwartner, D. Seehofer, H-M. Tautenhahn, M. Struck, T. Denecke, H-J. Meyer; Leipzig/DE
Purpose: Contrast enhancement of the adrenal gland defined by computed tomography (CT) was previously analysed as a
prognostic factor for critically ill patients due to various diseases. However, no study investigated this quantitative parameter in
patients with acute mesenteric ischaemia, a potentially lethal disease. Therefore, the aim of this study was to evaluate the prognostic
value of the contrast enhancement of the adrenal glands in patients with clinically suspected AMI.
Methods or Background: All patients with clinically suspected AMI were retrospectively assessed between 2016 to 2020. All
patients underwent surgical exploration. Overall, 134 patients (52 female patients, 38.8%) with a mean age of 69.2 ± 12.4 years were
included into the present analysis. For all patients, the preoperative CT was used to calculate the contrast media enhancement of the
adrenal glands and the spleen.
Results or Findings: Overall, 27 (20.1%) patients died within the 24-hour period, and 94 (70.1%) within 30 days. There were
statistically significant differences regarding the mean values for adrenal-to-spleen ratio for 24h-mortality (p= 0.001) and 30-day
mortality (p= 0.004), whereas the radiodensity of the inferior vena cava and the radiodensity of the spleen was statistically significant
between survivors and non-survivors after 30-days (p= 0.037 and p= 0.028, respectively). In cox regression analysis mean adrenal
radiodensity was associated with mortality after 24h with an HR of 1.09 (95% CI 1.02-1.16, p= 0.01).
Conclusion: The contrast media enhancement of the adrenal gland is associated with the 24h- and 30-day mortality in patients with
AMI. However, the identified associations are worse compared to previous analyses regarding other critical diseases.
Limitations: First, it is a retrospective single-centre study with known inherent bias. Second, despite the quantitative nature of the
measurements, there can still be some intrareader variability.
Comparison of brain injury in patients with and without facial fractures (7 min)
Iulia Tatiana Lupașcu; Bucharest / Romania
512
Author Block: I. T. Lupașcu, S. Hostiuc, C. Adrian, B. Popa, C. A. Minoiu; Bucharest/RO
Purpose: This study aimed to examine the association between facial fractures and brain injury and to compare brain injuries in
facial fracture patients with non-facial fracture patients.
Methods or Background: Cerebral CT of 492 polytrauma patients, who were admitted to the hospital between January 2019 to July
2023, were retrospectively evaluated.
Results or Findings: From the total of 492 patients, 129 (26%) had facial fractures (102 men, 27 women, mean age 45 ±,17 years)
and 363 (74%) were without facial fractures (253 men, 110 female, mean age 46 ±,17 years).
Facial fractures were significantly correlated with brain injuries (p< .001), showing a greater incidence of brain lesions (54%, n= 68),
compared to the non-facial fractures group (31%, n= 111). Subdural hematoma was the most frequent lesion (29%, n= 44) in the
facial fracture group, while intracerebral haemorrhage was the most frequent lesion (33%, n= 73) in the non-facial fracture group.
The zygomatic bone was the most frequently fractured (28%, n= 67) and it was significantly correlated to the presence of
intracerebral haemorrhage, subdural and subarachnoid haemorrhage, but also with pneumocephalus and diffuse brain swelling (p<
.05).
Frontal sinus fracture (12%, n= 29) and its right or left side location were correlated with the presence and location of epidural
haematoma and intracerebral haemorrhage (p< .05).
Maxillary fractures were observed in 27% (n= 65) patients, nasal bones in 25% (n= 58) and mandible fractures in 8% (n= 19).
In the facial fractures group, diffuse brain swelling was present in 18% (n= 23) patients, pneumocephalus in 10% (n= 12) and brain
herniation in 6% (n= 8), while in the non-facial fractures group, diffuse brain swelling was observed in 7% (n= 27), pneumocephalus
in 2% (n= 6) and brain herniation in 3% (n= 10) of patients.
Conclusion: There is a significant association between facial fractures and traumatic brain injury, with zygomatic bone being the
most frequently fractured and subdural hematoma the most associated brain lesion.
Limitations: The limitation of the study was the retrospective design.
Assessment of Syngo.CT LVO: detection in photon counting CT and conventional CT (7 min)

Jan Boriesosdick; Porta Westfalica / Germany
Author Block: J. Boriesosdick, N. Haag, S. Saeed, C. Mönninghoff, J. Borggrefe, J. R. Kroeger; Minden/DE

Purpose: Syngo.CT LVO detection (SIEMENS Healthineers) is a prototype for AI-assisted detection of large vessel occlusion in the
anterior circulation area in patients with suspected ischaemic stroke. We investigated the sensitivity and specificity of the algorithm in
photon counting CT (PCCT; NAEOTOM Alpha) and conventional CT (SOMATOM GoTop/X.ceed/GoAll). Furthermore, we investigated
whether there were differences in detection rate between proximal (internal carotid artery, M1 segment of middle cerebral artery)
and distal (M2 segment) vessel occlusions.
Methods or Background: We retrospectively analysed 443 cases with CT-Angiography of the head (Bv44-Kernel) with Syngo.CT
LVO Detection. 176 cases showed vascular occlusion in the anterior circulation area. 150 cases were investigated with the PCCT with
a total of 50 occluded vessels, 293 cases were investigated in conventional CT scanners with a total of 126 occluded vessels.
Results or Findings: The algorithm showed an overall sensitivity of 70.5% and a specificity of 98.5%. Thereby, we found significant
differences (p= 0.013) in sensitivity between PCCT (84%) and conventional CT (65.1%). Proximal vessel occlusions were detected
significantly better than M2- occlusions in all scanners with a sensitivity of 86.2% versus 26.1% (p< 0.001). While sensitivity regarding
proximal vessel occlusions did not differ significantly between scanners (90.2% versus 84.3%; p= 0.359), sensitivity regarding M2-
occlusions showed a significant difference in favor of PCCT (55.6% versus 18.9%; p= 0.039).
Conclusion: Syngo.CT LVO detection could assist radiologists in the rapid detection of large vessel occlusions as a triage tool in
clinical routine. Thereby, the algorithm seems to work better with PCCT data, especially for M2-occlusions, probably due to a higher
image quality compared to conventional detectors. Overall, the algorithm should be further trained specifically regarding the
detection of M2- occlusions.
Limitations: The sign of the study was monocentric, and cases were prone to selection bias. Furthermore, there were only a few
cases with M2- occlusions in PCCT.
Ethics committee - additional information: The study was conducted according to the guidelines of the Declaration of Helsinki
and approved by the Institutional Ethics Committee of the Ruhr-University Bochum (19 July 2021, reference number 2021-827).
Incidence of contrast-induced acute kidney injury (CI-AKI) in trauma patients undergoing contrast-enhanced computed
tomography using iso-osmolar contrast agent (7 min)
Minji Gim; Suwon-si, Gyeonggi-do / Korea, Republic of
513
Author Block: M. Gim, H. S. Lee, K. Lee, J. K. Kim, J. Huh; Suwon/KR
Purpose: In trauma patients, no report has been published on the preventive effect of iso-osmolar contrast agent for contrast-
induced acute kidney injury (CI-AKI). We aimed to evaluate the incidence and severity of CI-AKI, and its predictive factors in trauma
patients.
Methods or Background: From the trauma registry in a regional trauma centre in Korea, patients who underwent CT scans with
iodixanol and were followed up for at least 72 hours were consecutively included. Patient demographic details, co-morbidities, and
laboratory test results were collected. CI-AKI was defined by the 2012 kidney disease improving global outcomes guideline. The
severity of CI-AKI was classified by the RIFLE criteria. Trauma severity was assessed by the injury severity score (ISS) category.
Predictive factors of CI-AKI were evaluated by univariate and multivariate logistic regression.
Results or Findings: Of 1115 patients who underwent CT with iodixanol, 799 were included in this study. The incidence of CI-AKI
was 3.80% (30/799). Severe renal failure according to RIFLE criteria was 2.87% (23/799). Incidence of CI-AKI according to the ISS was
1.52% (9/592) in the minor group, 3.95% (7/177) in the moderate group, 42.30% (11/26) in the serious group, and 75.0% (3/4) in the
severe group. In univariate analysis, significant predictive factors of CI-AKI based odds ratio (OR) included hypertension (2.87), heart
disease (4.06), ISS serious category (47.50), and ISS severe category (194.33). In multivariate analysis, significant predictive factors
were the ISS serious category [16.70 (95% CI, 16.70-138.96], and ISS severe category [245.22 (95% CI, 21.88-2748.38)].
Conclusion: In trauma patients who underwent CT with iodixanol, the overall incidence of CI-AKI of 3.8% (30/799) was considerably
low. The main predictive factors of CI-AKI involved hypertension, heart diseases, and the ISS serious and severe categories, but its
incidence and severity largely relied on the severity of trauma.
Ethics committee - additional information: This study was approved by IRB of Ajou University Hospital.
Utilising dual-layer detector spectral CT to predict clinical risk stratification in the diagnosis of acute pulmonary
embolism (7 min)
Author Block: Y. Zhou, Z. Wang, L. Lei, W. Cao, S. Dong, S. W. Yue, Z. Zhou; Zheng Zhou/CN
Purpose: The research utilised the iodine density map in conjunction with the effective atomic number map (ID-Z-eff map) to
evaluate the rate of emboli detection in individuals diagnosed with acute pulmonary embolism (PE). Additionally, the study
quantitatively assessed the burden of pulmonary thrombosis to predict the clinical risk stratification in these patients.
Methods or Background: This prospective study comprised a cohort of 83 individuals diagnosed with PE who underwent CTPA
utilising dual-layer detector spectral CT (DLCT). The spectral data were reconstructed to generate conventional CT images (CI) and ID-
Z-eff maps. The number of identified emboli in both cohorts was quantified, and the diagnostic efficacy was assessed. The distribution
of emboli was analysed using deep-learning lung segmentation. The severity of pulmonary embolism was evaluated using the Qanadli
and Mastora scores. The receiver operating characteristic (ROC) curves were used to assess each score in distinguishing clinical risk
stratification.
Results or Findings: 202 and 243 emboli were detected in conventional CT images (CI) and ID-Z-eff maps, respectively. The ID-Z-eff
map demonstrated superior embolus detection capabilities compared to CI images (p< 0.05). 42/35 patients were categorized as low-
/intermediate risk. There was a significant difference in pulmonary embolism index between the low-risk and medium-risk groups
(Qanadli: 14.8%, 39.5%, p< 0.001; Mastora: 11.2%, 34.3%, p< 0.001). The pulmonary embolism index Qanadli and Mastora scores
demonstrated promising discrimination in the low-/intermediate-risk groups (Qanadli: AUC=0.876, Mastora: AUC=0.875).
Conclusion: The ID-Z-eff map can enhance the sensitivity and accuracy of acute pulmonary embolism emboli detection. The
pulmonary embolism index based on the ID-Z-eff map utilising DLCT can accurately predict the clinical risk stratification of acute
pulmonary embolism.
Limitations: The sample size in this research is relatively limited.
Ethics committee - additional information: This study was approved by the Institutional Review Board
BRIXIA score, laboratory findings and vaccination status for prediction of mortality in severe COVID-19 pneumonia (7
min)
Valentina Opancina; Kragujevac / Serbia
514
Author Block: V. Opancina1, M. Muto2, N. Zdravkovic1, E. Ciceri3; 1Kragujevac/RS, 2Naples/IT, 3Milan/IT
Purpose: The aim of our study was to investigate correlations between main laboratory parameters, vaccination status, and Brixia
score, as well as to confirm if the Brixia score is a significant independent predictor of an unfavourable outcome (death) in
hospitalised COVID-19 patients.
Methods or Background: Chest X-ray has verified its role as a crucial tool in COVID-19 assessment due to its practicability,
especially in emergency units. Likewise, the Brixia score has proven itself as a useful tool for COVID-19 pneumonia grading. The study
was designed as a cross-sectional multicentric study. It included patients with a diagnosed COVID-19 infection who were hospitalised
between September 2021 and February 2022 and had RT-PCR-confirmed COVID-19 and initial CXR and laboratory results.
Results or Findings: This study included a total of 279 patients with a median age of 62 years. The only significant predictor of an
unfavourable outcome (death) was the Brixia score (adjusted odds ratio 1.148, p= 0.022). Also, the results of the multiple linear
regression analysis (R2= 0.334, F= 19.424, p< 0.001) have shown that male gender (B= 0.903, p= 0.046), severe COVID-19
(B=1.970, p< 0.001), and lactate dehydrogenase (B=0.002, p< 0.001) were significant positive predictors, while albumin level
(B=–0.211, p< 0.001) was a significant negative predictor of the Brixia score.
Conclusion: Our results provide important information about factors influencing the Brixia score and its usefulness in predicting
unfavourable outcomes in COVID-19 patients. These findings have clinical relevance, especially in epidemic and emergency
circumstances. In order to monitor these patients effectively and to achieve efficient patient management, CXR is a great tool due to
its wide availability. Also, the application of the Brixia score is clear and simple and benefits clinicians in daily praxis.
Limitations: The limitations of the study are its small sample size and the lack of a control group.
Ethics committee - additional information: This study was approved by the Ethics Committee of University Clinical Center
Kragujevac, 62/2022.
515
HW 13Sc - Imaging strategies in acute stroke

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1

1. To become familiar with imaging algorithms for a selection of acute stroke patients for IVT and EVT.
2. To learn about imaging protocols based on different imaging modalities.
3. To develop practical skills in detecting haemorrhage and large vessel occlusion.
4. To develop practical skills in detecting non-salvageable infarct "core" and potentially viable brain tissue.
5. To become familiar with treatment strategies for stroke following the latest research guidelines.

Myriam Edjlali-Goujon; Paris / France
516
E³ 1318 - Cardio-oncology
Categories: Cardiac, Imaging Methods, Oncologic Imaging

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Giuseppe Muscogiuri; Roma / Italy

Giuseppe Muscogiuri; Roma / Italy
Cardiac tumours (28 min)

Luigi Natale; Anguillara Sabazia / Italy
1. To describe the three most common cardiac tumours per cardiac location.
2. To assess the important features in tumour evaluation on CT and MRI.
3. To integrate imaging findings and location into the most likely tumour diagnosis.
Cancer treatment effect on the heart (28 min)

1. To list the three oncological drugs most commonly related to cardiac effects.
2. To analyse cancer treatment effects on cardiac function and morphology.
3. To advise on which imaging modality to use in specific clinical scenarios.
Myocarditis in immune checkpoint inhibitor treatment (28 min)

Bernd J. Wintersperger; Toronto / Canada
1. To describe the effect of immune checkpoint inhibitor (ICI) treatment on the heart.
2. To assess the imaging features of ICI myocarditis in MRI.
3. To reflect on the importance of timely ICI myocarditis diagnosis.
517
RC 1303 - Imaging of microvascular dysfunction
Categories: Cardiac, Contrast Media, Hybrid Imaging, Imaging Methods

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:

Invasive assessment of microvascular dysfunction (15 min)

Tommaso Gori; Mainz / Germany
1. To describe the pathophysiology of coronary perfusion.

2. To describe the clinical presentation, epidemiology and mechanisms of microvascular dysfunction.
3. To discuss the impact of therapies on microvascular resistance.
PET/CT and SPECT imaging of microvascular dysfunction (15 min)

Thomas Schindler; Villach / Austria
1. To learn the non-invasive assessment of myocardial blood flow (MBF) and myocardial flow reserve (MFR) with PET/CT and SPECT
imaging, strengths, and pitfalls.
2. To learn the description and characterisation of coronary microvascular angina.
3. To learn the different types of coronary microvascular dysfunction (CMD).
MRI imaging of microvascular dysfunction (15 min)

1. To learn the indications and appropriate use of the modality in microvascular dysfunction, including protocols.
2. To explore different clinical scenarios of microvascular dysfunction.
3. To learn the imaging appearances on cardiovascular MRI.
Panel discussion: When to think about microvascular dysfunction: who should we image and when? (10 min)
518
Categories: Breast, Cardiac, Chest, Genitourinary, Oncologic Imaging

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderators:
Ferdia Aidan Gallagher; Cambridge / United Kingdom

Digital breast tomosynthesis with advanced reading methods for the Dutch national breast cancer screening program
(STREAM): design and rationale of a prospective non-randomised screening trial (8 min)
Leonardus Bernardus van den Oever; 's-Hertogenbosch / Netherlands
Discussant (4 min)
Eva Maria M. Fallenberg; München / Germany
European lung cancer screening implementation: 4-IN-THE-LUNG-RUN trial (8 min)

Mario Silva; Parma / Italy
Discussant (4 min)
Constance De Margerie-Mellon; Paris / France
Artificial intelligence as concurrent reader in prospective European Lung Cancer Screening (4-IN-THE-LUNG-RUN) trial
(8 min)
Mario Silva; Parma / Italy
Discussant (4 min)
Diagnostic performance of DW-MRI in advanced ovarian cancer: first results of the Dutch prospective multicentre
MISSION trial (8 min)
Max Lahaye; Amsterdam / Netherlands
Discussant (4 min)
519
Deep learning-aided classification of disease distribution patterns in metastatic ovarian cancer (8 min)
Cathal Mccague; Cambridge / United Kingdom
Discussant (4 min)
Renato Cuocolo; Napoli / Italy
Sharing clinical and medical imaging data from the multicentre SCOT-HEART and DISCHARGE trials on coronary CT
angiography: first experience using the open-source medical image management system XNAT (8 min)
Discussant (4 min)
520
HW 13Cd - Which cardiac findings every general radiologist should know
Categories: Cardiac, Chest, General Radiology, Imaging Methods

ETC Level: LEVEL II
Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:


Sue Thomas; Bournemouth / United Kingdom
Nicola Galea; Roma / Italy
1. To become familiar with typical and atypical imaging findings of cardiac findings to be reported in non-cardiac CT and MRI exams.
3. To discuss the limits and technical drawbacks of non-cardiac CT and MRI for the identification and characterisation of cardiac
findings.
4. Understand the additional diagnostic value of ECG gating.
5. To learn how to report cardiac findings in non-cardiac CT and MRI.
521
RPS 1312 - Neuro- and skeletal-paediatric radiology
Categories: Head and Neck, Musculoskeletal, Neuro, Oncologic Imaging, Paediatric

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Ignasi Barber; EspluguesdeLlobregat / Spain
Evaluating AI and human-derived tumour volume estimations in paediatric osteosarcoma: association with histological
response and survival predictions (7 min)
Lisa C. Adams; München / Germany
Author Block: L. C. Adams1, K. K. Bressem1, W. Morakote2, V. Suryadevara2, A. Pribnow2, S. Spunt2, L. Baratto2, J. Rosenberg2, H. E.
Daldrup-Link2; 1Berlin/DE, 2Palo Alto, CA/US
Purpose: The study aimed to assess the potential of using human- and AI-assessed tumour volume changes as markers for
treatment response in paediatric osteosarcoma, in comparison to histological necrosis, and to evaluate their association with patient
survival.
Methods or Background: In paediatric osteosarcoma, the extent of tumour necrosis post-surgery is used to estimate treatment
response. This single-centre retrospective study involved examining pre- and post-chemotherapy MRI scans of 57 paediatric and
young adult patients with histologically confirmed osteosarcoma. A subset of these patients also underwent 18F-FDG PET. We
correlated human- and AI-assessed tumour volume changes with overall (OS) and recurrence-free survival (RFS) metrics using
concordance correlation (CC). Additionally, quantitative measures such as tumour volume/size changes and histological necrosis (%)
were assessed for their impact on OS/RFS through logrank tests and Cox regression. We further compared tumour ADC and SUV
between responders (≥90% histological necrosis) and non-responders.
Results or Findings: AI-assessed tumour volume change measurements were a stronger predictor of OS than histological necrosis
(Harrell's C=0.88 versus C=0.70). Human assessments did not outpace necrosis in predicting survival outcomes. When examining
histological necrosis ≥90% and AI-driven volume change of <25% post-chemotherapy, both were linked to improved OS (p=0.021
and p<0.0001 respectively). There was high agreement between human and AI measurements (CC coefficient >0.95). Furthermore,
patients with a histological necrosis of ≥90% had higher ADC values both pre- and post-chemotherapy (both p=0.004), and displayed
lower pre-therapy SUV values when compared to non-responders (p=0.001).
Conclusion: Both human and AI-derived tumour volume changes showed a correlation with histological necrosis, indicating their
potential as markers for treatment response and OS in paediatric osteosarcoma. AI-driven volume change measurements
demonstrated stronger predictive capabilities for OS than histological necrosis.
Limitations: Our research was based in a single centre, potentially limiting wider applicability.
Funding for this study: Funding was received from the National Cancer Institute (grant R01CA269231).
Ethics committee - additional information: This study was approved by the IRB, with approval number 48854.
Intracranial haemorrhage detected with prenatal MRI in foetuses with Chiari II malformation: indication or
contraindication for foetal surgery (7 min)
Marlene Stuempflen; Vienna / Austria
522
Author Block: S. Hui1, D. Prayer2, P. Kienast2, J. Binder2, K. Goeral2, C. Mitter2, T. Dorittke2, G. Kasprian2, M. Stuempflen2;
1
Guangzhou/CN, 2Vienna/AT
Purpose: Intracranial haemorrhages (ICH) have emerged as a notable association in Chiari II malformation (CM II), yet their origins
and clinical implications remain elusive. This study aims to validate the prevalence of ICH in CM II, investigate contributing factors to
ICH, and delineate the phenotypic attributes associated with CM II and ICH.
Methods or Background: A retrospective review of foetal MRI scans obtained in foetuses with CM II presenting (January 2007 to
December 2022) at Vienna General Hospital was performed for ICH utilising EPI-T2* blood-sensitive sequence. Foetuses with aqueduct
stenosis (AS) were also included as a control group. The incidence of ICH and corresponding gestational ages were compared between
CM II and AS cases, and morphometric measurements (inner/outer CSF spaces, posterior fossa, venous structure) were compared
among the three 1:1 age-matched groups: CM II+ICH, CM II-ICH, and AS+ICH. Additionally, a co-occurrence network was constructed
to visualise associations between diagnostic features in ICH cases.
Results or Findings: A total of 101 foetuses with CM II and 90 controls with AS were included. The prevalence of ICH in foetuses
with CM II was higher compared to the AS cases (28.7% vs 18.9%), accompanied by congested veins (deep vein congestion mainly in
young foetuses, and cortical veins also affected in older foetuses). ICH was correlated with more distal vermis ectopia, reduced outer
CSF spaces, and clivus-supraocciput angle. The co-occurrence network analysis underscored venous congestion and venous sinus
stenosis as pivotal components within the network.
Conclusion: The prevalence of ICH among foetuses with CM II accentuates the interplay of venous congestion, ICH, and vasogenic
oedema, hinting at potential anatomical attributes underlying the vicious cycle. Prenatal repair surgery should be applied to these
cases as soon as possible.
Limitations: The retrospective nature of this study was an identified limitation.
Ethics committee - additional information: This study was approved by the Ethics Committee with approval code: 1716/2017.
Using 'fast' MRI head scan to improve paediatric patient experience for investigation of headaches (7 min)
Laura Christine Walker; Pathhead / United Kingdom
Author Block: L. C. Walker, A. Quigley, S. Choi, L. Armstrong, D. Evans, R. Kirkbride; Edinburgh/UK

Purpose: There are numerous requests, per year, for Magnetic Resonance Imaging (MRI) head scans for paediatric patients who
present with headaches, usually returning a normal resultant MRI scan. A fast head (FH) imaging protocol (IP) was developed,
reducing scan length from 12 minutes 51 seconds (Standard Head (SH) IP) to four minutes 22 seconds, a 66% time reduction in scan-
time, potentially improving patient experience.
Methods or Background: 50 retrospective paediatric patients (mean age, 10 years ± 3.215 Standard Deviation (SD), referred for
headaches, who had both FH and SH IP, at the same appointment, were included. Imaging was split on a picture archiving and
communication system, randomised, then each folder blindly reported twice, separately, by two consultant neuro-radiologists. The
original report was used as gold standard and compared with all imaging. Scores on a pre-determined ranking system were recorded
by the radiologists using a four-point Likert scale (LS) for all scans. Intra-observer agreement was reviewed and SD was used as the
statistical evaluation.
Results or Findings: The FH IP showed adequate overall image quality and, on three sequences, less motion artefact with a score of
two or more points on the LS. In two cases on the T1 and fluid attenuation inversion recovery sequences the average score was less
than two (1.95; 1.98) when visualising the basal ganglia. Although these scores were less than the SH imaging, the overall image
quality was deemed to be sufficiently diagnostic across all points with a sensitivity of 82.7%, specificity of 76.5% and accuracy of
79.4%.
Conclusion: FH imaging shows pathology can confidently be identified despite a reduction in image quality and offers sufficient
diagnostic clinical information. FH IP imaging can become the new SH IP when investigating headaches in paediatric patients.
Multiparametric functional MRI quantitative assessment in children with methylmalonic acidemia (7 min)
Jiqing Song; Shandong / China
523
Author Block: J. Song; Shandong/CN
Purpose: The aim of this study was to describe the multiparametric functional MRI imaging features and to quantitatively evaluate
the macrostructural, microstructural and blood flow abnormalities of brain tissue in patients with methylmalonic aciduria (MMA).
Methods or Background: We enrolled sixty-two children with methylmalonic aciduria and another 62 age-matched and gender-
matched children as a control group. Multiparametric MRI including routine sequences, diffusion weighted imaging (DWI), diffusion
tensor imaging (DTI), susceptibility weighted imaging (SWI) and arterial spin labelling (ASL) were performed. A scoring system for
conventional structural MR imaging was developed to quantitatively evaluate the severity of brain macrostructural injury. Functional
MRI quantitative parameters, including ADC values, CBF values and FA values were acquired in eighteen ROIs. We compared the
parameters in both groups.
Results or Findings: A total of 47 children (75.8%) showed more than one structural abnormality on routine sequences. The highest
score was 9 in one patient. Cortical atrophy was the most common abnormality, followed by ventricular dilation. Basal ganglionic
abnormal signal was seen in 10 patients. Haemorrhage was detected in five patients. Compared to the control group, ADC values
increased in all eighteen ROIs, FA values decreased in widespread regions of white matter and CBF values decreased in anterior white
matter in cases younger than four years old in the patient group. Restricted diffusion was demonstrated on DWI in patients with acute
exacerbation.
Conclusion: It was more objective and reliable for scoring system to quantitatively evaluate the severity of brain macrostructural
injury. DWI may relate to acute exacerbation and stable episode of the lesions. Our study further confirmed the value of DTI for
quantitatively evaluating the brain white matter microstructural abnormality and initial demonstration of cerebral blood flow changes
on ASL in patients with MMA.
was obtained from the parents of all participants.
Reference values for knee ultrasound in children and adolescents: thickness of intercondylar cartilage, and length of
ossified patella (7 min)
Sílvia Costa Dias; Porto / Portugal
Author Block: S. C. Dias1, D. C. Carvalho1, M. Castro1, C. C. Dias1, I. M. Ramos1, I. Brito1, K. Rosendahl2; 1Porto/PT, 2Tromsø/NO
Purpose: The study aimed to establish ultrasound-based reference values for intercondylar cartilage thickness and patellar length by
age and sex.
Methods or Background: This was a prospective, cross-sectional study on healthy volunteers aged 3–17 years. Exclusion criteria
were those of chronic medicated diseases that affect the skeletal system or a recent trauma. Bilateral knee ultrasound was performed
by one of two experienced examiners, with the child supine, knee 90° flexed for measurement of intercondylar cartilage thickness
and extended for the patellar length. Interobserver variation on 27 participants was assessed.
Results or Findings: A total of 127 volunteers (67 females) with a median age of 10.9 years were included. Median thickness of the
intercondylar cartilage was 3.1 mm (percentiles 2.5th – 97.5th: 2.1- 4.2 mm) for females and 3.4 mm (percentiles 2.5th – 97.5th:
2.3-4.7 mm) for males (p=0.005). There were no differences according to right or left side (p=0.757). The cartilage thickness
decreased with age, with a median of 3.6 mm for 3–6-year-olds and 2.9 mm in 14-17 year-olds (p < 0.001). Median length of the
ossified patella was 33.6 mm (percentiles 2.5th – 97.5th: 0.0 – 44.2 mm) on the right side versus 32.4 mm (percentiles 2.5th – 97.5th:
0.8 – 42.2 mm) on the left side (p < 0.001). No differences were seen according to sex (p=0.259 for the right and p=0.233 for the left
side). The reproducibility between the two readers was high (intraclass correlation coefficient values > 0.9).
Conclusion: The intercondylar cartilage thickness decreased with increasing age, with females having thinner cartilage than males.
The length of the ossified right patella was significantly longer than the left. The presented reference values can help discriminate
between normality and pathology.
Ethics committee - additional information: The study registration number is: CE 191/22.
The diagnostic performance of magnetic resonance imaging in the categorisation of paediatric neck lymph nodes:
radiological and pathological correlations (7 min)
Naz Paytoncu; Istanbul / Turkey
524
Author Block: N. Paytoncu, E. Çalışkan, H. G. Düzkalır, M. Arifoğlu, N. Fıstıkçıoğlu, H. P. Günbey; Istanbul/TR
Purpose: In literature, there isn't a compact magnetic resonance imaging (MRI) study of the neck in the paediatric population
covering both numerical data and morphological criteria. Therefore, the aim of this study was to present certain MRI features of neck
lymph nodes in benign and malignant conditions in children.
Methods or Background: Contrast-enhanced MRI of the neck of 51 paediatric patients aged 1-18 years (40 boys, 11 girls
[10.08±4.73]) who underwent neck lymph node biopsy were retrospectively analysed. They were grouped as benign, including
reactive (27 [52.9%]), lymphadenitis (11 [21.6%]), and malignant (13 [25.5%]). The groups were evaluated multiparametrically in
terms of quantitative and qualitative variables.
Results or Findings: Medians and ranges (25-75th percentile) of long axis, short axis, area and apparent diffusion coefficient (ADC)
values of the largest lymph node were 21 (17-24) mm, 14 (12-18) mm, 228.60 (144.79-351.82) mm2, 2531 (2457-2714) mm2/s for
reactive, 24 (19-27) mm, 15 (11-20) mm, 271.80 (231.43-412.20) mm2, 2534 (2425-2594) mm2/s for lymphadenitis, 27 (23.50-31.50)
mm, 20 (15-22) mm, 377.08 (260.47-530.94) mm2, 2337 (2254-2466) mm2/s for malignant, respectively. The lymph nodes with a
long axis greater than 22 mm, a short axis greater than 16 mm, an area greater than 319 cm2, and an ADC value less than 2367
mm2/s may be malignant. Those with an atypical location (especially supraclavicular) have a higher rate of malignancy (p=0.003).
Perinodal signal changes, nodal heterogeneity with cystic/necrotic areas and posterior cervical triangle location are common in
lymphadenitis (p<0.001). Reactive lymph nodes are distributed symmetrically in both neck halves (p<0.001).
Conclusion: In an MRI-based approach to reactive, lymphadenitis and malignant lymph nodes for the paediatric era, not only
numerical data such as long axis, short axis, surface area and ADC, but also morphological criteria such as location, distribution,
internal structure and perinodal heterogeneity should be used.
Limitations: The study cohort could be expanded to include a greater number of participants. ADC measurements result from the
solid millimetric section of mostly necrotic lymph nodes, which might not be optimal.
Ethics committee - additional information: The ethics review board at our institution approved this study (decision number:
2023/514/244/16) which is performed in accordance with the principles of the Declaration of Helsinki.
Glymphatic system dysfunction and white matter connectivity alterations for diagnosing autism spectrum disorder in
children (7 min)
Miaoyan Wang; Wuxi / China
Author Block: M. Wang1, K. He2, D. Xu1, L. Wang1, G. Zhang3, B. Peng4, Y. Dai4, L. Zhang1, H. Jiang1; 1Wuxi/CN, 2Changchun/CN,
3
Shenzhen/CN, 4Suzhou/CN
Purpose: Multiparametric MRI analysis can help elucidate the pathogenesis of autism spectrum disorder (ASD). This study aimed to
analyse the glymphatic system and alterations of white matter connectivity for diagnosing ASD.
Methods or Background: In this retrospective case-control study, the data of children aged 3–6 years with ASD, and typically
developing (TD) children from two tertiary medical centres were collected. Patients with ASD were assigned to the mild-moderate,
and severe groups using the Childhood Autism Rating Scale. Automated diffusion tensor imaging along the perivascular space (aDTI-
ALPS) index and fractional anisotropy values were obtained on DTI. Node efficiency values were derived from the construction of
structural brain networks. A general linear model was used to evaluate the group differences in aDTI-ALPS index, fractional anisotropy
values, and brain network properties. Multiple MRI parameter features were evaluated for diagnostic efficacy using a machine
learning framework.
Results or Findings: Seventy children with ASD (mean age, 4.25 years; 51 male patients) and 45 TD children (mean age, 4.54
years; 25 male patients) were included in the study. The aDTI-ALPS index was lesser in the mild-moderate ASD group than that in the
control group (1.43 versus 1.58, p<.001). A further reduction in the aDTI-ALPS index was observed in the severe ASD group compared
with that in the mild-moderate ASD group (1.43 versus 1.33, p<.001). The accuracies of diagnosing ASD and distinguishing severe
ASD using aDTI-ALPS combined with extra-axial cerebrospinal fluid volume, white matter fractional anisotropy, and node efficiency
were 82.50% and 85.00% respectively, and area under the curve were 0.88 and 0.89, respectively.
Conclusion: The multiparametric predictive model based on glymphatic system dysfunction and alterations in white matter
connectivity enables the effective diagnosis and stratification of severity levels in autism spectrum disorder.
Funding for this study: Funding was provided by the Precision Medicine Key Project of Wuxi Health Commission (grant number:
J202107); Sanming Project of Medicine in Shenzhen (SZSM202011005); and Wuxi Science and Technology Development Project (CN)
(grant number: N20192005).
Microstructural white matter change, emotional dysfunction and visual working memory in adolescents with borderline
personality disorder (7 min)
Xiaoping Yi; Changsha / China
525
Author Block: X. Yi1, B. T. Chen2; 1Changsha/CN, 2Duarte, CA/US
Purpose: Emotional dysfunction is one of the core symptoms in adolescent borderline personality disorder (BPD), with the underlying
mechanisms remaining unclear. The aim of this study was to assess the alteration of microstructural white matter fibres and its
association with visual working memory and emotional dysfunction in adolescent BPD.
Methods or Background: A total of 53 adolescents with BPD aged 12–17 years and 39 age- and gender-matched healthy controls
(HCs) were enrolled into this study. Based on diffusion tensor imaging (DTI) data, radial diffusivity (RD) and axial diffusivity (AD) were
generated using Tract-Based Spatial Statistics (TBSS) method. Correlative analysis of microstructural alterations with visual working
memory, non-suicidal self-injurious behaviours (NSSI) and childhood trauma were performed.
Results or Findings: Compared with HCs, adolescents with BPD showed lower AD values in the splenium of the corpus callosum, left
anterior corona radiata, and left external capsule. In adolescents with BPD, higher RD values were observed in the genu of the corpus
callosum, body of the corpus callosum, right anterior corona radiata, and right uncinate fasciculus. There were significant correlations
between increased RD of genu and body of corpus callosum was negatively correlated with visual working memory (visual
reproduction), NSSI (Ottawa Self-Injury Inventory-4C), and childhood trauma (Childhood Trauma Questionnaire-E) (P<0.05).
Conclusion: There were brain microstructural alterations within the cortical-limbic system in adolescents with BPD, and these
changes were found to be associated with visual working memory, NSSI and childhood trauma in BPD. These results implicate that the
microstructural alterations may serve as a potential neuroimaging biomarker for underlying pathological mechanisms in adolescents
with BPD.
Limitations: The sample size was relatively small, and this was a cross-sectional study.
Ethics committee - additional information: The present study was approved by the ethics committee of our institute (IRB:
2022020227), and all participants and their legal guardians provided written informed consent.
Lesion volume and spike frequency impact perfusion in focal cortical dysplasia: a paediatric arterial spin labelling
study (7 min)
Antonio Giulio Gennari; Zurich / Switzerland
Author Block: A. G. Gennari, G. Bicciato, S. Lo Biundo, R. Kottke, I. Yakoub, D. Cserpan, R. Tuura O’Gorman, G. Ramantani; Zurich/CH
Purpose: Arterial spin labelling (ASL) has yielded promising results in the presurgical workup of children with FCD-related epilepsy.
Despite the increased utilisation of ASL, the interpretation of perfusion patterns and their correlations with other patient
characteristics remains unclear. Our study evaluated the perfusion changes captured by ASL in this vulnerable subgroup and
investigated their clinical, EEG, and MRI determinants.
Methods or Background: We included children with an MRI-detectable FCD, who underwent ASL. We assessed ASL perfusion
changes qualitatively by visual inspection and quantitatively by estimating the asymmetry index (AI). We correlated perfusion
patterns and their extent, as well as the AI values, with clinical, EEG, and MRI features.
Results or Findings: We considered 18 scans from 15 children with FCD-related epilepsy; seven underwent resective epilepsy
surgery. A total of 16 of 18 (89%) scans showed FCD-related perfusion changes: 10 FCDs were hypoperfused, whereas six were
hyperperfused. Nine scans had perfusion changes larger than, and seven equal to, the FCD extent on anatomical images.
Hyperperfusion in ASL was determined by frequent EEG spikes (p=0.047, Chi-square test). Perfusion changes in ASL larger than the
FCD corresponded to larger lesion volumes (p=0.017, Wilcoxon-Mann-Whitney test). Higher AI values were determined by frequent
EEG spikes (p=0.004, Welch t-test) and smaller lesion volumes (after controlling for age at MRI) in univariate analysis, but only
frequent EEG spikes retained their significance in multivariate analysis.
Conclusion: ASL showed FCD-related perfusion changes in most cases, including smaller volume lesions, which may escape
detection in anatomical MRI. Higher spike frequency may increase ASL yield in affected children. These observations may facilitate
the interpretation of ASL findings, improving treatment management, counselling, and prognostication in children with FCD-related
epilepsy.
Limitations: The main limitations are the retrospective design and the small sample size.
Funding for this study: We thank the Anna Mueller Grocholski Foundation and the Swiss National Science Foundation (SNSF:
208184) (to G.R.) and the Swiss Government Excellence Scholarship (to A.G.G.) for funding. The funders had no role in the design or
analysis of the study. None of the authors has any conflict of interest to disclose.
Ethics committee - additional information: The study was approved by and performed according to the guidelines and
regulations of the local ethics committee (KEK-ZH 2019-01854). All parents gave written informed general consent to reuse clinical
data for research.
Assessing the validity of MR imaging severity score for predicting the clinical outcome in acute necrotising
encephalopathy of childhood (7 min)
Kumail Khandwala; Karachi / Pakistan
526
Author Block: K. Khandwala, K. Hilal, S. Kaleem, M. Mufarrih; Karachi/PK
Purpose: Acute necrotising encephalopathy of childhood (ANEC) is a unique entity with bilateral grey and white matter involvement.
The purpose of this study is to determine whether severity of MR imaging findings can predict the prognosis and the clinical outcome
of patients with ANEC.
Methods or Background: A retrospective cross-sectional study was conducted on 42 patients diagnosed with ANEC. An MR imaging
severity score was devised for each patient according to a point system derived from the presence of haemorrhage, cavitation,
enhancement, diffusion restriction and location of lesions. The scoring was categorised into mild, moderate, and severe. Clinical
outcomes were determined at the time of discharge and at one-yearly follow-ups as mild disability, moderate disability, severe
disability, and death.
Results or Findings: The study included 21 boys and 21 girls with a mean age of 71.5 months. No statistically positive correlation (r
0.1198) was found between the MR grading and the clinical outcome. A statistically significant correlation of diffusion restriction (p
0.03) and cerebellar involvement (p 0.05) with worse clinical outcome was seen. Those with presence of shock also correlated with
worse outcomes (p 0.01).
Conclusion: Cerebellar involvement and presence of diffusion restriction on imaging, and presence of shock on presentation were
associated with a worse clinical outcome in our study. The MR imaging severity score overall, however, did not correlate significantly
with clinical outcome. Therefore, our results suggest the importance of combined clinical, laboratory, and neuroimaging findings in
determining the prognostic outcome of patients with ANEC rather than imaging severity alone.
Limitations: This was a single centre study with limited resources to conduct extensive virological studies, follow-up MRI or assess
RANBP2 target mutation for determining a genetic analysis. Our follow-ups were also limited because not all patients could be
clinically examined.
Ethics committee - additional information: The study was approved with approval number: 2020-5529-14796.
Bone involvement in children with lymphoma: results of a PET/MR study (7 min)

Lucia Pilati; Teolo / Italy
Author Block: L. Pilati1, G. Fichera1, S. Bertin1, M. Pillon1, D. Cecchin1, P. Zucchetta2, C. Giraudo1; 1Padua/IT, 2Treviso/IT
Purpose: The objective of this study was to assess the diagnostic value of 18F-FDG-PET/MR for bone involvement in pediatric
patients with lymphoma
Methods or Background: Children and adolescents (i.e., <21 years-old) with lymphoma who underwent a 18F-FDG-PET/MR at
diagnosis, from January 2017 to December 2022 were included. For each PET/MR site (skull, upper and lower extremities, and axial
skeleton) and number of skeletal lesions (up to five) were recorded with a separate evaluation of T1w, TIRM and PET datasets. PET/MR
was considered positive when a lesion was visible on T1W and/or TIRM and had high metabolic activity (higher than the mediastinal
blood pool). Biopsy or a combined reference standard, including other radiological techniques and follow-up were used. The diagnostic
value of T1W, TIRM, PET, and PET/MR was assessed by computing per-patient sensitivity (Se) and specificity (Sp).
Results or Findings: 59 patients (mean age 13.8±3.1years old; 32 females) matched the inclusion criteria. Overall, 11 (18.6%)
patients, eight with Hodgkin-lymphoma, had bone involvement. Overall, 60 lesions were detected with the lower extremities being
mostly affected (30 lesions), followed by the axial skeleton (16 lesions). The average size of the largest lesions was 26.5±16 mm. On
T1w images were identified five false positive and four false negative patients (Se=63.6%, Sp=89.6%) while with TIRM six patients
turned out to be false positive (Se=100%, Sp=87.5%). With PET and PET/MR occurred only one false negative (Se=90.91%,
Sp=100%).
Conclusion: PET/MR including TIRM allows an accurate characterization of bone involvement in children and adolescents with
lymphoma.
Pre- and postoperative foetal MRI for the evaluation of foetoscopic hybrid surgery in foetuses with spina bifida aperta
(7 min)
Maximilian Schulze; Marburg / Germany
527
Author Block: M. Schulze1, C. Keil1, I. Bedei2, V. Hohmann1, B. Saß1, S. Köhler1, R. Axt-Fliedner2, A. Kemmling1; 1Marburg/DE,
2
Giessen/DE
Purpose: Foetal MRI is important for the preoperative evaluation of spina bifida aperta (SBA) and its intracranial findings and is
therefore relevant for treatment decisions. Postoperative MRI enables intrauterine evaluation of the surgical result, also with regard to
the regression of ACM and possible further postpartum therapy requirements.
Methods or Background: SBA is a congenital malformation of multifactorial aetiology. The sensorimotor impairment of function
depends on the level of the spinal lesion. In addition, intracranial malformations such as Arnold Chiari malformation (ACM),
hydrocephalus, callosal disorders, and heterotopias are found. The MOMS trial demonstrated the superiority of intrauterine therapy
over postpartum therapy of SBA. The aim of the study is to evaluate the prenatal hybrid surgery (laparotomy with foetoscopic closure
of the SBA) performed at the centre.
From 2021- 2023, fifteen foetuses were operated on. Preoperative MRIs were performed in the 19th- 26th periods.
SSW and postoperative MRIs were performed in the 30th-32nd week of pregnancy according to standard protocol, T2 HASTE, DWI.
Results or Findings: The ratio of boys to girls was 6:9. 10 foetuses (66%) showed a meningomyelocele, 5 (33%) had rachischisis,
and lesion length was median of five segments (range 48). Motor levels were 4/15 foetuses (27%) at L5 and 11/15 foetuses (73%) at
S1. 100% had an ACM; 4 showed heterotopias, 3 had CC dysgenesis and 2 had syringomyelia.
Postop MRI showed complete closure of the SBA and regression of the ACM in all foetuses. 4/15 foetuses (27%) had a hydrocephalus,
which was treated postnatally by means of a VP shunt. The postop foetal MRI findings were confirmed by postpartum clinical
examination and postpartum MRI control.
Conclusion: Foetal MRI provides important morphological and functional information for the pre- and postoperative evaluation of
intrauterine hybrid surgery.
Limitations: The study is retrospective.
528
OF 13R - Optimising practice in interventional radiology
Categories: EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Interventional Radiology, Radiographers

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
This session will be dedicated to providing a comprehensive exploration of key aspects in the field of interventional radiology. With
four insightful talks, this session aims to enhance practice, improve safety, and elevate the quality of care in interventional radiology
procedures. Whether you are an experienced practitioner or new to the field, this session offers a wealth of knowledge and practical
strategies to elevate and optimise the quality of care, improve safety, and enhance patient outcomes in interventional radiology
procedures.
Moderator:
Silvia Svetlic; Milan / Italy

Bridging the gap: dose considerations in interventional radiology (12 min)

Lee O'Hora; Letterkenny / Ireland
How do we enhance safety in interventional imaging? (12 min)

Andrea Roletto; Milan / Italy
Top tips for radiographers performing paediatric interventional radiology procedures (12 min)
Emma Rose; London / United Kingdom
Strategies to enhance acute stroke patient pathways (12 min)

Lara Marie Deguara; St Pauls Bay / Malta
529
RPS 1304 - Advanced imaging of thoracic oncologic patients
Categories: Chest, Interventional Oncologic Radiology, Interventional Radiology, Oncologic Imaging

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Rok Cesar; Golnik / Slovenia
DEB-BACE followed by systemic chemotherapy vs systemic chemotherapy alone for advanced lung adenocarcinoma: a
propensity score match study (7 min)
Jianfei Tu; Lishui / China
Author Block: J. Tu; Lishui/CN

Purpose: This retrospective study aimed to investigate the effectiveness and safety of sequential bronchial arterial
chemoembolisation with drug-eluting beads (DEB-BACE) and chemotherapy versus systemic chemotherapy alone for advanced lung
adenocarcinoma progressing on targeted therapy.
Methods or Background: Stage III or IV lung adenocarcinoma patients in the chemotherapy group received an intravenous injection
of pemetrexed 500 mg/m2 plus cisplatin 75 mg/m2 on day 1 for 4 cycles, with each cycle lasting for 21 days. Patients in the DEB-
BACE plus chemotherapy group underwent DEB-BACE using CalliSpheres drug-eluting beads loaded with gemcitabine (400 mg) and
received via the microcatheter cisplatin 75 mg/m2 and gemcitabine 600 mg/m2 followed by intravenous chemotherapy 3 weeks post
DEB-BACE. The primary outcome was overall survival.
Results or Findings: Sixty-two patients who received chemotherapy and 69 who received DEB-BACE plus chemotherapy were
included, with 36 patients in each group after PSM. After PSM, the median OS was 18.3 months in the chemotherapy group and 33.1
months in the DEB-BACE plus chemotherapy group (P < 0.001). Multivariate Cox regression analysis showed that DEB-BACE plus
chemotherapy was associated with an 82% reduction in the risk of death versus chemotherapy only (P < 0.001). After PSM,
treatment-emergent adverse events of grade 3 or worse occurred in 2 of 36 patients in the DEB-BACE plus chemotherapy group and
16 of 36 patients in the chemotherapy alone group.
Conclusion: DEB-BACE plus chemotherapy improves the response rate and extends the survival of III-IV lung adenocarcinoma
patients progressing on targeted therapy. This offers this patient population, who otherwise have a rather dismal clinical outcome, an
effective and safe treatment option.
Limitations: DEB-BACE has a learning curve in terms of target tumour feeding vessel selection, the number of embolised vessels,
levels of embolisation and size of embolisation beads.
Funding for this study: Funding was received from the Medical and Health Science and Technology Plan of Zhejiang Province (grant
number WKJ-ZJ-1932).
Ethics committee - additional information: The study was approved by the Research Ethics Group of the Ethics Committee at
Lishui Central Hospital: ethical review of research (2022) number 373.
Tree-based models for predicting clinically significant pneumothorax in patients undergoing percutaneous coaxial core
lung biopsy: a retrospective cohort study (7 min)
Miguel Emilio Chevasco Hanze; Barcelona / Spain
530
Author Block: M. E. Chevasco Hanze, D. Castellon Plaza, S. A. Bolivar, B. Del Rio Carrero, H. H. J. Jofre; Barcelona/ES
Purpose: This study aimed to create a prediction model for the development of significant pneumothorax following a CT-guided
coaxial core lung biopsy (CT-CCLB) by employing machine learning tree-based models.
Methods or Background: A total of 469 patients who underwent CT-CCLB were retrospectively included. A list of 22 patient,
procedure and lesion characteristics were retrieved. Boruta analysis was used for selection of feature predictors. Afterwards, four tree
models, namely CART, AdaBoost, GB and XGBoost, were applied. The final model was chosen based on PPR, PLR and AUC values.
Final model and predictors behaviour were further evaluated by tree plot and SHAP analysis.
Results or Findings: Significant pneumothorax rate was 12.79%. GB classifier was found to have the best discriminating power
(AUC = 76.82%; PPR = 3.80; PLR = 5.08 43.48%). The top five predictors were lesion size/depth, DLCO, BMI and involvement of
fissures/bullae/emphysema during biopsy. Scenarios for encountering the highest significant pneumothorax occur when: 1) the
mentioned surfaces are compromised and the procedure is done with the patient in a supine or lateral body position; 2) lesion size is
lower than 22.37 mm and BMI is lower than 26.5; 3) lesion size is higher than 22.37 mm and depth is higher than 36.86 mm.
Conclusion: Significant pneumothorax after CT-CCLB was more likely to develop among patients with small lesions, lower BMI, higher
depth, and biopsies done in prone/lateral position and with fissures/bullae/emphysema. Machine-learning models demonstrated a high
predictive performance, with results being easy to visualise and read.
Limitations: Outweighing outcome was variable. Biopsies were made by radiology residents and attending radiologists
Ethics committee - additional information: No ethics committee approval was required because the present study was
retrospective, with no intervention throughout the research.
CT-guided core needle biopsy is safe and accurate for the assessment of pulmonary lesions associated with cystic
airspaces (7 min)
Maurizio Balbi; Monza / Italy
Author Block: M. Balbi1, N. C. Culasso1, M. Barba1, R. Senkeev1, S. Capelli2, A. Caroli2, F. Filipello3, L. Righi1, A. Veltri1; 1Orbassano/IT,
2
Bergamo/IT, 3Verduno/IT
Purpose: This study aimed to evaluate the safety and diagnostic capability of CT-guided core needle biopsy (CNB) in pulmonary
lesions associated with cystic airspaces (PLACAs).
Methods or Background: Consecutive pulmonary biopsies performed at the San Luigi Gonzaga Hospital (Orbassano, Italy) from
February 2010 to January 2022 (n=3069) were retrospectively reviewed to identify patients who underwent CNB for PLACAs (n=90,
case group; median age, 69.5 years, 95% confidence interval [CI], 62.0-75.0; 28 females). A group of CNB patients with non-cystic
lesions matched for age, sex, emphysema, and lesion depth and dimensions (n=180, control group) was selected to compare the
diagnostic yield and complication rate. The diagnostic performance for the final diagnosis was calculated. Univariate and multivariate
logistic regressions were performed in case patients to identify risk factors for complications and a non-diagnostic specimen (i.e.,
nonspecific benignity, atypical cells, insufficient specimen). PLACAs’ specimens were reviewed to assess histopathology.
Results or Findings: There were no significant differences between cases and controls in complication rate (overall: 40% versus
38%; major: 4% versus 6%, respectively) and non-diagnostic specimens (12% versus 9%). The diagnostic performance was similar in
both groups (accuracy: 97.78% vs. 97.78%, sensitivity: 97.53% vs. 97.63%, specificity: 100% vs. 100%). Among the patient,
procedural, and lesion-related data, the length of the needle pathway through the lung (odds ratio, [OR], 2.86; [95% CI, 1.08-7.80];
p=0.036) and the procedure time (OR, 10.93; 95% CI, 3.77-35.85; p < 0.001) were significant risk factors for complications. No
variables predicted a non-diagnostic specimen. In most cases, PLACAs were adenocarcinoma (54%), and the cystic airspaces
corresponded to tumour cystification (22 out of 31 resected specimens, 71%).
Conclusion: CT-guided CNB was safe and effective for assessing PLACAs. A long needle pathway and procedure time increased the
complication risk.
Limitations: Identified limitations were (1) that this was a single-centre study and (2) the limited number of cases.
Funding for this study: The authors state that this work has not received any funding.
Ethics committee - additional information: The ethics committee approved the present study and waived the need for written
informed consent.
Haemorrhage risk prediction after computed tomography-guided lung biopsy: combining clinical parameters and
quantitative pulmonary vascular analysis (7 min)
Keng-Chian Lin; Taipei / Taiwan, Chinese Taipei
531
Author Block: K-C. Lin, Y-S. Huang, Y-C. Chang; Taipei/TW
Purpose: This study aimed to evaluate the utility of combining quantitative pulmonary vasculature measures with clinical factors for
predicting pulmonary haemorrhage after computed tomography (CT)-guided lung biopsy.
Methods or Background: Patients who underwent CT-guided lung biopsy were retrospectively included in this study. Clinical and
radiographic variables were evaluated as predictors of pulmonary haemorrhage. The radiographic pulmonary vascular analysis
included vessel count, vessel density, vessel diameter, vessel area, blood volume in small vessels with a cross-sectional area ≤ 5
mm2 (BV5), and total blood vessel volume (TBV) in the lungs. Univariate and multivariate logistic regressions were used to identify
the independent risk factors of higher-grade pulmonary haemorrhage and establish the prediction model, which was presented in the
form of a nomogram.
Results or Findings: A total of 126 patients was included (discovery cohort n=103, validation cohort n=23). Any pulmonary
haemorrhage, higher-grade (grade ≥2) pulmonary haemorrhage, and hemoptysis occurred in 42.9%, 15.9%, and 3.2% of patients
who underwent CT-guided lung biopsies. In the discovery cohort, patients with greater lesion depth (p=0.013), higher vessel density
(p=0.033), and higher BV5 (p=0.039) were more likely to experience higher-grade haemorrhage. The nomogram prediction model for
higher-grade haemorrhage built by the discovery cohort showed similar performance in the validation cohort.
Conclusion: Higher-grade pulmonary haemorrhage may occur after CT-guided lung biopsy. Lesion depth, vessel density, and BV5 are
independent risk factors for higher-grade pulmonary haemorrhage. Nomograms that integrate both clinical parameters and
radiographic pulmonary vasculature measures offer enhanced capability for the assessment of haemorrhage risk following CT-guided
lung biopsy, thereby facilitating improved clinical care for patients.
Ethics committee - additional information: This study was approved by the Institutional Research Ethics Committee: approval
number 202306051RIN.
De novo low-dose CT-guided lung biopsy technique: minimising radiation with maintained safety and diagnostic yield
rates (7 min)
Avik Banerjee; Leicester / United Kingdom
Author Block: S. Vijayakumar1, A. Banerjee2, G. Tsaknis2; 1Leicester/UK, 2Kettering/UK

Purpose: The purpose of this study is to present our innovative approach in developing a de novo low-dose CT-guided lung biopsy
technique, aiming to significantly reduce radiation exposure (DLP less than 100 mGy-cm, only slightly more than CT fluoroscopy)
without compromising histopathological accuracy or increasing complication rates.
Methods or Background: We conducted a retrospective analysis of lung biopsy procedures performed using our novel low-dose CT-
guided biopsy technique. A total of 100 patients with suspected lung lesions were included. The procedure involved precise planning
and real-time image guidance, striking a balance between kVp and mAs to ensure that diagnostic image quality is maintained without
unnecessary radiation exposure to the patient. Patient demographics, lesion characteristics, procedural details, radiation dose,
histopathological findings, and post-procedural complications were meticulously recorded and analysed.
Results or Findings: The detailed findings are still in process, but some of the highlights include: 1) over 90-95% of cases had DLP
below 100 mGy·cm, with almost all cases under 150 mGy·cm; and 2) varied kVp and mAs permutations were explored. Our protocol
utilised 100kVp and 50mA, dropped to 30mA during biopsy, utilising precise planning and positions for challenging lung lesions,
effectively reducing radiation doses.
Conclusion: Our study introduces a novel low-dose CT-guided lung biopsy technique, reducing radiation exposure to DLP
<100mGy·cm while preserving high histopathological accuracy and patient safety. This approach is especially significant for patients
needing recurrent scans and biopsies. Its effectiveness positions it as a potential standard, enhancing healthcare quality.
Limitations: For patients with very high BMI, increased radiation dose was necessary to target the lesion, usually higher than our
target of 100kVp.
Dual-energy CT-based radiomic analysis for predicting pathological grading of lung invasive adenocarcinoma (7 min)
Yuting Zheng; Wuhan / China
532
Author Block: Y. Zheng; Wuhan/CN
Purpose: To investigate the value of radiomics based on dual-energy CT (DECT) for predicting pathological grading of lung invasive
adenocarcinoma.
Methods or Background: In this retrospective study, a total of 107 patients (80 G low and 27 G high) with lung adenocarcinoma
before surgery were included. Clinical, radiographic features and quantitative parameters were recorded for the clinical-DECT model.
The DECT-based radiomics model was constructed from features extracted from virtual monoenergetic images (VMI), including 50 kev
and 150 kev images. The DECT radiomics, clinical-DECT, and conventional CT radiomics models were established.
Results or Findings: For predicting lung invasive adenocarcinoma grades, the DECT radiomics model achieved excellent
performance with an area under curve (AUC) of 0.997 and 0.743 in the training and test set, respectively. Tumour density, lobulation,
and effective atomic number at AP were included in the clinical-DECT model with an AUC of 0.836 in the training set, lower than the
DECT radiomics model. In comparison to the conventional CT radiomics model (AUC, with 0.998 and 0.529 in the training and test
set), the DECT radiomics model demonstrated a higher AUC value and a better net benefit to the patients in the test cohorts.
Conclusion: DECT-based radiomics features were useful in predicting pathological grading of lung invasive adenocarcinoma, yielding
better predictive performance than the clinical-DECT and conventional CT radiomics models.
Limitations: Firstly, this was a single-centre retrospective study, and the sample size was relatively small, which may constrain the
generalisability of our findings. Secondly, as the patients were examined from 2021 to 2022 with a limited follow-up time, the
influence of tumour differentiation on patient outcomes was not assessed.
Ethics committee - additional information: This study was approved by the Ethics Committee of Tongji Medical College.
Adherence to CT surveillance guidelines in early stage post-treatment lung cancer recurrence (7 min)
Sian Kneafsey; Dublin / Ireland
Author Block: S. Kneafsey, D. P. Garrahy, D. Byrne, P. Beddy; Dublin/IE

Purpose: Detection of lung cancer recurrence in patients post treatment can result in survival benefit, and enable early detection of
new primary lung cancer. However, adherence to post-treatment surveillance protocols is highly variable. We assessed the adherence
to CT surveillance guidelines in all patients treated with curative resection for non-small cell lung cancer (NSCLC) at our tertiary
referral lung cancer centre.
Methods or Background: All patients treated with curative intent surgery are followed up in a nurse led outpatient clinic. Patient
demographics, postoperative imaging studies, histopathological results and follow-up appointment schedule were recorded. We
defined the minimal standard of surveillance imaging studies (MSSIS) as ≥5 CT studies in the first 5 years as specified in our follow-up
guidelines, based on the ESMO guidelines.
Results or Findings: A total of 1243 patients were included. The mean age was 66 ± 9.7. Of these, 579 (46.6%) were male, 665
(53.4%) were female. In total, 787 (63.3%) achieved the minimal standard of surveillance imaging studies directly within our centre,
and are undergoing ongoing active surveillance. A further 170 (13.6%) patients achieved the minimal standard of surveillance
imaging studies and were discharged from the surveillance service at five years post-surgery. A further 127 (10.2%) patients were
discharged to surveillance services in satellite referral hospitals. A total of 149 (12%) patients achieved MSSIS but have since died.
Five (0.4%) patients achieved MSSIS but are no longer under surveillance due to palliation. Five (0.4%) patients are no longer under
active surveillance due to unknown reasons.
Conclusion: A total of 1111 (89.3%) patients achieved MSSIS at our institution during this surveillance period. Over two-thirds
(63.6%) of patients are undergoing active surveillance. Existence of a dedicated nurse-lead lung cancer surveillance outpatient clinic
had led to higher surveillance adherence rates than international literature reports.
Ethics committee - additional information: This study was approved by St. James Hospital Research and Innovation Office:
project 8410.
Coronary calcification and interstitial lung disease are both independently associated with increased mortality in
patients undergoing radiotherapy for stage 3 non-small cell lung cancer (7 min)
Emily Hughes; Glasgow / United Kingdom
533
Author Block: C. P. McKeag1, E. Hughes1, E. McGarry1, S. Ghatorae2, G. Cowell1, J. Maclay1; 1Glasgow/UK, 2Larbert/UK
Purpose: Patients diagnosed with lung cancer are often multi-morbid. Assessment of comorbidities is possible on diagnostic CT and
may influence survival.
Methods or Background: We looked at overall survival outcomes in patients diagnosed with stage 3 non-small cell lung cancer
undergoing radical radiotherapy (+/- systemic treatment) in the West of Scotland between 2017-2020. Mortality was right censored at
two years, and the cohort consisted of 431 individuals. This was an update to a previous review of outcomes 2017-2019.
We reviewed the diagnostic CT to identify common comorbidities including emphysema, coronary artery calcification, interstitial lung
disease, and pleural effusion. Each of these was individually assessed for any impact on survival, and Kaplan-Meier curves were
generated, with the aim of identifying potential markers of increased mortality.
Results or Findings: We showed a significant increase in overall mortality for patients with interstitial lung disease (ILD) (p<0.005),
and for patients with severe coronary artery calcification compared to those with mild or less calcification (p<0.005). There was no
statistically significant difference in mortality between moderate and mild or less coronary calcification (p=0.13), nor between
moderate and severe (p=0.23).
There was no statistically significant difference in mortality in patients with or without pleural effusion (p=0.37), nor between severe
emphysema and mild or no emphysema (p=0.49).
Conclusion: Severe coronary artery disease and ILD are associated with reduced overall survival in patients undergoing radical
radiotherapy for NSCLC. Further studies investigating cardiovascular complications of radiotherapy are required.
Radiogenomics relationship of non-small cell lung cancer: preliminary results (7 min)

Maria Paola Belfiore; Naples / Italy
Author Block: M. P. Belfiore, M. Sansone, V. Patanè, R. Monti, F. Grassi, G. Ciani, R. Grassi, S. Cappabianca; Naples/IT
Purpose: The aim of this study was to correlate the radiomics features with the genetic results obtained from liquid biopsy in
patients with lung tumours.
Methods or Background: We included 53 patients suffering from NSCLC who underwent pre-surgery CT (GE Revolution 128 MDCT)
at the Radiology Department of the Campania University. Every patient performed liquid biopsy subject to informed consent for the
genetic analysis. For the radiomic analysis, image processing CT volumes were manually delineated using ITK-snap 3.8.0. Radiomics
features (first order, GLCM, GLRLM, GLSZM, GLDM, NGTDM) were computed using Pyradiomics in Python 3.7 environment. For the
statistical analysis, association between radiomic features and gene mutations were assessed using feature importance based on ROC
analysis; moreover, a machine learning approach based on SVM was used to evaluate the ability of radiomic features to predict gene
mutations.
Computations have been performed in the R environment using the CARET package.
Results or Findings: From the genetic analysis it turns out that the accuracy, i.e. the number of correct predictions and the total
number of patients, obtainable using the selected group of features is of the order of 0.67. Some correlations between gene and
features were found to be the case: ROS.miss.6.43.Arg167Gln with a feature group that included first-order glcm and glszm, ROS-
miss.42.43.Asp2213 Asn with a broader set of features and ALK.miss.29.29.Asp1529 Glu with the same previous group but less
intense.
Conclusion: Radiomics could better determine the accuracy of malignancy of pulmonary nodules, which have been detected by CT
scan in order to treat curatively, select patients with early-stage lung cancer who are appropriate for post-surgical treatment, and
determine patients with stage III NSCLC who can tolerate immunotherapy as consolidation therapy after concurrent treatment with
chemotherapy-radiation therapy.
Limitations: These results are preliminary and require a greater number of observations.
A decade of image-guided pleural biopsies: a multicentre retrospective study examining diagnostic yield and
complications (7 min)
Liam Peng; Glasgow / United Kingdom
534
Author Block: L. Peng, S. Tsim, K. Blyth, G. Cowell; Glasgow/UK
Purpose: Pleural disease is common, with diagnostic options including pleural effusion aspiration, local anaesthetic thoracoscopy
(LAT) and image guided biopsy. Despite diagnostic and therapeutic applications, LAT availability varies and is unfeasible with a
complex pleural space. Ultrasound (US-) and Computed Tomography (CT-) guided biopsy remain commonly performed, yet post-
procedural data is relatively sparse. A high prevalence of malignant pleural mesothelioma (MPM) in the West of Scotland offers an
excellent opportunity to assess complication rate and diagnostic yield associated with image guided biopsy relative to other
investigation strategies, to shape the consent process, develop care models and aid decision making around timely MPM diagnosis.
Methods or Background: US- and CT-guided pleural biopsy procedures performed in six hospitals between 01.01.13 and 31.03.23
were identified by searching study codes and a key word search. Patient demographics, biopsy pathology, final diagnosis (either by
repeat biopsy or consensus diagnosis via regional/national multidisciplinary team meeting) and procedural complications were
recorded.
Results or Findings: 194 CT-guided pleural biopsies were performed using a coaxial technique. Seven (3.6%) had a peri-procedural
pneumothorax on CT, with two (1.0%) evident on post-procedural chest radiograph. None required intervention. Two cases (1.0%) had
haemoptysis. For diagnosis of pleural malignancy, sensitivity of CT-guided biopsy was 93.9% (95% CI 88.9-97.0%), specificity 100%
(89.1-100%), positive predictive value (PPV) 100% (97.6-100%) and negative predictive value (NPV) 76.2 (63.7-85.4%).
A total of 79 US-guided pleural biopsies were performed, without post-procedural pneumothorax. For diagnosis of pleural malignancy,
sensitivity of US-guided biopsy was 93.1% (95% CI 84.5-97.7%), specificity 100% (59.0-100%), PPV 100% (94.6-100%) and NPV 58.3%
(37.5-76.5%).
Conclusion: CT- and US-guided pleural biopsy is safe, offering excellent sensitivity and PPV in the diagnosis of pleural malignancy if
LAT is not available or feasible.
Ethics committee - additional information: This study received Caldicott approval granted by the NHS board.
Structured reporting quality of chest CT for lung cancer staging: a double cohort study involving radiology residents (7
min)
Valeria Peruzzi; Udine / Italy
Author Block: V. Peruzzi, L. Cereser, C. Vecchia, F. Cortiula, M. Bortolot, G. Como, R. Girometti, C. Zuiani; Udine/IT
Purpose: To compare radiology residents' (RRs) report quality using the structured radiological model (SRM) from the Royal College
of Radiologists with narrative reporting (NR) for chest CT staging of lung cancer. To assess reporting times for NR and SRM.
Methods or Background: A study coordinator preliminarily selected 30 non-small cell lung cancer patients who underwent a
baseline staging contrast-enhanced chest CT examination between 2014 and 2022 at our University Hospital. After attending a
dedicated training lesson, four third-year RRs (RR1-4) independently reported all the CT examinations in two 2-month-apart separate
reading sessions. In the first reading, all the RRs used the NR, while in the second reading, RR1-2 used the NR, and RR3-4 used SRM.
Two chest-devoted radiologists, in consensus, rated the completeness and accuracy of all the NRs and SRMs. Two thoracic
oncologists, in consensus, expressed the perceived clarity for the reports from the second reading session. All the quality indicators
were expressed on a 100-point scale. The Wilcoxon test was used for statistical analysis.
Results or Findings: Comparing reading sessions, RR3-4 report completeness was significantly higher when using SRM vs. NR (90.7
vs. 74.0, p<0.001), while RR3-4 accuracy and RR1-2 completeness and accuracy values were not significantly different. In the second
reading, report completeness, accuracy, and clarity of RR3-4 were significantly higher than RR1-2, with median values of 90.7 vs. 72.8
(p<0.001), 63.1 vs. 58.7 (p=0.04), and 87.3 vs. 68.3 (p<0.001), respectively. Median RR3-4 reporting time was significantly longer
than RR1-2 (13.5 min vs. 10.6 min, p<0.001).
Conclusion: The completeness, accuracy, and clarity of SRM were superior to NR at the price of a longer reporting time.
Limitations: Retrospective, monocentric study, with a limited number of patients.
Funding for this study: None
535
RPS 1305 - Strategic deployment of AI
Categories: Artificial Intelligence & Machine Learning, Medico-legal, Research

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Early platform release of the federated European cancer imaging infrastructure (7 min)
Ignacio Blanquer; Valencia / Spain
Author Block: A. S. Alic1, D. Arce Grilo1, M. Birhanu2, E. Bron2, V. Kalokyri3, T. Kussel4, K. Lang5, K. Majcen5, I. Blanquer1; 1Valencia/ES,
2
Rotterdam/NL, 3Heraklion/GR, 4Heidelberg/DE, 5Graz/AT
Purpose: EUCAIM (https://cancerimage.eu/) is a pan-European federated infrastructure for cancer images, fueling AI innovations.
Methods or Background: This federated infrastructure is built upon a set of core services that comprise a public metadata
catalogue, a federated search service following a common hyperontology, an access negotiation system, a coherent AAI and a
distributed processing service. EUCAIM has recently released an early prototype with 40 image datasets from nine cancer types
(breast, colon, lung, prostate, rectum, liver, diffuse intrinsic pontine glioma, neuroblastoma, glioblastoma) registered, related to the
five projects in the AI4HI network (EUCANIMAGE, ProCAncer-I, INCISIVE, CHAIMELEON and PRIMAGE - https://future-ai.eu/), for a total
of more than 200,000 image series from approximately 20,000 individuals. These collections follow a common metadata model
defined in the EUCAIM project.
Results or Findings: This early prototype comprises a dashboard with guiding instructions, a public catalogue, a federated search
engine and an access negotiation system in beta version.
Conclusion: This platform will permit users to discover, search, request, access and process medical imaging and associated clinical
data in a flexible manner, supporting federated providers with different access levels and a future centralised repository. EUCAIM is
based on cloud and container technologies, and it will be linked to intensive computing infrastructures such as EGI and
supercomputing centres.
Limitations: The access negotiation service is currently in beta version and access requests will be forwarded to the providers.
Funding for this study: This project is co-funded by the European Union under grant agreement 101100633
Ethics committee - additional information: The project acts as a broker for accessing data and relies on the ethical approvals of
the providers and requesters.
Radiology AI deployment and assessment rubric (RADAR) for value-based AI in radiology (7 min)
Jacob Johannes Visser; Rotterdam / Netherlands
536
Author Block: B-J. Boverhof1, K. Redekop1, D. Bos1, M. P. A. Starmans1, J. Birch2, A. G. Rockall3, J. J. Visser1; 1Rotterdam/NL, 2Poole/UK,
3
Godalming/UK
Purpose: The aim is to provide a comprehensive framework for value assessment of AI for radiology.
Methods or Background: This paper presents the RADAR framework, which has been adapted from Fryback and Thornbury's
imaging efficacy framework and facilitates valuation of radiology artificial intelligence (AI) from conception to local implementation.
Special attention is placed on local efficacy to underscore the importance of appraising an AI system in its local environment. The
RADAR framework is illustrated through a myriad of study designs that help conduct adequate valuation.
Results or Findings: The RADAR approach constitutes a seven-levelled-hierarchy, providing radiologists, researchers, and decision-
makers with a conceptual framework for comprehensive AI valuation in radiology. RADAR is dynamic, catering to varying valuation
throughout the AI's developmental cycle. Technical and diagnostic efficacy (RADAR-1 and RADAR-2) is assessed before clinical
implementation and can be addressed by in-silico clinical trials and cross-sectional studies. The next phases, encompassing diagnostic
thinking to patient outcome efficacy (RADAR-3 to RADAR-5) necessitate clinical integration and can be addressed through randomised
controlled trials and cohort studies. Societal efficacy (RADAR-6) delves into broader societal implications, assessed through health-
economic evaluations. Concluding the hierarchy, the extent to which previous assessments generalise locally (RADAR-7) are gauged
with budget impact analysis and multi-criteria decision analysis.
Conclusion: The RADAR framework stands as a comprehensive solution for valuing radiology. With its progressive and hierarchical
approach, as well as an emphasis on local efficacy, RADAR provides a comprehensive framework to illustrate radiology AI's value
conform to the notion of value-based radiology.
Knowledge of AI governance, perceived challenges, opportunities, and suggestions for AI implementation by UK

radiographers (7 min)
Author Block: N. Stogiannos1, T. J. T. O'Regan1, M. Pogose1, H. Harvey1, A. Kumar1, R. Malik2, A. Barnes1, M. F. F. McEntee3, C.
Malamateniou1; 1London/UK, 2Farnworth/UK, 3Cork/IE
Purpose: Radiographers are key stakeholders in AI use for clinical imaging and radiation therapy. AI Implementation is key to
harness the potential benefits of AI innovation. Knowledge of AI governance by all healthcare professionals is vital for AI
implementation in clinical practice. This study aims to explore UK radiographers’ knowledge and perceptions on AI governance.
Methods or Background: An online survey on Qualtrics was distributed to UK-practicing radiographers via social media. Eligible
respondents needed to have theoretical knowledge and/or practical expertise in the use of AI in medical imaging and/or radiation
therapy. Descriptive and inferential statistics was used to analyse quantitative data and content analysis for open-ended questions.
Results or Findings: There were 88 valid responses. Lack of training, guidance, and funding are the most important challenges to AI
implementation, as perceived by radiographers. Many radiographers (36.9%) were unaware of evaluation methods for AI tools, whilst
56.6% hadn’t received any AI-specific training. Robust governance frameworks (30.7%), customised training (27.3%), and patient and
public involvement (21.6%) were noted as strategic priorities by respondents.
Conclusion: Effective leadership, allocated time, and tailored training will contribute to successful AI implementation. Further
research is needed to ensure radiographers can harness the benefits and minimise risks of AI.
Limitations: Selection bias might have occurred in this study, since data was collected online. Also, the skewed geographical
distribution of the respondents may further limit the generalisability of the results.
Funding for this study: This study received funding from the College of Radiographers CORIPS grant scheme (grant number: 209)
and the City Radiography Research Fund.
Ethics committee - additional information: This study was approved by the City, University of London School of Health and
Psychological Sciences Research Ethics Committee (reference: ETH2122-1015).
Black box no more: a survey to explore AI adoption and governance in medical imaging and radiation therapy in the UK
(7 min)
537
Author Block: N. Stogiannos1, T. J. T. O'Regan1, A. Barnes1, A. Kumar1, R. Malik2, M. Pogose1, H. Harvey3, M. F. F. McEntee4, C.
Malamateniou1; 1London/UK, 2Farnworth/UK, 3Banstead/UK, 4Cork/IE
Purpose: The clinical use of AI tools in medical imaging and radiation therapy (MIRT) has highlighted challenges to AI adoption and
governance for healthcare professionals. This study aims to map the perceived challenges around clinical adoption of AI.
Opportunities associated with AI and suggestions for future implementation are explored.
Methods or Background: A multidisciplinary online survey on Qualtrics® was designed using expert focus groups and published
literature and piloted (n=9) before distribution. It was shared via social media and professional networks to all MIRT professionals in
the UK. Data was analysed using descriptive and inferential statistics on the SPSS software, whilst content analysis was employed for
the open-ended questions.
Results or Findings: A total of 245 valid responses were received from different MIRT professionals. Lack of knowledge of AI
governance frameworks was noted (42.1%). Prior AI training was significantly correlated with understanding of AI governance
concepts (p=0.007 for MHRA and 0.001 for ISO standards). Respondents indicated that clear governance frameworks (11.4%), AI
training (9%) and effective leadership (8.5%) are vital for successful AI adoption.
Conclusion: Knowledge of, and confidence in AI technologies correlate with prior AI-related training. Different professionals were
familiar with frameworks related to their practice. Tailored AI training is needed to address knowledge gaps for a safe and successful
AI adoption in medical imaging and radiation therapy in the UK.
Limitations: The small sample size of this study means results cannot be generalised to the broader UK medical imaging and
radiation therapy AI ecosystem.
Funding for this study: This study received funding from the College of Radiographers CORIPS grant scheme (grant number: 209)
and the City Radiography Research Fund.
Ethics committee - additional information: This study was approved by the City, University of London School of Health and
Psychological Sciences Research Ethics Committee (reference: ETH2122-1015).
Radiographer education and learning in artificial intelligence (REAL-AI) (7 min)

Geraldine Doherty; Belfast / United Kingdom
Author Block: G. Doherty1, L. McLaughlin1, R. Bond1, J. McConnell2, C. Hughes1, S. L. McFadden1; 1Belfast/UK, 2Leeds/UK
Purpose: Artificial intelligence (AI) is widespread in medical imaging, yet there is a paucity of information on education and training
available for staff. Further research is required to identify what training is available, and what preparations are required to bring AI
knowledge to levels that will enable radiographers to work competently alongside AI. This study aimed to: a) investigate current
provision of AI education at UK higher education institutes (HEIs); b) explore the attitudes and opinions of educators.
Methods or Background: Data were collected through two online surveys: 1) UK HEIs; 2) medical imaging educators. The surveys
were distributed in the UK by the heads of radiography education (HRE), The Society of Radiographers and as part of the Research
Hub at ECR 2023. The study was promoted on LinkedIn and Twitter (X), and through university channels.
Results or Findings: Responses were received from 22 HEIs in the UK and 33 educators from across Europe. Data analysis is
ongoing, but preliminary findings show that 68.2% (n=15) of responding HEIs claim to have introduced AI into the curriculum already.
84.8% (n=28) of educators claim they themselves have received no training on AI despite having to embed it into the curriculum. The
main reason for this, as cited by HEIs, is limited resources. 69.7% (n=23) of educators believe that AI concepts should be taught by an
AI expert.
Conclusion: By surveying educators and HEIs separately, this study captured two different perspectives regarding the provision of AI
education. This unique insight highlighted disharmony between HEIs and educators. Preliminary insights highlight that educators feel
unprepared to deliver AI content, and HEIs are under pressure to add AI concepts to an already full curriculum.
Limitations: An identified limitation was that surveys, focus groups and interviews were conducted in the English language only.
Funding for this study: This project has been part-funded by a College of Radiographers Industry Partnership Scheme, grant
number 229 (AI).
Ethics committee - additional information: This study was approved by the Ulster University Filter Committee. Reference
numbers: FCNUR-23-051 / FCNUR-23-006-A.
International medical students' perceptions towards artificial intelligence in medicine: a multicentre, cross-sectional
survey among 192 universities (7 min)
538
Author Block: F. Busch1, L. Hoffmann1, D. Truhn2, M. Makowski3, K. K. Bressem1, L. C. Adams3; 1Berlin/DE, 2Aachen/DE, 3Munich/DE
Purpose: Artificial intelligence (AI) is set to fundamentally change the educational and professional landscape for the next generation
of physicians worldwide. This study aimed to explore the current international attitude of medical students towards AI in the medical
curriculum and profession on a large, global scale and identify factors that shape their attitudes.
Methods or Background: This multicentre, multinational cross-sectional study developed and validated an anonymous online
survey of 15 multiple-choice items to assess medical, dentistry, and veterinary students' preferences for AI events in the medical
curriculum, the current state of AI education, and students' AI knowledge and attitudes towards using AI in the medical profession.
Subgroup analyses were performed considering gender, age, study year, tech-savviness, prior AI knowledge and AI events in the
curriculum, and university location.
Results or Findings: Between April and October 2023, a total of 4,313 medical, 205 dentistry, and 78 veterinary students from 192
faculties and 48 countries responded to the survey. Most participants came from European countries (n=2,350), followed by
North/South America (n=1,070) and Asia (n=944). Students showed predominantly positive attitudes towards AI in medicine (67.6%,
n=3,091) and expressed a strong desire for more AI education (76.1%, n=3,474). However, they reported limited general knowledge
of AI (75.3%, n=3,451) and felt inadequately prepared to use AI in their future careers (57.9%, n=2,652). Subgroup analyses revealed
differences in attitudes between students from the Global South and North and on the continental level, among others.
Conclusion: This large-scale international study underlines the generally positive attitude of medical students towards the
application of medical AI and explores variables that influence such attitudes. Our study highlights the necessity for a greater
emphasis on AI education within medical curricula.
Limitations: The unequal regional representation and selection bias were identified as limitations.
Funding for this study: The authors report the results on behalf of the COMFORT consortium, an initiative of the Horizon Europe-
funded COMFORT project (101079894).
Ethics committee - additional information: This study was approved by the IRB, with approval code: EA4/213/22.
AI in routine teleradiology use: results of a large-scale test across Germany and Austria (7 min)
Torsten Bert Thomas Moeller; Dillingen / Germany
Author Block: T. B. T. Moeller, P. F. W. Sögner; Dillingen/DE

Purpose: The objective of this study was to answer the question of whether the use of AI is already having a quality-improving effect
in routine teleradiological reporting throughout Germany and Austria.
Methods or Background: We performed a study of 2,707 native CCT scans from the CT departments of 140 hospitals in Germany
and Austria between March and April 2022, which were analysed using AI with haemorrhage analysis. The results were compared with
the findings of more than 70 teleradiologists who did not have the AI results at that time. Possible discrepant findings were evaluated
by two radiologists with specific neuroradiological CCT experience.
Additionally, the in-house error statistics from 2021 and 2022-23 were reviewed.
Results or Findings: Of the 2,707 CCT examined by both radiologists and AI, 189 cases (approximately seven percent) were found
to have intracranial haemorrhage described by both radiologists and AI. In 30 patients there was a discrepancy: the AI had seen a
haemorrhage that had not been described by the radiologist. These cases were subsequently re-evaluated. Twelve (approximately
40%) of the 30 unclear examinations were classified as false positives by the AI, eight cases as questionable positives, and 10 cases
as true positives. Thus, there were 199 cases with ICB in the studied patient group, of which > 5% were primarily missed by
radiologists without AI support.
A review of in-house error statistics also revealed a significant decrease in reported false findings for intracranial haemorrhage (from
16 in 2021 to one between 08/31/2022 and 08/31/2023).
Conclusion: The positive effects of AI on the quality of radiological reporting postulated in several studies can also be confirmed in
practice and especially in the teleradiological context.
Limitations: This assumption should be substantiated by further studies.
Artificial intelligence should only read a mammogram when it is certain: a hybrid breast cancer screening reading
strategy (7 min)
Sarah Delaja Verboom; Nijmegen / Netherlands
539
Author Block: S. D. Verboom, J. Kroes, S. Pires, M. Broeders, I. Sechopoulos; Nijmegen/NL
Purpose: The aim of this study was to incorporate and evaluate uncertainty quantification metrics in an artificial intelligence (AI)
breast cancer detection model and test their ability to guide a novel hybrid reading strategy in breast cancer screening in which recall
decisions are only made by standalone AI when it exhibits high certainty.
Methods or Background: Uncertainty quantification metrics were obtained from a modified version of a commercial AI breast
cancer detection model by structured Monte Carlo dropout. The metrics were defined as the variance or entropy of one or all
suspicious regions and used to estimate the certainty of the AI malignancy-present decision. With the proposed hybrid reading
strategy, the recall decision is based on AI only when the predictions are classified as certain, and by standard radiologist-double
reading otherwise. The new reading strategy was retrospectively tested on a previously-unseen subset of all digital mammographic
screening examinations acquired between 2003-2018 from a unit of the Dutch National Breast Cancer Screening (n=41,469) with
minimal 2-year follow-up.
Results or Findings: The best-performing uncertainty metric was the entropy of the mean output for the most suspicious region per
case. The hybrid reading strategy using this uncertainty metric and a recall rate equal to the standard radiologist-double-reading
strategy (27 per 1000) resulted in 46% of cases read by AI only and a cancer detection rate of 8.1 per 1000, which does not differ
from the standard strategy (8.0 per 1000, p=0.217). The mean AUC of the AI model increased from 0.957 (95% CI 0.944-0.969) for all
cases to 0.984 (95% CI 0.970-0.995) for the 46% of cases classified as certain (p<0.001).
Conclusion: Leveraging AI uncertainty to guide a hybrid AI-radiologist screening reading strategy can potentially reduce workload by
~46% without decreasing performance.
Limitations: Identified limitations were that this was a retrospective study with single-site data.
Funding for this study: aiREAD was financed by NWO, KWF, HH.
Ethics committee - additional information: According to the Dutch Central Committee on Research involving Human Subjects,
ethical
approval was not necessary.
Setting up a complaint data registry for research on the human: a Swiss experience (7 min)
Benoît Dufour; Sion / Switzerland
Author Block: B. Dufour1, B. Rizk1, C. Thouly1, H. Brat1, N. Heracleous1, D. Goyard2, P. Petetin3, F. Zanca4; 1Sion/CH, 2Paris/FR, 3Berre
l'Etang/FR, 4Leuven/BE
Purpose: Since 2014, the Law on Human Research (LRH) in Switzerland protects individuals participating in human research projects,
while ensuring quality and transparency.
We detail the establishment of a Complaint Data Registry (CDR) within a private radiology network in Switzerland.
Methods or Background: Data in the registry encompass DICOM images, examination reports, and clinical/demographic
information.
Key elements in creating the registry included defining its purpose and objectives, establishing governance (legal structure, general
informed consent, access rights), and outlining operational procedures (data storage duration, pseudonymisation, encryption key
access).
For governance, we structured the organisational framework and designated responsible individuals.
A workflow for informed consent, including consent for AI-based image analysis, was implemented. Patients receive an SMS before
appointments, granting access to information about the data registry and consent process. Patients can opt in or out for research by
digitally signing the consent form on their smartphone or at the centre on the day of the exam. Signed consents are stored in our RIS,
allowing radiologists to identify approved research and AI-analysed data.
For the operational processes, data are collected on a gateway, pseudoanonymised and sent to a cloud platform for storage, while
ensuring segregation based on the data's source sites and projects.
Results or Findings: Results showed that 780,000 research consents were automatically stored in the RIS database between
18.01.2023-03.10.2023, with 678,235 consenting research data reuse (87%). Since implementing the registry, patient consent for AI-
based data analysis increased from 56% to 92%.
Conclusion: Our experience in setting up a CDR could serve as a promising model for other institutions seeking to improve
healthcare outcomes by leveraging complaint data.
Limitations: The Swiss context might be different in other countries and other RIS systems might not guarantee the same level of
integration.
Ethics committee - additional information: The study is about setting up a data registry.
Advancements in generative AI for radiological imaging (7 min)

Can Ozan Tan; Enschede / Netherlands
540
Author Block: E. Hofmeijer, X. Zu, C. O. Tan; Enschede/NL
Purpose: Generative artificial intelligence (AI) has emerged as a transformative force in the field of radiology. It can empower
radiologists with tools to enhance image quality, reconstruct degraded data, and synthesize realistic images, improving diagnostic
accuracy and efficiency. In particular, generative AI enables creation of synthetic datasets that facilitate training algorithms, as well
as residents and fellows, when real-world data is scarce or difficult to obtain due to privacy concerns.
Methods or Background: We have recently developed a pipeline for creating artificial 2D radiologic images. Publicly available
standard and low-dose chest CT images (805 scans; 39,803 2D images, 17% containing lung nodules) were used to generate
synthetic image. Five radiologists with experience in chest and thoracic imaging were asked to assess synthetic image quality
compared to the real ones.
Results or Findings: Radiologists rated artificial images as 3.13 ± 0.46 (1 [unrealistic] to 4 [indistinguishable to the original image]),
close to their rating of the original images (3.73 ± 0.31). An extended diffusion-based model was then used to identify features of the
lung nodules that distinguish malignant versus benign ones and to generate further synthetic images that should reflect these
features. The accuracy of malignant/benign classification based on synthetic images reached an accuracy of 85.5%.
Conclusion: Our results show that synthetic radiologic images are realistic and reliably adhere to the key radiographic features that
are reflective of pathological changes. These results, when shown to be reliable across imaging modalities, organs, and pathologies,
can enable tailored synthetic images on individual, personalised, patient profiles ("digital twins").
Limitations: The ethical considerations surrounding the use of generative AI in radiology need to be addressed.
Funding for this study: This study was funded by a ZonMw Innovative Medical Devices Initiative (IMDI) subsidy for the B3CARE
project (dossier number: 10-10400-98-008).
Ethics committee - additional information: The data used for this work is based on publicly available sources: Lung Image
Database Consortium (LIDC) and Image Database Resource initiative (IDRI).
Improving CT justification practices with machine learning and deep learning: a multi-site study (7 min)
Jaka Potočnik; Dublin / Ireland
Author Block: J. Potočnik, E. M. Thomas, A. Lawlor, D. Kearney, R. P. Killeen, E. J. Heffernan, S. J. Foley; Dublin/IE
Purpose: The aim of this study was to compare human experts with machine learning (ML) and deep learning (DL) models for
assessing justification of CT brain referrals. Multiclass classification of the anonymised referrals with ML and DL determined if
prediction models could generalise and automate this process.
Methods or Background: Anonymised adult brain CT referrals performed in 2020 and 2021 were sourced from three Irish CT
centres. A total of 3,000 referrals were randomly selected. Two radiologists and radiographers retrospectively categorised the
referrals using iGuide as: justified, unjustified, or potentially justified. The final justification label for each referral was determined by
majority vote or consensus.
Prior to the feature extraction with bag-of-words (BoW), term frequency-inverse document frequency, and Word2vec models, word
tokenisation, stop words removal, and Enchant spell correction of unstructured clinical indications was performed. The dataset was
randomly split into stratified training and test sets (80/20). Downsampling to the minority class ensured class balance. Support vector
machines, logistic regression, gradient boosting classifier (GBC), multi-layer perceptron, and bidirectional long-short term memory
neural network were evaluated. Their hyperparameters were tuned on the training set.
Results or Findings: A total of 11,090 referrals were collected and a random sample of 3,000 were reviewed. 238 (8.1%) were
categorised as unjustified, 811 (27.4%) potentially justified, and 1,909 (64.5%) justified by raters.
The best-performing classifier (BoW+GBC) achieved 94.4% accuracy and macro precision, recall, and F1 scores of 0.94.
Conclusion: ML and DL-based approaches can generalise and accurately predict the justification of radiology referrals in accordance
with the iGuide categorisation. This may help in addressing poor European justification practices.
Limitations: Downsampling resulted in a smaller dataset for multiclass classification, which, in turn, led to suboptimal performance
in DL. A larger, more representative dataset, along with a validation set, may provide better insights into performance.
Ethics committee - additional information: The University ethics committee has approved an ethics exemption (LS-E-21-216-
Potocnik-Foley) based on the outcomes of local DPIAs.
Artificial Intelligence in automated protocolling for Finnish brain MRI referrals (7 min)
Heidi Huhtanen; Turku / Finland
541
Author Block: H. Huhtanen, M. J. Nyman, A. Karlsson, J. Hirvonen; Turku/FI
Purpose: Advancements in AI-driven models for natural language processing has offered opportunities to automate many menial
tasks that require understanding written text. Automating the protocolling of incoming MRI referrals could reduce interruptions in
radiologists’ workflow. The purpose of this study is to test different AI models in assigning a suitable protocol and the need for
contrast medium for emergency brain MRI referrals.
Methods or Background: For training and testing the models, we collected 1,563 and 390 Finnish emergency brain MRI referral
texts, respectively. Data was labelled according to suitable imaging protocol and the need for contrast medium. We trained baseline
machine learning (ML) models (three different algorithms) and newer deep learning (DL) models (BERT and GPT3) for classification.
We also tested whether using less training data (50% of the training set) or using less data but upsampling it with augmentation
affected model performance.
Results or Findings: In protocol and contrast medium prediction, GPT3 outperformed other models with accuracies of 84% and
91%, respectively. BERT models had accuracies of 78% and 89%, and the best ML models 77% and 86%, respectively. For DL models,
using less training data affected performance negatively. Upsampling the data with augmentation boosted BERT’s accuracy in the
protocol task but not in the contrast medium task. For ML models, neither dataset size nor augmentation seemed to affect
performance.
Conclusion: Our results show that there is potential in using AI in automatic protocolling. Although GPT3 outperformed other
algorithms, BERT and ML models also performed well. However, the DL models seem to have more potential to improve performance
with increasing dataset size, than the ML models.
Limitations: The limitations of this study are the high imbalance between MRI protocol classes and using data from only one
institute.
Funding for this study: Funding was provided by the Emil Aaltonen Foundation (grant number: 230049), and the Radiological
Society of Finland.
Ethics committee - additional information: Review by the ethics committee was waived due to the retrospective nature of this
study in accordance with national legislation.
542
OF 13E - Leave a lasting impression with your work
Categories: Education, Professional Issues

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
In this session, organised by the ESR Journal Family, you will learn about boosting your work through the smart use of AI assistance in
the writing process, presenting it successfully, and disseminating it through social media.
Moderator:
Doenja Marina Johanna Lambregts; Amsterdam / Netherlands

How to make your paper sound intelligent, not artificial (15 min)
Daniel Pinto Dos Santos; Frankfurt / Germany
How to shine when presenting your work (10 min)

Gennaro D'Anna; Legnano / Italy
How to grow your audience and build a scientific network using social media (15 min)
Brendan S Kelly; Dublin / Ireland
543
RC 1310 - Imaging of the shoulder
Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Chiara Giraudo; Padova / Italy

Chiara Giraudo; Padova / Italy
Shoulder instability (15 min)

Christoph Schäffeler; Chur / Switzerland
1. To name and identify the stabilising structures of the shoulder.

2. To discuss different categories of shoulder instability.
3. To list the important imaging findings that may determine therapy.
Rotator cuff pathology (15 min)

Philip Robinson; Leeds / United Kingdom
1. To describe the anatomy and function of the rotator cuff tendons.

2. To explain the development of common rotator cuff pathologies and their imaging appearances.
3. To compare the sensitivity and specificity of imaging techniques available to assess rotator cuff pathology.
Other causes of shoulder pain (15 min)

Marco Zanetti; Baden / Switzerland
1. To name and identify the anatomic structures and pathologic conditions in the rotator cuff interval.
2. To recognise and classify the acromioclavicular (AC) joint injuries.
3. To know the differential diagnoses of shoulder muscle abnormalities not related to rotator cuff tendon tears.
Panel discussion: What are the challenges in evaluating various disorders of the shoulder? (10 min)
544
E³ 1323 - Gynaecology and obstetrics
Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Vlad Mr. Bura; Cluj-Napoca / Romania

Vlad Mr. Bura; Cluj-Napoca / Romania
Imaging of the uterus (28 min)

Ramona Anna Woitek; Cambridge / United Kingdom
1. To comprehend the imaging anatomy of the uterus and its changes throughout life and during pregnancy.
2. To understand the typical imaging features and local imaging-based staging of cervical and endometrial cancer.
3. To become familiar with the typical imaging features of benign disorders of the uterus, especially uterine leiomyomas,
adenomyosis, and endometriosis.
Disorders of the adnexa (28 min)

Rosemarie Forstner; Salzburg / Austria
1. To describe the imaging features of benign tumours of the ovaries.

2. To understand the diagnostic evaluation and imaging features of malignant tumours of the ovaries.
3. To identify the imaging features in regard to the stage and extent of adnexal tumours.
4. To become familiar with the imaging features useful for differentiating adnexal masses.
Acute gynaecological and obstetric disorders (28 min)

1. To become familiar with the typical and atypical imaging features of acute disorders of the uterus and the ovaries.
2. To understand the common emergencies associated with acute gynaecological disorders, including ectopic pregnancy, placenta
previa, and emergencies related to abortion.
545
RC 1308 - Imaging of the skull base and cranial nerves
Categories: Head and Neck, Imaging Methods, Oncologic Imaging

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Agnieszka Trojanowska; Lublin / Poland

Agnieszka Trojanowska; Lublin / Poland
Updates in cranial nerve imaging (15 min)

Christoph Kenis; Bruges / Belgium
1. To understand the basics of cranial nerve imaging.

2. To optimise imaging algorithms for cranial nerve palsies.
3. To review the novel techniques to delineate the cranial nerves.
New and advanced MRI techniques for the diagnosis of skull base lesions (15 min)
Philip Touska; London / United Kingdom
1. To review the advanced MRI sequences used at the skull base.

2. To present applications of multimodality imaging of the skull base.
3. To discuss the technical challenges and limitations of skull base imaging.
Skull base after treatment (15 min)

1. To review treatment-related changes at the skull base.

2. To discuss tools for early tumour detection and assessment of residual/recurrent disease.
3. To present new advances in surveillance imaging.
4. To predict early tumour detection and imaging markers for treatment response.
Panel discussion: Clinical applications, tips, and tricks? (10 min)
546
OF 13T - Tackling inequality in radiology
Categories: Education, General Radiology, Management/Leadership, Professional Issues, Students

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:

The need for diversity in radiology (10 min)

Judy Yee; New York / United States
1. To explain the need for diversity in radiology.

2. To highlight the current challenges in radiology.
Tackling gender disparities (10 min)

1. To highlight findings of a gender disparity study in Europe.

2. To discuss how inequality can be tackled in Europe.
Tackling inequality from the perspective of young radiologists (20 min)

Lorraine Murray; Limerick City / Ireland
1. To highlight inequality challenges as a young resident.

2. To discuss opportunities and ways forward as a young radiologist.
Open forum discussion: Tackling inequality from the perspective of young radiologists (15 min)
547
RC 1304 - From pulmonary nodule detection to histopathological diagnosis and treatment: which
route to choose?
Categories: Chest, Interventional Radiology

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Anna Rita Larici; Rome / Italy

Anna Rita Larici; Rome / Italy
The endobronchial route (15 min)

Arschang Valipour; Vienna / Austria
The transthoracic route (15 min)

Peter Beddy; Dublin / Ireland
1. To discuss a safe technique for transthoracic image guided lung biopsy.

2. To learn tips for the biopsy of lesions in tricky locations.
3. To understand how to identify and treat complications.
Thermal ablation of malignant pulmonary nodules (15 min)

Carole Ridge; Dublin / Ireland
1. To understand the principles of thermal ablation and its application in treating malignant pulmonary nodules.
2. To identify the different types of thermal ablation techniques, such as radiofrequency ablation (RFA), microwave ablation (MWA),
and cryoablation.
3. To comprehend the patient selection criteria for thermal ablation, considering nodule size, location, and co-morbidities.
4. To analyse the clinical outcomes and long-term efficacy of thermal ablation in treating malignant pulmonary nodules.
Panel discussion: Building bridges between radiologists and pulmonologists in the interventional approach of
pulmonary nodules (10 min)
548
E³ 1320 - Organ biopsy
Categories: Abdominal Viscera, Chest, Head and Neck, Interventional Oncologic Radiology, Musculoskeletal
ETC Level: LEVEL II
Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
The session summarises indications, techniques and results of percutaneous imaging-guided biopsies in the most common target
organs. The audience will also familiarise themselves with how to deal with the increasing number of biopsy procedures coming to the
interventional radiology department.
Moderator:

Neck organs (18 min)
Thoracic organs (18 min)

Paul Habert; MARSEILLE / France
Abdominal organs (18 min)

Ciara O'Brien; Toronto / Canada
Bone and soft tissues (18 min)

Richard Fawcett; Leeds / United Kingdom
Panel discussion: How can the IR clinic deal with the increasing number of biopsies? (13 min)
549
RW 13 - How to improve chest CT reports
Categories: Chest, Professional Issues

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Tips on improving chest CT reports (15 min)

Joe Barnett; St Albans / United Kingdom
1. To write clear reports addressing the clinical question.

Short cases review, interactive discussion and critiquing of reports (45 min)
Joe Barnett; St Albans / United Kingdom

2. To show how to make reports brief yet clinically pertinent.
550
NH 13 - Cancer screening and detection: can AI change the game?
Categories: Artificial Intelligence & Machine Learning, Oncologic Imaging, Research

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
The session will include a brief overview of the role of AI in the screening of several cancer tumours. Cancer prevention, screening and
early detection offer the best chance of beating cancer and saving lives. The extension of screening programmes and shortage of
radiologists are jeopardising elements for the fully implementing of population-based cancer programmes. The lectures will focus on
the feasibility, testing and planning of innovative AI solutions, with the identification of and addressing barriers and facilitators for the
utilisation of AI in screening services within the different health systems.
Moderator:
Chairperson's introduction: AI and screening (5 min)

AI and breast cancer screening (18 min)

Sarah Vinnicombe; Cheltenham / United Kingdom
AI and lung cancer screening (18 min)

Bram Van Ginneken; Nijmegen / Netherlands
AI and pancreas cancer detection (18 min)

Vincenza Granata; Vitulazio / Italy
AI and prostate cancer detection (18 min)

Olivier Rouviere; Lyon / France
Panel discussion: Are we prepared to lead early preclinical cancer diagnosis? (13 min)
551
EIBIR 13 - Innovative tools for comprehensive risk assessment in radiology: insights from the
SINFONIA Project
Categories: Artificial Intelligence & Machine Learning, EuroSafe Imaging/Radiation Protection

Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

Innovative tools for the accurate estimation of organ doses from radiological examinations (25 min)
1. To learn about the SINFONIA web-based tools for patient-dose assessment in x-ray imaging.
2. To understand the role of AI in patient dosimetry.
3. To understand why accurate patient dosimetry is needed not only in radiotherapy but also in diagnostic radiology.
AI-based algorithms for dose optimisation in CT (25 min)

Habib Zaidi; Genève / Switzerland
1. To learn about SINFONIA's AI-based algorithms for CT dose optimisation.

2. To understand how AI can assist in CT dose optimisation.
3. To appreciate the impact of AI on dose optimisation.
A novel software tool for risk appraisal (25 min)

1. To learn the basic concepts behind risk estimation in the context of medical exposures.
2. To understand how the new SINFONIA risk appraisal software tool works.
3. To understand the limitations and uncertainties in the risk estimates for individual patients and what this means for using the risk
appraisal tool in daily clinical routine.
Discussion (10 min)
552
RC 1313 - High-field and low-field MRI: what is the difference?
Categories: Imaging Methods, Multidisciplinary, Physics in Medical Imaging

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Sebastian Kozerke; Zurich / Switzerland

Sebastian Kozerke; Zurich / Switzerland
High-field MRI (15 min)

Siegfried Trattnig; Vienna / Austria
1. To learn about the signal-to-noise ratio (SNR), susceptibility and chemical shift dependence on the static magnetic field and its
impact on achievable spatial and spectral resolution and susceptibility-weighted imaging (SWI).
2. To understand the trade-offs, limitations and safety issues with high-field and ultra-high-field MRI.
3. To get an insight into the possible clinical applications where high-field MRI makes the difference.
Low-field MRI (15 min)

Andrew G. Webb; Leiden / Netherlands
1. To learn about the signal-to-noise ratio (SNR) dependence on the static magnetic field and its impact on the achievable contrast,
resolution, and acquisition times.
2. To understand the trade-offs and limitations associated with low-field MRI.
3. To get an insight into the possible clinical applications of low-field MRI systems.
Panel discussion: High-field vs low-field: room for both? (25 min)
553
SF 13 - The future is here: scales and numbers in neuroimaging
Categories: Neuro
ETC Level: LEVEL II
Date: March 1, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
The session will include an overview of quantitative neuroimaging biomarkers currently used for diagnostic purposes, treatment
decision and prognosis. The lectures intend to familiarise the audience with these scales and numbers derived from structural MRI,
molecular imaging or perfusion MRI or CT in clinical setting, and how they should be presented in neuroradiology reports.
Moderator:
Catherine Oppenheim; Paris / France

Catherine Oppenheim; Paris / France
Scales and numbers in stroke imaging (20 min)

Pedro Vilela; Lisbon / Portugal
Scales and numbers in brain tumour imaging (20 min)

Marion Smits; Rotterdam / Netherlands
Scales and numbers in neurodegeneration (20 min)

Panel discussion: Quantification in neuroradiology reports: does it impact patient management? (25 min)
554
RC 1314 - Pros and Cons: simulators will replace traditional clinical placements
Categories: Education, Imaging Methods, Professional Issues, Radiographers, Students

Date: March 1, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderators:
Asuncion Torregrosa Andres; Valencia / Spain

Pro (15 min)

Naomi Shiner; Nottingham / United Kingdom
This house believes that simulators will replace traditional clinical placements.
Con (15 min)

Paul Bezzina; Msida / Malta
This house believes that simulators will not replace traditional clinical placements.
Panel discussion: What does an optimum radiographer clinical training curriculum look like? (25 min)
555
MD 6 - Advanced prostate cancer (APC): new developments in imaging and treatment -

recommendations for clinical practice and directions for the future
Categories: Abdominal Viscera, Genitourinary, Multidisciplinary, Oncologic Imaging, Research

Date: March 1, 2024 | 10:15 - 11:15 CET
CME Credits: 1
Moderator:

1. To highlight new developments in the treatment of prostate cancer.

The imaging specialist's perspective (8 min)

The urologist's perspective (8 min)

Francesco Sanguedolce; Barcelona / Spain

Maria De Santis; Milan / Italy

556
CUBE 14 - The Good, the Bad and the Ugly' - case studies in aorto-iliac intervention

Date: March 1, 2024 | 10:30 - 11:00 CET
being explored.
Moderators:

The Good, the Bad and the Ugly' - Case studies in aorto-iliac intervention (28 min)
Frank Carey; Norwich / United Kingdom
1. To analyse case studies to determine appropriate interventional strategies in aorto-iliac disease.

2. To discuss complications and post-intervention care.
3. To develop critical thinking and decision-making skills.
557
EFRS 6 - Collaboration and shared resources of EFRS
Categories: Management/Leadership, Multidisciplinary, Professional Issues, Radiographers

ETC Level: LEVEL I
Date: March 1, 2024 | 10:30 - 11:30 CET
This session aims to gain greater knowledge of EFRS related projects and to share resources and the impacting of the development of
our profession.
Moderator:
Karoliina Paalimäki-Paakki; Oulu / Finland

Highlights on the ESCO MRI project (10 min)

Highlights on the EU JUST CT project (10 min)

Highlights on the EU REST project (10 min)

Francis Zarb; Msida / Malta
Discussion (20 min)
Closing (5 min)
558
AI-SC 5 - Quality assurance of AI applications: which role for the medical physicist?

Date: March 1, 2024 | 11:15 - 12:15 CET
Moderator:
Hilde Bosmans; Leuven / Belgium

Quality assurance of AI applications: which role for the medical physicist? (57 min)
1. To understand the importance of safety and quality assurance of AI in healthcare.

2. To discuss examples of QA in practice.
3. To appreciate the role of the medical physicist.
559
MD 7 - Rectal cancer: new developments in imaging and treatment - recommendations for clinical

Date: March 1, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:

1. To highlight new developments in the treatment of rectal cancer.

2. To critically review recent evidence in imaging and put this in perspective of the new treatment developments.
The radiation oncologist's perspective (8 min)

Vincenzo Valentini; Rome / Italy

Geerard L. Beets; Amsterdam / Netherlands

560
PL 2 - Radiology's transformation: new roles for talent and technology
Categories: Management/Leadership, Professional Issues

Date: March 1, 2024 | 11:30 - 12:00 CET
CME Credits: 0.5
Moderator:
Radiology's transformation: new roles for talent and technology (28 min)
Geoffrey D. Rubin; Tucson / United States
561
CUBE 15 - Strategies in the management of acute pulmonary embolism - How do I do it?

Date: March 1, 2024 | 12:00 - 12:30 CET
Moderators:

Strategies in the management of acute pulmonary embolism - How do I do it? (28 min)
AbdulRahman Jan Alvi; Riyadh / Saudi Arabia
1. To receive a summary of strategies for managing acute pulmonary embolism.

2. To understand interventional techniques in the management of acute pulmonary embolism using case examples.
562
RC 1411 - When neuro meets musculoskeletal: the craniocervical junction
Categories: Emergency Imaging, Musculoskeletal, Neuro

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Anatomy, normal variants and congenital pathology of the craniocervical junction (15 min)
Andrea Rossi; Genoa / Italy
1. To revise the fundamental element of craniocervical junction anatomy with a focus on radiological markers of instability.
2. To become familiar with the most common normal variants.
3. To recognise the principal congenital anomalies involving the craniocervical junction.
Traumatic findings at the craniocervical junction (15 min)

Üstün Aydingöz; Ankara / Turkey
1. To name fractures, dislocations, and other traumatic findings that occur at the craniocervical junction.
2. To estimate how pathophysiology pertains to imaging findings in craniocervical junction trauma.
3. To describe how new imaging techniques can help change diagnostic algorithms related to craniocervical trauma.
Inflammatory and tumoral lesions at the craniocervical junction (15 min)

Cem Çalli; Izmir / Turkey
1. To define typical inflammatory lesions of the craniocervical junction.

2. To address the possible tumoral lesions of the craniocervical junction.
3. To describe radiological findings of inflammatory and tumoral lesions of the craniocervical junction
Panel discussion: Craniocervical junction: the terrain of musculoskeletal, neuro or both? (10 min)
563
EIBIR 14 - From pixels to probes: unveiling the potential of integrated photonics in medical imaging
Categories: Imaging Methods, Neuro, Physics in Medical Imaging, Research

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Antonio Pifferi; Milan / Italy

Antonio Pifferi; Milan / Italy
HyperProbe: transforming brain surgery by advancing functional-guided neuronavigational imaging (24 min)
Luca Giannoni; Sesto Fiorentino / Italy
1. To learn novel optical imaging technologies to aid neurosurgery.

2. To appreciate the potential of hyperspectral imaging for non-invasive diagnosis.
3. To understand how hyperspectral imaging can overcome current limitations in neuronavigation.
Med-IPUT, integrated photonic ultrasound transducers (IPUTs): revolutionising medical ultrasound imaging with light
(24 min)
Paul van Neer; The Hague / Netherlands
1. To learn how medical ultrasound images can be created using light.

2. To appreciate the exciting new possibilities IPUTs enable.
3. To understand the potential impact of IPUTs on medical ultrasound imaging.
Discussion (6 min)
564
HW 14Ma - Ultrasound of the muscles and nerves of the upper limb

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1

1. To become familiar with muscle and nerve anatomy of the upper limb with the usual landmarks.
2. To develop practical skills in performing ultrasound.
3. To understand the main limitations of ultrasound.
4. To learn tricks and tips to improve daily practice.

Lionel Pesquer; Bordeaux / France
Georgina Marian Allen; Oxford / United Kingdom
Tutors
Maria Pilar Aparisi Gomez; Valencia / Spain
Žiga Snoj; Ljubljana / Slovenia
Salvatore Gitto; Milano / Italy
Saulius Rutkauskas; Kaunas / Lithuania
565

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1


Philippe Puech; LILLE / France
566
E³ 25C - Peripheral arterial disease
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Atherosclerotic peripheral arterial occlusive disease (23 min)

Giles Hannibal Roditi; Glasgow / United Kingdom
1. To discuss the role of CT and MR in imaging suspected or known atherosclerotic peripheral arterial disease.
2. To show examples of atherosclerotic disease of the peripheral arteries.
3. To discuss the essential elements of the radiology report in patients with suspected or known atherosclerotic peripheral arterial
disease.
Non-atherosclerotic peripheral arterial disease, including compression syndromes (23 min)

Tim Leiner; Rochester / United States
1. To discuss the role of CT and MR in imaging suspected or known non-atherosclerotic peripheral arterial disease.
2. To show examples of non-atherosclerotic peripheral arterial disease, including compression syndromes.
3. To discuss the essential elements of the radiology report in patients with suspected or known non-atherosclerotic peripheral
arterial disease, including compression syndromes.
567
ESR/UEMS - The voice of the future
Categories: Education, Management/Leadership, Professional Issues, Students

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderators:
Paolo Ricci; Roma / Italy
Miraude Adriaensen; Heerlen / Netherlands

Paolo Ricci; Roma / Italy
Miraude Adriaensen; Heerlen / Netherlands
ESR: past, present, and future (10 min)

1. To learn about the mission and vision of the ESR.

2. To understand the organisation of the ESR.
3. To appreciate the leadership tools available to shape next-generation radiology.
UEMS: past, present, and future (10 min)

Othmar Engelbert Haas; Klagenfurt / Austria
1. To learn about the mission and vision of the UEMS.

2. To understand the organisation of the UEMS.
3. To appreciate the leadership tools available to shape next-generation radiology.
Opportunities and involvement of junior doctors within the UEMS Radiology Section (8 min)
Katrine Riklund; Umeå / Sweden
Antanas Montvila; Kaunas / Lithuania
1. To learn about the involvement of radiology residents within the UEMS Radiology Section.
2. To appreciate the role of the European Junior Doctors.
3. To learn about the ETAP 2.0 programme.
Opportunities and involvement of medical students within the ESR (8 min)

Anna Detoraki; Nicosia / Cyprus
1. To learn about the involvement of medical students within the ESR.

2. To appreciate the Rising Star programme.
3. To learn about the Invest in the Youth programme.
Opportunities and involvement of radiology residents within the ESR (8 min)

568
1. To learn about the involvement of radiology residents within the ESR.

2. To understand the role of the RTF.
3. To appreciate the input and potential of the next generation.
How to become an EDiR holder (8 min)

Lucian Beer; Vienna / Austria
1. To learn about the EDiR.

2. To know about the EDiR holder group.
3. To understand the various ways to prepare yourself for the EDiR.
Panel discussion: Is there a future for radiologists? (5 min)
569
OF 14R - Nurturing the caregiver: prioritising radiographers' mental health and wellness
Categories: Education, Management/Leadership, Medico-legal, Professional Issues, Radiographers

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
In the demanding landscape of healthcare, radiographers serve as the cornerstone of diagnostic and radiotherapy services, playing a
pivotal role in patient care. The session centres on the crucial theme of fostering the mental health and well-being of radiography
professionals and will consist of three insightful talks that collectively address the critical issue of sustaining and supporting those
who dedicate their careers to caring for others. The session will examine the multifaceted challenges faced by radiographers and the
collective responsibility to prioritise their mental health and wellness. Through practical tips, compassionate leadership, and
resilience-building strategies, this session aims to empower radiography professionals to continue providing exceptional care while
nurturing their own well-being.
Moderator:
Megan Brydon; HALIFAX / Canada

Megan Brydon; HALIFAX / Canada
Tips for addressing staff burnout, sustainability and retention (16 min)
From empathy to action: compassionate leadership in combating workplace harassment (16 min)
Amy Hancock; Exeter / United Kingdom
Building resilience: enhancing radiographers' mental health in high-stress environments (16 min)
Barry Hallinan; Dublin / Ireland
570
RC 1407 - Pros and Cons: prostate-MRI reporting should be done with the aid of AI systems
Categories: Artificial Intelligence & Machine Learning, Genitourinary, Imaging Informatics, Oncologic Imaging, Professional Issues
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Pro (15 min)

This house believes that prostate-MRI reporting should be done with the aid of AI systems.
Con (15 min)

Maarten De Rooij; Nijmegen / Netherlands
Prostate-MRI reporting should be done without the aid of AI systems.
Panel discussion: Will AI replace radiology stars, or will it be an aiding tool? (25 min)
571
RPS 1416 - Latest CT technology in oncologic imaging
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Oncologic Imaging
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Rho/Z maps derived from dual-energy CT for differentiation of hypodense liver lesions in patients with colorectal
carcinoma (7 min)
Mirela Dimitrova; Frankfurt a. Main / Germany
Author Block: M. Dimitrova, A. Gökduman, V. Koch, L. D. Gruenewald, L. S. Alizadeh, S. Martin, T. J. Vogl, I. Yel, C. Booz; Frankfurt/DE
Purpose: This study aims to evaluate the diagnostic accuracy of dual-energy computed tomography (DECT)-based Rho/Z-maps for
differentiation of hypodense liver lesions in patients with diagnosed colorectal carcinoma compared with conventional CT value
measurements.
Methods or Background: This retrospective study included 364 patients (mean age, 70 ± 13 years; 183 men and 181 women)
suffering from colorectal carcinoma who had undergone third-generation dual-source DECT as part of tumour staging between
January 2016 and January 2023. For this study, we measured Rho (electron density) and Z (effective atomic number) values as well as
Hounsfield units (HU) in hypodense liver lesions. Values were compared, and diagnostic accuracy for differentiation was computed
using receiver operating characteristic (ROC) curve analyses. MRI or biopsy served as a standard of reference for lesion
characterisation.
Results or Findings: A total of 822 lesions (351 metastases and 471 cysts) in contrast-enhanced DECT images were evaluated.
Mean Rho, Z and HU120kV values showed significant differences between both liver lesion types (p<.0001). The greatest difference
between measured values for liver metastases and benign cysts was observed for Rho (MD, -35.31 ± 3.217, 92%; HU120kV MD,
-21.46 ± 3.017, 60%). Rho and Z measurements showed a higher AUC value (Rho, 0.962; 95% CI, 0.936-0.989; Z, 0.918; 95% CI,
0.845-0.954) compared with HU120kV (0.802; 95% CI, 0.741-0.862) for lesion differentiation.
Conclusion: Rho and Z measurements derived from DECT allow for improved differentiation of liver metastases and cysts in patients
with colorectal carcinoma compared with conventional CT value measurements.
Limitations: The single-centre retrospective study design was an identified limitation.
Photon-counting CT in depicting pancreatic ductal adenocarcinoma (PhDAC): a pilot study (7 min)

Mischa Woisetschlager; Linköping / Sweden
572
Author Block: H. Blomstrand1, W. C. Bartholomä1, E. Tesselaar1, B. Björnsson1, T. Bjerner1, M. Sandborg1, N. Elander2, M.
Woisetschlager1; 1Linköping/SE, 2Liverpool/UK
Purpose: Pancreatic ductal adenocarcinoma (PDAC) is a challenging condition with low survival rates. Accurate imaging is vital for
treatment decisions. Photon-counting CT (PCCT) is an emerging technology with potential benefits over standard CT. This pilot study
assessed PCCT's usefulness in pancreatic and PDAC imaging compared to energy-integrating detector (EID) CT.
Methods or Background: Eight PDAC patients received equivalent radiation doses of EID and PCCT scans, with five experienced
evaluators rating image quality using a Likert scale. Quantitative measures including contrast-to-noise ratio (CNR), signal-to-noise
ratio (SNR), noise levels, and tissue sharpness (line density profiles) were also recorded.
Results or Findings: PCCT images matched EID CT quality overall, with a slight preference for EID CT in arterial phase images for
three parameters (p=0.015). Venous phase image quality showed no significant difference (p>0.05). Quantitatively, PCCT displayed a
trend toward higher CNR and SNR, indicating potential advantages in tissue differentiation. Noise levels were lower in PCCT images,
especially in pancreatic tissue (p=0.064). Line density profiles hinted at sharper tissue edges in PCCT, although not statistically
significant.
Conclusion: In conclusion, this pilot study suggests that PCCT is a promising imaging tool for pancreatic cancer patients. Despite
slight preferences for EID CT in certain criteria, PCCT's quantitative advantages and noise reduction trends indicate its potential for
enhancing soft tissue discrimination and clinical outcomes, pending further advancements in PCCT technology.
These findings suggest PCCT's promise in pancreatic imaging, offering similar image quality to EID CT. Quantitative assessments hint
at potential benefits in tissue differentiation and noise reduction. While EID CT was preferred in some criteria, PCCT's advantages may
improve with evolving technology.
Limitations: This is a pilot study with a low number of patients. Comparison was made only to one scanner (Somatom Force).
Ethics committee - additional information: This study was approved by the Swedish Ethical Review Authority.
Correlation between iodine concentration in lung cancer on dual-energy CT and non-mucinous adenocarcinoma grading
(7 min)
Hikaru Watanabe; Tokyo / Japan
Author Block: H. Watanabe, Y. Sato, K. Oikado, T. Terauchi; Tokyo/JP

Purpose: The fifth edition of the WHO classification introduced a grading system for invasive non-mucinous adenocarcinoma. Lesions
are categorised into Grade 1 (well-differentiated), Grade 2 (moderately differentiated), and Grade 3 (poorly differentiated) based on
the percentage of histological subtypes with a high-risk pattern for poor prognosis. This study investigates the correlation between
the iodine concentration (IC) of lung adenocarcinoma measured by dual-energy CT (DECT) and this WHO grading.
Methods or Background: We examined preoperative contrast DECTs from 111 patients (46 men, 65 women; median age: 70 years;
37 Grade 1, 25 Grade 2, 49 Grade 3) diagnosed with primary lung adenocarcinoma between January and December 2022. Cases with
a solid component diameter under 5 mm on CT or those with invasive mucinous adenocarcinoma were excluded. IC measurements
were taken during the equilibrium phase, 3 minutes after contrast administration. The corrected IC was determined by dividing the IC
in the solid component of the lesion by the IC in the descending aorta at the tracheal bifurcation level. An ordinal logistic regression
model, accounting for corrected IC, smoking status, nodule type (solid or part-solid), gender, and age, was employed for the multiple
regression analysis.
Results or Findings: The multivariable analysis identified that significant independent predictors for a higher grade of invasive non-
mucinous adenocarcinoma were the corrected IC (OR, 2.74 [95% CI, 1.46–5.98]; p = .005) and the presence of a pure solid nodule
(OR, 5.02 [95% CI, 2.21-11.8]; p<.001).
Conclusion: IC measurements using DECT may predict the grade of pulmonary invasive non-mucinous adenocarcinoma.
Limitations: This study's limitations include its single-centre focus and retrospective design.
Funding for this study: This study was financially supported by Bayer Pharma Japan.
Ethics committee - additional information: The institutional review board approved this study, and there was a waiver for the
need for written informed consent.
Tumour response in non-small cell lung cancer patients treated with chemotherapy and targeted therapy: can spectral
CT predict response to treatment based on baseline data? (7 min)
Ronghua Mu; Guilin / China
573
Author Block: R. Mu1, X. Zhu1, X. M. Liu2, Z. Song2; 1Guilin/CN, 2Guangzhou/CN
Purpose: To assess the efficacy of baseline morphological features and quantitative parameters of lung spectral CT images for
predicting tumour response in patients with non-small cell lung cancer (NSCLC) receiving first-line chemotherapy and targeted
therapy.
Methods or Background: A total of 86 patients with stage III or IV NSCLC who received chemotherapy and targeted therapy
between August 2020 and August 2022 were retrospectively included. All patients underwent lung spectral CT scans before
chemotherapy and within one week after two cycles of treatment. Patients were divided into response and non-response groups
based on the Response Evaluation Criteria in Solid Tumours (RECIST). Binary logistic regression analysis was used to select the most
valuable CT imaging features, and three prediction models were constructed based on morphological features, spectral quantitative
parameters, and their combined model, respectively. Receiver operating characteristic curve (ROC) analysis was used to evaluate the
ability of these models to distinguish between RECIST responders and non-responders, and different models were compared using the
DeLong test. A novel predictive model was established based on CT morphological features and spectral quantitative parameters, and
the radiomics nomogram was evaluated using calibration curve, Hosmer-Lemeshow test, and decision curve analysis.
Results or Findings: Four CT morphological features (vessel convergence sign, invasion of blood vessels, lobulation sign, spicule
sign) and two spectral quantitative parameters (Ln AID, Ln AEF) performed well in predicting treatment response. The combined
nomogram achieved the highest performance (average AUC=0.887).
Conclusion: Spectral CT parameters can improve the accuracy of predicting short-term tumour response to treatment in NSCLC
patients. Spectral CT parameters can be surrogate imaging biomarkers for treatment evaluation of NSCLC patients.
Parameters of dual-layer spectral detector CT can improve diagnostic efficiency of microvascular invasion in
Jian Lv; Guilin / China
Author Block: J. Lv1, R. Mu1, X. Qin1, W. Zheng1, P. Yang1, B. Huang1, X. Li1, X. M. Liu2, X. Zhu1; 1Guilin/CN, 2Guangzhou/CN
Purpose: It is difficult to distinguish microvascular invasion (MVI) in hepatocellular carcinoma (HCC) with traditional computed
tomography (CT). This study aimed to assess MVI in HCC by dual-layer spectral detector CT (DLCT) combined with morphological
parameters.
Methods or Background: In this study, spectral images of 105 HCC patients were retrospectively analysed in the arterial phase
(AP), portal-venous phase (PP) and delayed phase (DP). Morphological parameters and three-phase enhanced spectral parameters
were calculated. Measurements were statistically compared using the independent samples t-test. The combination models of
morphological parameters (Model Ⅰ), spectral parameters (Model Ⅱ), spectral parameters and morphological parameters (Model Ⅲ)
were constructed. The receiver operating characteristic curve (ROC) analysis was used to evaluate the diagnostic performance of
tumours' MVI assessment.
Results or Findings: Tumours' MVI was found in 52 (49.5%) HCC patients confirmed by pathology. There were statistically
significant differences of mosaic architecture tumour capsule, corona enhancement, nodule-in-nodule architecture, iodine density in
arterial phase (ID-A), iodine density in portal-venous phase (ID-P) and iodine density in delayed phase (ID-D) between the MVI group
and none MVI group (P<0.05). Among single parameters, the ID-P demonstrated the best predictive efficiency for MVI with an area
under the curve (AUC) of 0.832. Model Ⅰ and Model Ⅱ showed predictive performance with AUCs of 0.846 and 0.872. Model Ⅲ displayed
higher predictive performance with AUC of 0.904. The results of the DeLong test indicated that there was no statistical difference in
the AUCs among all parameters and models.
Conclusion: The spectral parameters combining morphological parameters provides a promising tool to improve the evaluation for
predicting MVI in HCC.
Deep learning-based iodine contrast-augmenting algorithm for low-contrast-dose liver CT to assess hypovascular
hepatic metastasis (7 min)
Taehee Lee; Seoul / Korea, Republic of
574
Author Block: T. Lee1, J-H. Yoon1, J. Park2, J. Lee1, J. W. Choi1, C. Ahn1, J-M. Lee1; 1Seoul/KR, 2Busan/KR
Purpose: The study aimed to investigate the diagnostic value of low-contrast-dose liver CT using a deep learning-based iodine
contrast-augmenting algorithm (DLICA) to assess hypovascular hepatic metastases.
Methods or Background: This retrospective study included 128 patients who underwent contrast-enhanced DECT for hepatic
metastasis surveillance between July 2019 and June 2022 using a 30% reduced iodine contrast dose in the portal phase. Three image
types were reconstructed: 50-keV virtual monoenergetic images (50-keV VMI); linearly blended images simulating 120-kVp images
(120-kVp); and post-processed 120-kVp images using DLICA (DLICA 120-kVp). Three reviewers evaluated lesion conspicuity, and
contrast-to-noise ratios (CNRs) were measured from the regions of interest in the metastatic lesions and liver parenchyma. The
detection performance for hepatic metastases was evaluated using a jackknife alternative free-response receiver operating
characteristic method, with the consensus of two independent radiologists as the reference standard.
Results or Findings: DLICA 120-kVp demonstrated significantly higher CNR of lesions to liver (5.7±3.1 vs. 3.8±2.1 vs. 3.8±2.1) and
superior lesion conspicuity (4.0[3.3–4.3] vs. 3.7[3.0–4.0] vs. 3.7[3.0–4.0]) compared with 50-keV VMI and 120-kVp (P<0.001 for all).
Although there was no significant difference of merit in the figure for lesion detection among the three methods (P=0.105), DLICA
120-kVp had a significantly higher figure of merit for lesions with a diameter <20 mm than 50-keV VMI (0.677 vs. 0.648, P=0.007).
DLICA 120-kVp also demonstrated higher sensitivity on a per-lesion basis than the 50-keV VMI (81.2% vs. 72.9%, P<0.001).
Conclusion: DLICA 120-kVp provided higher lesion conspicuity and similar diagnostic performance to detect hypovascular hepatic
metastases compared with 50 keV VMI. DLICA 120-kVp can serve as an alternative to 50keV VMI for improved detection of liver
metastases, particularly in the case of smaller lesions, circumventing the need for high-end DECT equipment.
Limitations: The retrospective study design and the small sample size were identified limitations.
Funding for this study: This study was supported by a research grant from ClariPi (No. 06-2022-4530).
Ethics committee - additional information: This retrospective study was approved by the Institutional Review Board of Seou
National University Hospital and the requirement for written informed consent was waived.
The value of dual-layer spectral-detector CT in detecting neuroendocrine tumour liver metastases: comparison with
polyenergetic CT and Gd-EOB-DTPA-enhanced MR (7 min)
Tiansong Xie; Shanghai / China
Author Block: T. Xie, W. Liu, W. Deng, Y. Wang, W. Tang, Z. Zhou; Shanghai/CN

Purpose: The study aimed to investigate the value of dual-layer spectral-detector CT (DLCT) in detecting neuroendocrine tumour
liver metastases (NETLM) and to compare the diagnostic performance of polyenergetic CT images (PEI), DLCT, and Gd-EOB-DTPA-
enhanced MR.
Methods or Background: Seventy-two patients with suspected NETLM who underwent both DLCT and Gd-EOB-DTPA-enhanced MR
within three weeks were retrospectively enrolled, and they were divided into dataset 1 and dataset 2. In dataset 1, virtual
monoenergetic images (VMI) at 40-140 keV (VMI40-140, 10-keV interval) were reconstructed. Signal-to-noise ratio (SNR) and contrast-
to-noise ratio (CNR) of NETLM were measured and compared between PEI and VMIs. In dataset 2, two readers independently
evaluated the diagnostic performance of PEI (Set 1), VMI + PEI (Set 2), and Gd-EOB-DTPA-enhanced MR (Set 3) for detecting NETLM.
Subgroup analysis stratified by lesion size was performed.
Results or Findings: In dataset 1, VMI40 exhibited highest SNR and CNR across all energy levels, which was significantly higher
than that of PEI (P<0.01). In dataset 2, a total of 477 lesions were identified (396 metastases, 81 benign lesions). Adding VMI40 into
PEI resulted in a significantly improved per-lesion AUC and sensitivity (AUC, Set 1: 0.85; Set 2: 0.90, P < 0.01; Sensitivity, Set 1: 0.76;
Set 2: 0.86, P < 0.01). Set 3 achieved the highest per-lesion AUC of 0.97 and per-lesion sensitivity of 0.95. Subgroup analysis
demonstrated that the difference in diagnostic performance was concentrated on lesions <10 mm.
Conclusion: Gd-EOB-DTPA-enhanced MR yielded the highest diagnostic accuracy for NETLM detection. DLCT should be an alternative
choice, given the superior image quality and incremental diagnostic value of VMI40.
Limitations: The retrospective study design introduces the potential risk of selection bias.
Funding for this study: Funding was received from the Hong Kong, Macao, and Taiwan Cooperation Project of Scientific and
Innovative Plan of Shanghai (22490760800).
Ethics committee - additional information: This retrospective study was approved by the Institutional Review Board of the Fudan
University Shanghai Cancer Center (1612167–18). Informed consent was waived owing to the retrospective nature of the study.
Photon-counting computed tomography derived spectral reconstructions: a promising approach to diminish beam-
hardening artefacts in thoracic imaging (7 min)
Florian Haag; Heidelberg / Germany
575
Author Block: F. Haag1, N. Große Hokamp2, D. P. Overhoff1, G. Dasegowda3, A. Hertel1, D. Nörenberg1, S. O. Schönberg1, M. K. Kalra3,
M. Frölich1; 1Mannheim/DE, 2Cologne/DE, 3Boston, MA/US
Purpose: The use of port implants in cancer patients during follow-up CT scans can result in beam hardening artefacts, which can
significantly impact the quality and accuracy of the acquired images. The aim of this study is to improve image quality and accuracy
in thoracic imaging using photon-counting computed tomography (PCCT) derived spectral reconstructions.
Methods or Background: This retrospective single-centre study enrolled 49 patients with port systems undergoing CT for staging of
oncologic disease. 8 ROIs for 19 spectral reconstructions (polyenergetic imaging, monoenergetic reconstructions from 40 to 190 keV
[in steps of 10] as well as iodine maps and virtual non contrast [VNC]) were measured, resulting in a total of 7,448 mean and SD
Hounsfield unit measurements of port-chamber associated hypo- and hyperdense artefacts, bilateral muscles and vessels.
Additionally, a subjective analysis of the degree of artefacts was performed in consensus reading by two radiologists with more than
25 years and more than 5 years of experience in thoracic imaging.
Results or Findings: In quantitative analysis, a significant association of keV and iodine contrast as well as artefact intensity was
noted (p<0.001). In qualitative assessment, utilisation of 120 keV monoenergetic reconstructions could reduce severe and
pronounced artefacts completely, when compared to lower keV reconstructions (p<0.001). For imaging findings, no significant
difference between monoenergetic reconstructions could be noted (all p>0.05).
Conclusion: PCCT-derived, monoenergetic CT reconstructions are capable of reducing beam-hardening artefacts in chest imaging.
They may prove helpful to improve diagnostic confidence in these cases.
Limitations: The present study deals mainly with the potential of spectral reconstructions for artefact suppression. Other algorithms
for artefact reduction were not investigated and could be the subject of further research.
Funding for this study: This research was funded by the Photon-counting Consortium Baden-Württemberg (PC3), BMBF-Research
Campus M2OLIE and Hector Foundation, as well as a travel grant from ICON (DFG).
576
HW 14Sc - Imaging strategies in acute stroke

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1

1. To become familiar with imaging algorithms for a selection of acute stroke patients for IVT and EVT.
2. To learn about imaging protocols based on different imaging modalities.
3. To develop practical skills in detecting haemorrhage and large vessel occlusion.
4. To develop practical skills in detecting non-salvageable infarct "core" and potentially viable brain tissue.
5. To become familiar with treatment strategies for stroke following the latest research guidelines.

Karl-Olof Loevblad; Geneva / Switzerland
577
BS 14 - Abdominal viscera: imaging the liver
Categories: Abdominal Viscera, General Radiology, Imaging Methods

ETC Level: LEVEL II
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Nikolaos Kartalis; Stockholm / Sweden

Imaging diffuse liver disease (19 min)

Maxime Ronot; Paris / France
1. To present the current imaging techniques for evaluating diffuse liver disease.
2. To describe the typical imaging features of diffuse liver disease.
Imaging of cystic liver lesions (19 min)

1. To present the current imaging techniques for evaluating cystic liver lesions.
2. To become familiar with the typical imaging findings of cystic liver lesions.
Imaging of solid liver lesions (19 min)

Giuseppe Brancatelli; Palermo / Italy
1. To present the current imaging techniques for solid liver masses.

2. To become familiar with the typical imaging findings of solid liver masses.
578
RPS 1415 - Exploring imaging and interventional treatment of atherosclerotic lesions: fresh
perspectives
Categories: Cardiac, Imaging Methods, Interventional Radiology, Vascular

Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Semi-automatic measurement of carotid intima-media thickness in atherosclerosis (7 min)

Jiatong Xu; Shanghai / China
Author Block: L. Zhang, J. Xu, M. Chen; Shanghai/CN

Purpose: This study aims to demonstrate if semi-automatic measurement can improve the accuracy of carotid intima-media
thickness (CIMT) .
Methods or Background: This study analysed the results of carotid ultrasound examinations of 1637 patients. During the complete
cardiac cycle, three sonographers used semi-automatic measurement of the systolic and diastolic and manual measurement to
measure CIMT. The relevant data of the patients was also collected, including age, gender, diabetes status, and other cardiovascular
risk factors.
Results or Findings: For semi-automatic measurement of CIMT, the intraclass correlation coefficient (ICC) of the three-operator
agreement analysis is 0.99 (95% confidence interval: 0.984-0.994. There is a statistical difference between the semi-automatic and
manual measurement. There is no statistical difference between the left CIMT and the right CIMT measured by semi-automatic and
manual measurement. Compared with the cerebrovascular event group, the bilateral semi-automatic CIMT values between the
normal group and the cerebrovascular event group were statistically different.
Conclusion: Using the semi-automatic measurement of CIMT, the consistency between operators is better, and the repeatability is
higher. It can replace the manual measurement of CIMT, provide a more accurate and stable inspection method, and better evaluate
the degree of atherosclerosis.
Limitations: The study also has certain limitations. In addition to being subject to the inherent limitations of retrospective research,
due to technical problems, it needs to use specific machine measurements. Since the samples of this study come from the routine
examination of the hospital, a larger community study is needed at a later stage, to further prove the conclusion of this study.
Funding for this study: This research received no specific grant from any funding agency in the public, commercial or not-for-profit
sectors.
Ethics committee - additional information: Ethics approval was obtained from the Ethics Committee of Shanghai Tongren
Hospital.
Accuracy of an artificial intelligence-based algorithm for fully automated detection of aortic wall calcifications in chest
CT (7 min)
Christopher Schuppert; Heidelberg / Germany
579
Author Block: C. Schuppert1, R. Saffar1, J. Sperl2, J. Oravcová3, M. Kreibich1, J. Weiß1, F. Bamberg1, M. Czerny1, C. L. Schlett1; 1Freiburg
im Breisgau/DE, 2Erlangen/DE, 3Bratislava/SK
Purpose: This study aimed to assess the accuracy of an artificial intelligence-based algorithm for fully automated detection of aortic
wall calcifications in chest CT with a special focus on the surgically relevant clamping zone in the upper ascending aorta.
Methods or Background: We retrospectively included 100 chest CT scans from 92 patients who were examined on a third-
generation dual-source scanner. Subsamples comprised 47 scans with an aortic angiography using iodinated intravenous contrast
media and ECG-gating and 53 unenhanced scans. A previously validated deep learning algorithm performed aortic landmark
detection and aorta segmentation. The segmentation mask was divided into eight anatomic segments and aortic wall calcifications
were detected using a mean-based HU thresholding, yielding a binary score for their presence. Algorithm parameters (calcium cluster
size threshold, aortic mask dilatation) were varied to determine optimal performance. A visual rating served as a reference. Standard
estimates of diagnostic accuracy and interrater agreement using Cohen’s kappa were calculated.
Results or Findings: Aortic wall calcifications were observed in 74% of the examinations with a prevalence of 27% to 70% by aorta
segment. Using different parameter combinations, the algorithm provided results for 97% to 100% of the examinations and 95% to
99% of the combined aortic segments. The best-performing parameter combination for the presence of calcifications in the aortic
clamping zone yielded a sensitivity of 93% and a specificity of 82%, with an area under the receiver operating characteristic curve of
0.874. Using these parameters, the interrater agreement ranged from κ 0.66 to 0.92 per segment.
Conclusion: Fully automated detection of aortic wall calcifications in chest CT performs with good accuracy. This includes the critical
preoperative assessment of the aortic clamping zone.
Limitations: Intravenous contrast media and ECG-gating were not individually assessed as factors influencing algorithm
performance, which is a limitation.
Ethics committee - additional information: This retrospective single-centre study was approved by the local institutional review
board.
Quantitative parameters from dual-layer detector spectral CT (DLS-CT) in distinguishing non-calcified plaques and
chronic total occlusion (CTO) components in lower limb arteries (7 min)
Li Zhou; Xi'an / China
Author Block: L. Zhou, N. Ding, Y. Cui, Y. Han, J. Yang, X. Zhang, P. Cao, X. Huang; Xi'an/CN
Purpose: This study aimed to explore the application value of quantitative parameters from dual-layer detector spectral CT (DLS-CT)
in distinguishing non-calcified plaques and chronic total occlusion (CTO) components in lower limb arteries
Methods or Background: Peripheral arterial disease (PAD) patients with DLS-CT examinations were prospectively enrolled from
October 2022 to February 2023. Non-calcified plaques with a stenosis degree > 50% and CTO lesions in the above-knee arteries were
analysed. Quantitative parameters of conventional images, 40keV, 70keV, virtual non-contrast, iodine density, Z-effective (Zeff),
electron density (ED), and slope of the energy spectrum curve (λHU, λHU = (CT40keV -CT70 keV)/-30) of the non-calcified plaques
and CTO lesions were evaluated for statistical analysis.
Results or Findings: Ninety-three lesions in 24 patients were evaluated, consisting of 72 non-calcified plaques and 21 CTO lesions.
Analysis of the results indicated that CT40keV, iodine density and Z-effective of CTO tended to decrease and λHU tended to increase
compared with non-calcified plaques, with significant difference of all (P＜0.05). The area under the curve of ROC (AUC) for CT40keV,
iodine density, Z-effective, and λHU in distinguishing non-calcified plaques from CTO were 0.71 (95% CI 0.60-0.79), 0.74 (95% CI
0.52-0.725), 0.70 (95% CI 0.59-0.79), and 0.63 (95% CI 0.53-0.73), respectively. The AUC for the combination of these parameters
was 0.87 (95% CI 0.727-0.929) for distinguishing non-calcified plaques from CTO.
Conclusion: DLS-CT quantitative parameters have the potential to identify the composition characteristics of non-calcified plaques
and CTO, providing new imaging indicators for understanding the pathological changes and progression mechanism of PAD.
Ethics committee - additional information: The study was approved by the Ethics Committee of the First Affiliated Hospital
(ethics approval number: XJTUIAF2021LSY-223).
Clinical value of identifying noncalcified atherosclerotic plaques using effective atomic number maps and electron
density maps derived from non-contrast dual-layer spectral-detector CT (7 min)
Fan Xu; Guangzhou / China
580
Author Block: F. Xu, C. Y. Xiong, C. X. Pei, Y. Liao, X. Zeng; Guangzhou/CN
Purpose: This study aimed to evaluate the clinical feasibility of effective atomic number (Zeff) maps and electron density (ED) maps
derived from non-contrast enhanced (NCE) dual-layer spectral-detector computed tomography (DLCT) in identifying non-calcified
atherosclerotic plaque.
Methods or Background: This retrospective study included 205 patients with 386 noncalcified atherosclerotic plaques confirmed by
contrast-enhanced DLCT (CE-DLCT). Zeff maps, ED maps, and conventional images (CI) were reconstructed from both NCE-DLCT and
CE-DLCT scans. The presence of plaque on Zeff maps, ED maps, and CIs from NCE-DLCT scans was independently assessed by two
radiologists. CT attenuation, Zeff values, and ED values of plaque and blood reconstructed from NCE-DLCT were recorded. The
contrast-to-noise ratio (CNR) of plaques using CT attenuation, Zeff values, and ED values was calculated and compared. Finally,
interobserver agreement was assessed.
Results or Findings: A total of 254 of 386 plaques (65.8%) were identified on Zeff and ED maps from NCE-DLCT scans, whereas only
38 plaques (9.8%) were identified on CI. For all 386 plaques, there were no significant difference in CT attenuation, Zeff value, and ED
value between blood and plaque on NCE-DLCT scans (42.53 vs. 35.14 HU; P=0.18; 7.32 vs. 7.31, P=0.71; 6.52 vs. 6.48, P=0.84).
However, the CNR of Zeff and ED maps was significantly higher than the CIs obtained from NCE-DLCT scans (2.78 vs. 1.12, P<0.05;
2.12 vs. 1.12, P<0.05). Inter-reviewer agreement was good (ICC =0.86 and 0.91).
Conclusion: Zeff and ED maps derived from NCE-DLCT scans provide a potentially feasible method for identifying non-calcified
atherosclerotic plaques, which may be clinically useful in screening asymptomatic high-risk patients.
Limitations: An identified limitation was that this was a retrospective, single-centre study.
Ethics committee - additional information: This study was approved by an ethics committee. The necessity to obtain informed-
consent was waived.
Voxel-based calcified plaque analysis: a comparison between a novel silicon-based photon-counting CT prototype, a
conventional energy-integrating detector CT and µCT (7 min)
Emma Verelst; Brussels / Belgium
Author Block: E. Verelst1, G. Van Gompel1, B. Keelson1, J. De Mey1, D. Crotty2, P. D. Deak3, N. Buls1; 1Brussels/BE, 2Cork/IE, 3Ittigen/CH
Purpose: The study aimed to investigate the accuracy of a second-generation silicon-based photon-counting CT (Si-PCCT) prototype
in a voxel-wise evaluation of calcified plaques, compared to conventional energy-integrating detector CT (EIDCT) with high-resolution
micro-CT as a reference.
Methods or Background: Three human-resected arteries were embedded in a 2% agar-water phantom (d=20 cm). Helical scans
were acquired using a second-generation Si-PCCT prototype (GE Healthcare, Milwaukee, USA) and a conventional EIDCT system (GE
Revolution, Milwaukee, USA) at similar scan parameters (120 kV, 40 mm collimation, 0.9 pitch, 1s rotation and 17 mGy CTDIvol).
Images were reconstructed using a bone kernel, 1024 matrix, 150 mm field-of-view and slice thickness of 0.42 mm (Si-PCCT) and 0.63
mm (EIDCT). Micro-CT images were acquired with a benchtop system (X-cube, Molecubes) using a spiral high-resolution acquisition
protocol at 0.075 mm xy-pixel size and 0.2 mm slice thickness. Plaques (n=7) were analysed by a voxel-wise computation of plaque
volume (mm3), plaque burden (%) and calcified plaque volume (%), i.e. voxels with CT-values >1000 HU. Micro-CT images were used
as a reference. The differences between the two systems were tested with a paired sample t-test.
Results or Findings: Overall, Si-PCCT allowed for more accurate volume measurements, as the mean error towards the reference
volume (12.2 mm3) was significantly reduced when compared to EIDCT (27 mm3), p=0.03. The mean error towards the reference
plaque burden was also found to be significantly lower for Si-PCCT (6.6%) when compared to EIDCT (11.4%), p<0.001. Si-PCCT also
allowed for a significant increase in detection of heavily calcified regions (60.5%), when compared to EIDCT (38.6%), p=0.01.
Conclusion: This study demonstrates an improved spatial resolution of a second-generation Si-PCCT prototype, allowing for an
improved voxel-wise analysis on plaque volume, burden and composition, when compared to conventional EIDCT.
Limitations: The study was an ex vivo phantom study.
Funding for this study: This study was funded by the Radiology Department of the University Hospital of Brussels (UZ Brussel).
Ethics committee - additional information: Ethical approval was obtained for the use of three human arteries (carotid, femoral
and iliac).
Role of dual-energy computed tomography (DECT) in detection of carotid artery monosodium urate deposition in
patients with gout (7 min)
Muhammad Danish Sarfraz; Vancouver / Canada
581
Author Block: M. D. Sarfraz, L. Treanor, S. Masood, N. Murray, A. Sheikh, S. Nicolaou; Vancouver, BC/CA
Purpose: The study aimed to find out if a dual-energy CT can detect monosodium urate deposition in carotid arteries and whether
the presence of monosodium urate crystals has any effect on atherosclerotic disease in terms of plaque volume.
Methods or Background: This is a retrospective study. All patients who underwent any dual-energy neck imaging (like carotid
angiogram, neck soft tissues or cervical spine) from January 2015 to December 2022 were included.
All patients’ charts were rigorously reviewed to find gout patients and avoid confounding variables. The same number of healthy
controls (without a clinical history of gout or other rheumatic diseases) were included during the same study period. These healthy
controls were then age (within 5 years), sex and confounders-matched to gout patients.
DECT datasets were post-processed using gout application and a volumetric analysis of atherosclerotic plaques was performed using
a calcium scoring application in matched gout patients with carotid monosodium deposition, gout patients without carotid
monosodium deposition and control/non-gout patients.
Results or Findings: Out of a total 2157 patients who underwent dual-energy neck imaging during the study period, 85 were
confirmed gout cases. From the 85 gout cases, two were excluded due to presence of streak artefacts from dentures, hence n=83.
Out of the 83, monosodium urate deposition was detected in the carotid arteries of 10 patients (12%).
None of the matched control patients demonstrated monosodium crystal deposition.
Volumetric analysis of atherosclerotic plaques demonstrated larger plaque volumes in gout patients with monosodium urate
deposition (n=10) than matched gout patients without monosodium urate deposition (n=10) and matched control/non-gout patients
(n=10) (p value of 0.03).
Conclusion: Carotid arteries are not a common site for monosodium urate deposition, but if present can lead to increased
atherosclerosis in the involved vessel, resulting in increased incidence of TIA/strokes.
Ethics committee - additional information: Ethics approval was obtained before starting the project.
Correlation between carotid NASCET-stenosis and plaque quality as characterised by the new Plaque-RADS score (7
min)
Gian Luca Chabert; Cagliari / Italy
Author Block: G. L. Chabert1, R. Cau1, A. Schindler2, G. Grassi1, M. Porcu1, T. Saam3, L. Saba1; 1Cagliari/IT, 2Munich/DE, 3Rosenheim/DE
Purpose: The study aimed to investigate the correlation between the degree of carotid plaque stenosis, as measured by the NASCET
method, and plaque composition as categorised by the new Plaque-RADS grading system.
Methods or Background: One hundred and twenty subsequent patients (61 males, 59 females; mean age 67.9±18.61 SD) who
were admitted for suspected cerebral ischaemic disease (CVD) and underwent CT-angiography of the carotid arteries were included in
the study. Both carotid arteries were evaluated for the presence and quality of atherosclerotic plaque via the recently introduced
Plaque-RADS score, and the degree of carotid stenosis was quantified according to NASCET.
Results or Findings: In carotid vessels of the right side the average degree of stenosis increased with increasing lesion severity
from 0±SD0% for Plaque-RADS 1 (n=41), 8.6±9.8SD% for Plaque-RADS 2 (n=36), 27.0±10.4SD% for Plaque-RADS 3 (n=33), to
46.4±23.8SD% for Plaque-RADS 4 (n=10). The same increase was present for the left side from 0±SD0% for Plaque-RADS 1 (n=41),
9.6±10.3SD% for Plaque-RADS 2 (n=25), 25.7±13.2SD% for Plaque-RADS 3 (n=45), to 62.4±30.2SD% for Plaque-RADS 4 (n=9). A t-
test was conducted with a significant difference result of <0.005. Pearson correlation was also used (r=.807 for right stenosis with
right plaque-RADS values; r=.763 for the left side), when considering only plaque-RADS values of 3 and 4 (r=.505 for right side;
r=.622 for left side).
Conclusion: Plaque-RADS scores positively correlate with degree of carotid stenosis, especially at lower lesion grades. On the
contrary, our results also show that already in vessels with low grade stenosis, features of high risk plaques may be encountered and
easily described in a standardised fashion using the novel Plaque-RADS score.
Limitations: The study's data collection was confined to a single institution and evaluated with a new, yet to be completely
confirmed, scoring method.
Imaging of the aorta in dual-source photon-counting CT: impact of low energy virtual monoenergetic imaging on image
quality, vascular contrast and diagnostic assessability (7 min)
Ibrahim Yel; Frankfurt a. Main / Germany
582
Author Block: I. Yel1, C. Booz1, V. Koch1, L. D. Gruenewald1, L. S. Alizadeh1, S. Martin1, T. J. Vogl1, S. Waldeck2, D. P. Overhoff2;
1
Frankfurt/DE, 2Koblenz/DE
Purpose: To evaluate the impact of low energy VMI+ reconstructions on quantitative and qualitative image quality, vascular contrast
and diagnostic assessability of the aorta in patients undergoing photon-counting CTA.
Methods or Background: A total of 125 patients (69 male) who had undergone dual-source photon-counting CTA scans of the aorta
were retrospectively analysed in this study. Standard 120 kV CT images and low keV VMI+ series from 40 to 100 keV with an interval
of 15 keV were reconstructed. Quantitative analyses included evaluation of vascular CT numbers, SNR and CNR. CT number
measurements were performed in the ascending aorta, the aortic arch, the thoracic and infrarenal descending aorta. Qualitative
analyses were performed by three board-certified radiologists independently using five-point scales to evaluate image quality,
vascular contrast and diagnostic assessability.
Results or Findings: Mean attenuation, CNR and SNR values were highest in 40 keV VMI+ reconstructions (HU, 1312 ± 13; CNR, 32
± 8; SNR, 34 ± 10) followed by 55-keV VMI+ reconstructions (HU, 731 ± 9; CNR, 24 ± 6; SNR, 27 ± 9); all three mean values at these
keV levels were significantly higher compared with the remaining VMI+ series and standard 120 kV CT series (HU, 160 ± 8; CNR, 18 ±
5; SNR, 26 ± 6) (p<.0001). The qualitative analysis showed the highest rating scores for 55 keV VMI+ reconstructions followed by 40
keV and 70 keV VMI+ series with a significant difference compared to standard 120 kV CT images regarding image quality, vascular
contrast and diagnostic assessability of the aorta (p<.0001).
Conclusion: Low keV VMI+ reconstructions at a level of 40-55 keV significantly improve image quality, vascular contrast and
diagnostic assessability of the aorta compared with standard CT series in photon-counting CTA.
Limitations: The single-centre retrospective study design was an identified limitation.
583
RPS 1411 - Neuroimaging in white matter disease
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Imaging Methods, Neuro, Research, Vascular
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Radosław Zawadzki; Białystok / Poland
A novel 7T MRI approach to detect posterior fossa demyelinating lesions in patients with multiple sclerosis (7 min)
Lela Okromelidze; Jacksonvill / United States
Author Block: E. Middlebrooks, V. Patel, X. Zhou, S. Straub, L. Okromelidze, V. Gupta, S. J. Singh Sandhu, S. Tao; Jacksonville, FL/US
Purpose: Detection of infratentorial demyelinating lesions in multiple sclerosis (MS) presents a challenge in MRI. The goal of this
study is to assess the efficacy of a novel MRI approach, lesion-attenuated MPRAGE acquisition (LAMA) in detection of demyelinating
lesion within the posterior fossa and upper cervical spine in patients with MS on 7T MRI.
Methods or Background: A retrospective cross-sectional study of 42 patients with MS who had 7T brain imaging with LAMA, 3D DIR
and 2D T2-weighted turbo spin echo sequences was performed. Three neuroradiologists assessed lesion count in the brainstem,
cerebellum, and upper cervical spinal cord using both DIR and T2-weighted images in one session, and LAMA in a separate one.
Additionally, contrast-to-noise ratio (CNR) between LAMA and the conventional sequences was compared.
Results or Findings: LAMA identified more lesions than DIR+T2 (6.4 versus 3.0; p<0.001), exhibited better interrater agreement
(ICC = 0.75 [95% CI = 0.41 – 0.88] versus 0.61 [95% CI = 0.35 – 0.78]), and higher contrast-to-noise ratio (3.7 ± 0.9) versus that of
DIR (1.94 ± 0.7) and T2 (1.2 ± 0.7) (all p < 0.001). In cases with no lesions detected using DIR+T2, at least one lesion was identified
in 83.3% with LAMA. Across all analysed brain regions, LAMA consistently detected more lesions than DIR+T2.
Conclusion: LAMA has demonstrated improved detection of infratentorial demyelinating lesions in patients with MS compared to
traditional methods at 7T through high spatial resolution, robust lesion contrast, and resilience against prevalent 7T MRI artefacts
such as B1+transmit heterogeneity. Integrating LAMA with standard MP2RAGE provides an additional tool for accurate
characterisation of the extent of MS.
Limitations: AMA is limited to 7T. Further investigation is needed to assess its comparative advantage over other sequences at the
reduced field strength.
Ethics committee - additional information: This retrospective study was approved by Mayo Clinic Institutional Review Board.
Comparative analysis of three commercially available artificial intelligence software programmes for brain volumetry
and lesion segmentation in multiple sclerosis and dementia (7 min)
Antonio Di Gioia; L'Aquila / Italy
584
Author Block: A. Di Gioia, F. Colarieti, P. Badini, C. Santobuono, I. Antonio, S. Martino, F. Bruno, A. Splendiani, E. Di Cesare;
L'Aquila/IT
Purpose: The aim of this study was to compare the characteristics, results and experience by using three commercially available
software programmes for segmentation and counting of multiple sclerosis lesions and measurement of dementia brain volume
assessment.
Methods or Background: Brain MRI with standard imaging protocol and including volumetric FLAIR and T1 sequences were
performed in 28 patients (11 famales and 17 males) affected by multiple sclerosis (13 patients) and dementia (15 patients). MRI scans
were segmented twice in each software package by two investigators. Intra-rater, inter-rater and between-software agreement was
compared.
Results or Findings: The main differences were found in the segmentation of white matter lesions, in particular the recognition of
hypointense lesions in T1 that were evidenced from each of the three software used. Furthermore differences were found in both total
lesion volume and number of lesions, with a median difference mostly shown in the evaluation and recognition of periventricular
lesions (-0.39ml; 95% CI from -1.745 to 1.0). Concerning volume calculation, the inter-software and interobserver analysis shows the
main difference being in the evaluation of the frontal (-49.3ml; 95% CI from -60.9 to -37.8) and parietal region volume (-46.8; 95% CI
from -51.2 to 42.4).
Conclusion: The use of artificial intelligence software is becoming an essential part of clinical evaluation in medicine, and in our
specific case in the evaluation of multiple sclerosis lesions and measurement of dementia brain volume assessment. Despite
technological progress, according to our clinical study, there doesn’t yet seem to be a high level of agreement within and between the
software in terms of intra-rater, inter-rater and intra-software programme differences. Therefore, users should be aware of the lack of
interchangeability between these software programmes when they are applied in clinical practice.
Limitations: The low study sample and the fact this this was a retrospective analysis were identified as limitations.
Ethics committee - additional information: All subjects gave their informed consent for inclusion before they participated in the
study. The study was conducted in accordance with the Declaration of Helsinki.
Effect of AI supported lesion detection in MS patients on reporting times (7 min)

Sönke Peters; Altenholz / Germany
Author Block: S. Peters1, G. Kellermann2, J. Watkinson1, M. Huhndorf1, K. Stürner1, O. Jansen1, N. Larsen1; 1Kiel/DE, 2Hamburg/DE
Purpose: Evaluation of MR images of patients with multiple sclerosis (MS) is part of daily clinical routine. Consequently, a growing
number of companies offer software for automated lesion detection. The aim of this study was to evaluate the effect of such software
on the radiological reporting.
Methods or Background: Four radiologists retrospectively evaluated MRI examinations of 50 MS patients and counted the lesions in
FLAIR images separated by the locations periventricular, cortical/juxtacortical, infratentorial and unspecific. After six weeks they
repeated the evaluation, this time additionally using the AI based software mdbrain. In both settings the required time for lesion
detection was documented. Furthermore, the four radiologists evaluated follow-up MRIs of 50 MS patients concerning new and
enlarging lesions in the same manner.
Results or Findings: To determine the lesion-load the average reporting time decreased from 286.85 seconds to 196.34 seconds
(p>0.001). For the evaluation of the follow-up images the average reporting time dropped from 196.17 seconds to 120.87 seconds
(p<0.001). The interrater reliabilities showed no remarkable change. The interclass correlation coefficient (ICC) was 0.83 for the
determination of the lesion-load without software support and 0.84 with software support. For the follow-up evaluation, the ICC was
0.92 without software support and 0.82 with software support.
Conclusion: Increasing workload is a growing problem in daily clinical routine. Usage of AI based support for image-interpretation
can be helpful in clinical routine and significantly lowers reporting times in a system with already limited resources. The interrater
reliability was not remarkably affected be the usage of the software, probably because of the already high agreement without
software support.
Limitations: With these results, a direct conclusion regarding the effect on the reporting quality cannot be made.
Ethics committee - additional information: The study was approved by the ethics committee of the medical faculty of the
Christian-Albrechts-University (CAU) Kiel.
Brain white matter microstructural alterations in Behcet’s patients correlate with cognitive impairment: a diffusion
tensor imaging (DTI) study (7 min)
Osman Aykan Kargin; Istanbul / Turkey
585
Author Block: O. A. Kargin, S. Arslan, U. Uygunoglu, B. C. Poyraz, B. Korkmazer, E. Seyahi, O. Kizilkilic; Istanbul/TR
Purpose: Behcet's syndrome (BS) is a chronic, recurrent, multisystemic vasculitis of unknown aetiology. Neuroaxonal damage in its
neurological involvement, known as Neuro-Behcet's syndrome (NBS), can lead to declines in neurocognitive functions, contributing to
disease-related morbidity. Our study aims to evaluate the microstructural integrity of cerebral white matter tracts in patients with
NBS and BS without neurological involvement via DTI, correlate DTI data with neurocognitive functions, and explore clinical-
anatomical relationships to identify potential neuroimaging biomarkers.
Methods or Background: The study comprised 34 NBS patients and 32 BS patients without neurological involvement, identified
based on the Behcet's Syndrome International Study Group (ISG) and the International Consensus Recommendation (ICR) criteria.
Additionally, 33 age- and educational status-matched healthy individuals were included as controls. All participants underwent DTI, as
well as standardised neuropsychological tests assessing various cognitive domains, including attention, memory, verbal fluency,
abstraction, executive control, visuospatial skills, and sensorimotor performance. DTI data were analysed using tract-based spatial
statistics (TBSS) and automated probabilistic tractography to investigate intergroup differences. Subsequently, the correlations
between DTI parameters of white matter tracts and neuropsychological test scores were examined.
Results or Findings: DTI revealed decreased fractional anisotropy and increased radial diffusivity, mean diffusivity, and axial
diffusivity in both supratentorial and infratentorial white matter in NBS patients, indicating widespread loss of microstructural
integrity. Moreover, this loss of integrity was also observed in BS patients without overt neurological involvement, albeit to a more
limited extent. In NBS patients, specific white matter tracts, including superior thalamic radiation, cingulum, and fornix, were
associated with poor cognitive performance across multiple domains.
Conclusion: Our results suggest that DTI findings can potentially serve as biomarkers to assess neurocognitive impairment
associated with central nervous system involvement in BS.
Limitations: This is a cross-sectional study without longitudinal follow-up.
Funding for this study: Funding was provided by the Scientific Research Projects Coordination Unit of Istanbul University-
Cerrahpasa. Project number TTU-2022-36692.
Ethics committee - additional information: The study was approved by Istanbul University-Cerrahpasa Clinical Research Ethics
Committee (reference number 162).
Usefulness of automatic fully convolutional neural network algorithm (TractSeg) in DTI determining cerebral white
matter integrity in patients with primary Sjögren's syndrome (7 min)
Michał Sobański; Wrocław / Poland
Author Block: M. Sobański1, M. Gajowczyk1, P. Rygiel2, A. Korbecki1, J. Korbecka1, M. Sobańska1, A. Sebastian1, J. Bladowska1;
1
Wroclaw/PL, 2Enschede/NL
Purpose: Primary Sjögren's syndrome (pSS) is an autoimmune disease in which central nervous system (CNS) involvement may
occur. Incidence and pathogenesis of changes related to CNS in pSS is not well understood. Diffusion tensor imaging (DTI) can be
used to demonstrate impaired integrity of cerebral white matter (CWM), especially by quantitative values such as fractional
anisotropy (FA).
Methods or Background: A study group containing 33 patients with primary Sjögren's syndrome and a control group containing 26
healthy patients were studied by performing DTI on a 3T MRI scanner. Data was reprocessed by automatic TractSeg algorithm, which
is a fully convolutional neural network (FCNN), resulting in a segmentation of 72 white matter (WM) tracts per patient, as well as a
value of FA for each tract.
Results or Findings: Reduction in FA in the study group relative to the control group was found in multiple CWM areas. Previously
unreported involvement of the cerebellar WM tracts and optic radiations were found. There were also numerous correlations between
reduced FA values in white matter tracts and rheumatological factors.
Conclusion: DTI examination is a sensitive advanced MRI technique for detecting impaired integrity of cerebral WM tracts in patients
with pSS. The greatest reduction in FA in the study group was found in the cerebellar peduncles, which is also a novel observation.
Other WM tracts also showed reduction in FA. Numerous associations between neuroradiological findings and rheumatological factors
have been demonstrated. The utility of automated methods of collecting measurements in the DTI, such as TractSeg's FCNN, allows
for a rapid and reproducible collection of qualitative and quantitative data.
Limitations: Identified limitations were the (1) small study group, and (2) lack of DTI study at the moment of diagnosis of pSS.
was obtained from all participants. Opinion number of the Bioethics Committee at the Medical University of Wroclaw - KB - 578/2020.
Quantitative assessment of lesion-cerebral cortex signal ratio in FLAIR sequence for discriminating white matter
lesions from non-specific T2 hyperintensities in systemic lupus erythematosus (7 min)
Mehmet Alperen Tezcan; Malatya / Turkey
586
Author Block: R. Akdoğan1, M. A. Tezcan2, Z. Tüzün1, S. A. Atalay1, K. Kaşali1, G. Polat1, M. Alkan Melikoğlu1; 1Erzurum/TR, 2Malatya/TR
Purpose: In our study, we aimed to determine the diagnostic value of measuring the lesion-cortex signal ratio (LCSR) in the FLAIR
sequence for distinguishing T2 hyperintense white matter lesions related to SLE from non-specific T2 hyperintensities, with the
intention of providing guidance to clinicians in the diagnosis and treatment process of the disease.
Methods or Background: A total of 30 SLE patients with white matter hyperintensities and without additional conditions were
included in the study. A control group of 30 patients with non-specific hyperintensities in the white matter in the FLAIR sequence was
selected. LCSR measurements were performed twice at different times by two radiologists. The data are presented as mean and
standard deviation, and the Mann-Whitney U test was used to compare LCSR means between SLE patients and the control group. ROC
analysis was conducted to assess whether LCSR measurement can be used to differentiate SLE white matter lesions from non-specific
hyperintensities.
Results or Findings: The Cronbach's alpha value between the measurements of the two observers at two different time points was
found to be 0.971 and 0.946, respectively, with a highly significant agreement value of 0.959 between them. The LCSR value for SLE
patients was measured as 1.30±0.12, while it was significantly higher at 1.13±0.10 for the control group (p<0.001). In the ROC
analysis conducted to determine whether LCSR measurement can be used to differentiate SLE white matter lesions from non-specific
hyperintensities, the area under the curve was calculated as 0.872±0.043, and it was found to be statistically significant (p<0.001).
Consequently, the LCSR cut-off value was determined as 1.1939, with a sensitivity of 0.836 and a specificity of 0.833.
Conclusion: Measurement of LCSR in the FLAIR sequence allows for a high-precision differentiation between white matter lesions
associated with SLE and non-specific T2 hyperintense lesions. Furthermore, the excellent interobserver agreement in LCSR
measurements attests to its reproducibility.
Limitations: The retrospective design of this study was an identified limitation.
Ethics committee - additional information: This study received ethics committee approval.
Relationship between nutritional status and white matter integrity in the geriatric population: a retrospective DTI
Study (7 min)
Naz Atila; Istanbul / Turkey
Author Block: B. Atasoy Badur, N. Atila, P. Soysal, O. Ipar, A. Kaya, S. Balsak, A. Alkan; Istanbul/TR
Purpose: Malnutrition is described as the clinical condition of decreased nutrient intake and/or inadequate nutrient absorption. It
may result in adverse effects in geriatrics. The aim of this study is to investigate whether the nutritional status in the geriatric
population is associated with the integrity of white matter tracts by using DTI.
Methods or Background: This is a retrospective cross-sectional study with a total of 224 geriatric patients. Mini-nutritional
assessment (MNA) was utilised to detect malnutrition risk. Routine brain MRI protocol and DTI were performed for all patients. 18
different white matter tracts in both cerebral hemispheres were evaluated by ROI-based approach. The MD, FA, AD and RD values
were recorded.
Results or Findings: According to the MNA: 86 patients have normal nutritional status, 107 patients are with malnutrition risk and
31 patients are diagnosed with malnutrition. There is a significant difference between these three groups in terms of age and lab
values of folate. There is no statistically significant difference between the groups in terms of gender, co-morbid disease and
laboratory values of vitamin B12, vitamin D or thyroid function tests. Regarding the DTI values, the most affected white matter tracts
are cingulum (CG), genu of corpus callosum, forceps minor, anterior limb of internal capsul, superior (SCP) and middle cerebellar
peduncle. After adjusting for the folate and age on the effect of DTI parameters, CG and SCP are still affected.
Conclusion: The findings indicate that malnutrition in geriatric patients may be associated with deteriorated DTI values of white
matter tracts, especially in cingulum and superior cerebellar peduncle, which might be considered as an indicator of impaired white
matter integrity.
Limitations: The ROI-based approach is operator-dependent.
Ethics committee - additional information: This study was approved by the local ethics committee.
Cerebral microstructural white matter changes measured by diffusion tensor imaging in asymptomatic carotid
endarterectomy patients: a prospective study into predictors and correlation with vascular anatomy (7 min)
Csongor Péter; Budapest / Hungary
587
Author Block: C. Péter, B. Szilveszter, M. Vecsey-Nagy, S. Borzsák, Z. Mihály, Z. Czinege, A. Varga, P. Sótonyi; Budapest/HU
Purpose: Diffusion tensor imaging (DTI) metric changes were assessed in a prospective cohort of asymptomatic carotid
endarterectomy (CEA) patients in correlation with patient characteristics, and vascular anatomy (4-axis; circle of Willis [CW] scores).
Methods or Background: Prospectively enrolled patients (54, males 54%, age 69.7±7 years) had DTI pre-CEA (mean 2, range 0-31
days) and post-CEA (mean 2, range 0-5 days). DTI metrics (FA–fractional anisotropy, axial, mean, radial diffusivities – AD, MD, RD)
were calculated for 16 white matter (WM) regions (8-8 regions of index and contralateral hemispheres) pre-CEA and post-CEA and
compared using the repeated measure Wilcoxon signed rank test (p<0.0125, Bonferroni corrected). Preoperative supraaortic plus
intracranial CTA was performed to create the 4-axis score (range 0-15,mthe lower the worse) and CW score (0–5, 0 if both the anterior
plus ispilateral posterior semicircles towards the operated internal carotid artery are incomplete). Univariate and multivariate linear
regression analyses were used to correlate main demographic, comorbidity data, vascular scores (p<0.05).
Results or Findings: Significantly decreased FA in 1/8 index WM region (p=0.007), singificantly increased AD in 7/8 index, 4/8
contralateral WM regions (p<0.0001-0.0120), significantly increased MD in 6/8 index, 4/8 contralateral WM regions (p<0.0001-0.005),
significantly increased RD in 5/8 index, 3/8 contralateral WM regions (p<0.0001-0.011) was found. In 19/26 regions with significant
DTI metric changes from pre-to-post-CEA, the 4-axis score was a significant predictor using univariate regression analysis
(p<0.001-0.038), remaining significant with multivariate regression analysis in the index temporal, contralateral frontal, parietal WM
regions (p=0.006-0.012) for MD, and in the index insular (p= 0.042) and temporal WM regions (p=0.039) for RD.
Conclusion: DTI metric changes suggesting microstructural deterioration were found from pre-CEA to post-CEA, more pronounced on
the index side, with the weighted 4-axis score as a significant predictor.
Limitations: Identified limitations were (1) the small cohort and (2) the lack of correlation with surgical/anaesthesiology data.
Funding for this study: Funding was received from the National Research, Development and Innovation Office – NKFIH, K 129277.
Ethics committee - additional information: This study was approved by the institutional ethical committee with approval code:
IV/667-1/2022/EKU.
588
RPS 1404 - From diagnosis to prognosis: chest imaging
Categories: Artificial Intelligence & Machine Learning, Chest, Imaging Methods, Vascular
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Galit Aviram; Tel-Aviv / Israel
AI vs senior radiologists in detecting thoracic abnormalities on chest radiographs compared to CT (7 min)

Souhail Bennani; Paris / France
Author Block: S. Bennani1, J. Ventre1, V. Marty1, E. Lacave1, C. Carrière1, D. Hayashi2, A. Kompel3, A. Gupta3, A. Guermazi4; 1Paris/FR,
2
Stony Brook, NY/US, 3Boston, MA/US, 4West Roxbury, MA/US
Purpose: The study aimed to assess the diagnostic performances of an artificial intelligence (AI) software in the detection of thoracic
abnormalities on chest radiographs compared to senior radiologists.
Methods or Background: We collected 319 chest radiographs of patients above 22 years old who underwent thoracic CT within 72
hours. A senior chest radiologist annotated the radiographs for four abnormality types (pleural abnormality, consolidation,
mediastinal-hilar abnormality, nodule) using CT findings as the ground truth. Three senior radiologists independently analysed the
dataset, knowing clinical indications without CT access. Discrepancies were resolved by consensus.
The AI (ChestView, Gleamer), a deep learning algorithm that detects the four abnormalities, was compared against the radiologists
and their consensual analysis for sensitivity and specificity.
Results or Findings: The dataset included 168 radiographs (age: 64±16 years, 91 women): 129 with at least one abnormality, 39
without any abnormality.
For consolidation, the sensitivities were 72% for AI, 54%, 80%, and 66% for the individual readers, and 71% for consensus, with
specificities of 92% for AI, 80%, 77%, 85% for the readers, and 92% consensus. The sensitivities for mediastinal-hilar abnormalities,
were 54% (AI), 43%, 27%, 48% (readers), and 54% (consensus); specificities were 95%, 88%, 96%, 93%, and 94%, respectively. For
nodules, the sensitivities were reported as 67% for AI, 57%, 55%, 55% for the readers, and 62% consensus, with specificities of 89%,
52%, 88%, 81%, and 86%, respectively. Lastly, for pleural abnormalities, the sensitivities were 84%, 89%, 87%, 73%, and 83%, and
the specificities were 93% for AI, 88%, 92%, 95% for the readers, and 94% for consensus.
Conclusion: The AI consistently matched or exceeded senior radiologists in detecting thoracic abnormalities.
Limitations: The dataset had few radiographs with high pathology prevalence.
Funding for this study: Funding for this study was received from Gleamer.
Ethics committee - additional information: The study was approved by WCG (number IRB00000533).
Detection of incidental pulmonary emboli in conventional contrast-enhanced chest CT: application of deep learning-
based augmented contrast enhancement and de-noising algorithms (7 min)
Seunghyun Song; Seoul / Korea, Republic of
589
Author Block: S. Song, E-S. Cho, T. H. Kim, C. Ahn, J. H. Kim; Seoul/KR
Purpose: The objective of this study was to investigate the feasibility of deep learning-based augmented contrast enhancement (DL-
ACE) and deep learning-based de-noising (DL-DN) algorithms to diagnose incidental pulmonary emboli (PE) in routine contrast-
enhanced chest CT.
Methods or Background: This study included 1081 routine contrast-enhanced chest CT scans. The commercial DL-ACE (ClariACE,
ClariPi) algorithm provided iodine-contrast enhancement boosting by 100%, and DL-DN (ClariCT.AI, ClariPi) algorithm decreased the
image noise. The performances of CT images applying DL-ACE and DL-DN were compared with that of conventional CT images based
on the vascular attenuation value, image noise and contrast-to-noise ratio (CNR) of the pulmonary arteries, subjective image
qualities, and the detection rate of incidental PE.
Results or Findings: DL-ACE significantly increased the attenuation value of the pulmonary arteries from 186.5 ± 50.2 HU to 336.9
± 99.6 (p< 0.001), while DL-DN significantly decreased the image noise from 24.3 ± 6.2 to 19.3 ± 6.6 (p< 0.001). Therefore, the CNR
of pulmonary arteries was significantly improved from 6.3 ± 2.3 to 16.6 ± 6.9 when DL-ACE and DL-DN algorithms were applied.
Subjective scores of the vascular enhancement and image noise were significantly better on CT images using DL-ACE and DL-DN (p<
0.001). Incidental PE was present in 83 studies (7.7%). DL-ACE and DL-DN showed an additional 3 segmental and 2 sub-segmental PE,
compared to conventional CT. However, there was no significant difference in the detection rate of incidental PE, regardless of
whether DL-ACE and DL-DN were applied.
Conclusion: DL-ACE and DL-DN algorithms significantly improved the vascular attenuation value, image noise, CNR of the pulmonary
arteries, and subjective image qualities, and showed a small incremental benefit for the detection of incidental pulmonary emboli,
compared to routine contrast-enhanced chest CT.
Limitations: This was a retrospective and single centre study.
Ethics committee - additional information: This study was approved by the Institutional Review Board (IRB), and written informed
consent was waived.
Impact of the recent global iodinated contrast agent shortage on positivity rate for pulmonary embolism in CT
pulmonary angiograms and chest CT with contrast at two major US health care systems (7 min)
Axel Wismueller; Pittsford / United States
Author Block: A. Wismueller1, J. P. Kanne2, L. Stockmaster1, E. Weinberg1, D. Shrier1, A. Vosoughi1, A. Kasturi1, N. Hadjiyski1;
1
Rochester, NY/US, 2Madison, WI/US
Purpose: The objective of this study was to quantitatively track the impact of the recent global shortage of iodinated contrast agents
on pulmonary embolism (PE) positivity rate in CT pulmonary angiograms (CTPA) using AI-based image analysis at two major US
healthcare systems.
Methods or Background: Using commercial AI-based image analysis (Aidoc Medical), we analysed daily volumes, PE and incidental
PE (iPE) positivity rates for 7,633 computed tomography pulmonary angiogram (CTPA) and 11,164 contrast-enhanced chest CT
(CT+C) exams before and during the contrast agent shortage (both comprising 01.04.2022 through 01.07.2022). For comparison, we
analysed daily volumes and positivity rates for intracranial haemorrhage (ICH) on 30,803 non-contrast head CT exams during the
same period. We compared two observational periods: a pre-shortage control period (10.04.2022 through 40.04.2022), and a contrast
shortage period (20.05.2022 through 10.06.2022). Percentage change metrics of case volumes and positivity rates for PE, iPE and ICH
were calculated.
Results or Findings: Case volumes of CTPA exams dropped from baseline during the shortage period by 38.9% while PE positivity
rates significantly increased by 31.1% (p< 10^-4, Welch's unequal variances t-test). Similarly, case volume of CT+C exams dropped
during the contrast agent shortage period by 49.9% while iPE positivity rates significantly increased by 20.2%, (p< 10^-4). For
comparison, non-contrast head CT volumes dropped by only 6.5%, and ICH positivity rates increased by only 15.9%, with no
significant difference (p> 0.05).
Conclusion: Our results suggest a significant increase of PE/iPE positivity rates in significantly decreased CTPA/CT+C total exam
volumes during the observed global contrast agent shortage period, while non-contrast head CT exam volumes and ICH positivity
rates remained essentially stable. Our observations can be explained by more restrictive ordering patterns for contrast-enhanced
studies during the shortage period.
Limitations: Ground truth was based on AI image analysis rather than NLP on radiology reports.
Funding for this study: Funding was received from the ACR Innovation Award for AW; Aidoc Medical.
Ethics committee - additional information: This study was approved by URMC and UWHealth institutional IRB approvals.
Correlation between postoperative pulmonary venous congestion on chest radiograph and outcome after cardiac
transplantation (7 min)
Charlotte Sonja Böttger; Düsseldorf / Germany
590
Author Block: C. S. Böttger, C. B. Fink, V. Hettlich, F. S. Jenkins, D. Scheiber, H. Aubin, A. Lichtenberg, S. Reinartz, U. Boeken;
Düsseldorf/DE
Purpose: After heart transplantation (HTX), chest radiography plays an important role in the simple, non-invasive detection of
pathological changes. Given the significant relevance of fluid balance following this procedure, particularly for hemodynamic
regulation, the assessment of potential pulmonary venous congestion (p.v.-congestion) in chest radiographs is crucial. The objective
of this analysis was to determine a potential correlation between early postoperative p.v.- congestion after cardiac transplantation
and subsequent outcomes.
Methods or Background: 302 patients underwent HTX between 09.2010 and 09.2023 in our department. Recipients were
retrospectively divided into 3 groups based on the severity of signs of pulmonary venous congestion on chest x-ray on the first
postoperative day: Group 1 had no signs of p.v.-congestion (n= 115), Group 2 showed signs of mild p.v.- congestion (n= 145) and
Group 3 showed signs of moderate to higher-grade p.v.- congestion (n= 41). One patient was excluded due to intraoperative death.
Results or Findings: The 30-day mortality was 4.5% in group 1, 6.2% in group 2, and 22% in group 3 (p< 0.05). Combined donor
heart/lung organ harvesting was performed significantly more often in group 2 compared to group 3. Intraoperative bypass time and
duration of operation were shorter in groups 1 and 2 compared to group 3. There were no significant differences between groups
regarding graft rejection or infection. Patients in group 3 developed significantly more neurological complications postoperatively. 1-
year and 3-year survival was reduced in patients with signs of moderate to higher grade congestion on X-ray (p> 0.05).
Conclusion: In this analysis, it was demonstrated that the presence of signs of p.v.-congestion and the severity on the first
postoperative day after HTX have an influence on survival. Therefore, a conservative estimate of survival can be made by assessing
the first postoperative chest radiograph.
Comparison of two CT-derived methods to assess alveolar collapse as a potential prognostic imaging marker in
pulmonary fibrosis (7 min)
Hoen-oh Shin; Hannover / Germany
Author Block: S. Scharm1, C. M. Schaefer-Prokop2, A. Schreuder3, J. Fuge1, F. Wacker1, B. Seeliger1, A. Prasse1, H-o. Shin1;
1
Hannover/DE, 2Amersfoort/NL, 3Nijmegen/NL
Purpose: The objective of this study was to evaluate two different CT-derived approaches for the assessment of alveolar collapse, a
potential precursor of pulmonary fibrosis.
Methods or Background: For this single-centre retrospective longitudinal study, 66 consecutive patients with idiopathic pulmonary
fibrosis underwent CT in inspiration and expiration and pulmonary function testing at baseline. The patient population was divided
into two subgroups according to their status at 3 years (death or transplantation versus clinical surveillance). Parametric response
maps were generated as scatterplots of voxel-wise attenuation values of paired inspiration and expiration scans after non-linear
registration. Inspiratory and expiratory attenuation histograms were also generated and analysed. Voxels with an abnormally high
increase in attenuation during expiration were interpreted as "collapsed" lung tissue.
A Mann-Whitney U test was performed to assess the difference in CT-derived measures between the two subgroups and logistic
regression was performed to test the predictive power of FVC and both CT-derived measures (PRM and histogram analyses).
Results or Findings: All CT-derived PRM and histogram measures were significantly different (p < 0.005) between the two patient
subgroups.
With a discriminatory performance of AUC = 0.788, 95% CI 0.679-0.898 for PRM and AUC = 0.791, 95% CI = 0.684- 0.899 these two
CT-derived measures had a superior predictive performance compared to FVC alone with AUC = 0.708, CI 95% 0.581- 0.836.
The advantage of the PRM analysis is a voxel-wise matching between inspiration and expiration due to the required registration. On
the other hand, the histogram parameter approach does not require advanced postprocessing but lacks the direct spatial correlation
of the attenuation change between inspiration and expiration.
Conclusion: Both methods perform similarly well in predicting patient outcome by directly or indirectly quantifying alveolar collapse,
and both outperform FVC.
Ethics committee - additional information: This study was approved by IRB, with the approval code: No. 10726_BO_K_2023
Effect of inspiratory lung volume on bronchial and arterial dimensions on chest CT (7 min)
Yuxin Chen; Rotterdam / Netherlands
591
Author Block: Y. Chen1, R. Latisenko2, D. Lynch3, P. Ciet1, J-P. Charbonnier2, H. A. W. M. Tiddens2; 1Rotterdam/NL, 2Nijmegen/NL,
3
Denver, CO/US
Purpose: Bronchus-artery (BA) dimensions on chest CT are influenced by inspiratory lung volume. We aimed to assess the effect of
inspiratory lung volume on BA-ratios in patients with COPD.
Methods or Background: A selected group participating in the COPDGene study received a full dose (FD, 120kVp, 200mAs) and a
reduced dose (RD, 120kVp 35mAs) CTs in the same imaging session. CTs were analysed using LungQ. For segmental (G0) and distal
generations, the following diameters were measured: bronchial outer (Bout), inner (Bin), wall thickness (Bwt), and artery (A), and the
following BA-ratios were computed: Bout/A, Blumen/A, Bwt/A, and bronchial wall area/bronchial outer area (Bwa/Boa). Total lung
capacity of the CT (TLC-CT) was computed. Differences between the volumes (ΔTLC-CT%) between the two CTs were expressed as %
of the highest TLC-CT. Mixed-effect models were used to investigate the influence of TLC-CT on BA-ratios adjusted for dose protocol.
AUC was used to define a cut-off value for BA-ratios in relation to ΔTLC-CT%.
Results or Findings: 1319 patients with a mean (SD) age of 64.4 (8.7) years were included. 329 (124) BA-pairs could be analyzed
per CT. No significant difference was found for TLC-CT between FD and RD scans. A ΔTLC-CT% of 5% significantly lead to 0.015 and
0.025 decreases in Bout/A and Bin/A and 0.004 and 0.015 decreases in Bwt/A and Bwa/Boa. The cut-off values to determine when BA-
ratios were influenced by lung volume were for Bout/A 4.9%, Bin/A 3.7%, Bwt/A 4.4%, and Bwa/Boa 5.4%.
Conclusion: The BA biomarkers provide a robust quantification of bronchial changes on CT when differences in ΔTLC-CT% are kept
below 5%. Standardising volumes for clinical follow-up and trials is recommended to optimise assessment for tracking airway disease
changes over time.
Limitations: The study examined a selected group of an elderly population,
Funding for this study: This study was funded by the Erasmus MC LungAnalysis supporting grant.
Ethics committee - additional information: The COPDGene study was approved by the institutional review boards at each of the
21 participating clinical sites.
Quantitative CT characterises baseline regional structure and function in idiopathic pulmonary fibrosis patients with
one-year diffusing capacity decline (7 min)
Hongseok Ko; Seoul / Korea, Republic of
1 2 3 2 2 4 2 1 2 3
Author Block: H. Ko , W. C. Chung , J. Choi , S. M. Choi , C-H. Lee , K. J. Chae , C. H. Lee ; Chuncheon/KR, Seoul/KR, Kansas City,
4
KS/US, Jeonju/KR
Purpose: We characterised regional lung structure and function in idiopathic pulmonary fibrosis (IPF) patients with a one-year
diffusing capacity of carbon monoxide (DLCO) decline using quantitative inspiratory and expiratory chest computed tomography (CT)
analysis.
Methods or Background: Baseline and one-year pulmonary function tests (PFTs) and baseline inspiration-expiration CT scans were
collected from 51 IPF patients (71.5±5.9 years). Commercial and in-house quantitative CT (qCT) software were used for the
segmentation and quantification of 113 regional lung structural-functional features. Baseline qCT features were compared between
patients with 5% or more percent-predicted (%pred) DLCO decline (IPF-A) and the rest (IPF-B), using Wilcoxon rank-sum test.
Results or Findings: Compared to IPF-B (n= 25, 1 female), IPF-A (n= 26, 4 females) had less baseline high attenuation area percent
(HAA%) in the right upper (RUL, -32.0%, p= 0.011) and lower (RLL, -29.3%, p= 0.027) lobes and greater anisotropic deformation index
(31.2%, p= 0.030) and relative displacement (26.2%, p= 0.047) in the left lower lobe (LLL). IPF-A also had moderately greater
normalised hydraulic diameter (Dh*) at the RLL segmental airways (sRLL) (9.3%, p= 0.054). Early traction bronchiectasis was
observed in the sRLL of IPF-A. IPF-A showed CT features of greater extent and severity of fibrosis in IPF. Demographics and PFTs were
not significantly different at baseline.
Conclusion: qCT analysis characterised baseline regional lung structure and function in IPF with DLCO decline and supported visual
CT interpretation. Findings suggest that advanced fibrotic changes and RLL segmental airway traction bronchiectasis may be
precursors of DLCO decline presumably with progressive pulmonary fibrosis.
Limitations: This study has a limited number of cases.
Funding for this study: This study was supported by Korea Environmental Industry & Technology Institute (KEITI) grant
2018001360001 funded by the Ministry of Environment (MOE), Republic of Korea.
Ethics committee - additional information: The study protocol was approved by the institutional review boards (IRB) of Seoul
National University Hospital (IRB: 1810-036-977)
Default vs selective application of artificial intelligence to classify viral pneumonia on CT imaging: impact on human
performance (7 min)
Francesco Rizzetto; Rho / Italy
592
Author Block: F. Rizzetto, L. Berta, L. A. Carbonaro, D. Artioli, F. Travaglini, G. Zorzi, A. Torresin, P. E. Colombo, A. Vanzulli; Milan/IT
Purpose: This study aimed to assess the impact of different artificial intelligence (AI)-powered approaches on human performance in
distinguishing COVID-19 from other viral pneumonia on CT imaging.
Methods or Background: Three experienced radiologists blindly evaluated 220 chest CT examinations of patients with viral
pneumonia (n=151 COVID-19; n=69 other viruses), assigning a CO-RADS score before (S1) and after (S2) receiving results from a
validated AI classifier. Inter-reader agreement with Gwet’s agreement coefficient type-2 (AC2) and performance metrics were
calculated for S1 and S2. Two different S2 scenarios were considered: one where the AI output was available for all cases (default
approach), and another where the AI prediction was applied only to CO-RADS=3 cases from S1 (selective approach).
Results or Findings: The readers showed good-to-excellent agreement in assigning CO-RADS for all scenarios (range
AC2=0.77-0.81). On average, CO-RADS changes between S1 and S2 occurred in 18% of cases, with 29% involving patients initially
assigned CO-RADS=3. In these cases, the use of AI output correctly classified 85% of patients. Conversely, when the radiologists were
confident in S1 diagnosis (CO-RADS≠3), class changes in S2 occurred in 7% of cases. This prevented incorrect diagnosis in 45% of
patients but led to a missed correct classification in the remaining 55%. In S1, the readers achieved 78% accuracy, with 15% of
patients classified as CO-RADS=3. In S2, accuracy was 81% with 16% of CO-RADS=3 for the default approach, and 79% with 10% of
CO-RADS=3 for the selective approach. Significant differences were only observed in the proportion of CO-RADS=3 cases in S2 with
the selective approach (p<0.009).
Conclusion: AI helps distinguish COVID-19 from other viral pneumonias in equivocal cases but its reliability diminishes when the
reader is already confident on diagnosis.
Limitations: The main limitation is the retrospective single centre study design.
Ethics committee - additional information: This study received approval from the Local Ethics Committee (decision number:
188-22042020)
593
EIBALL 14 - Clinical validation of DWI as imaging biomarkers
Categories: Imaging Methods, Oncologic Imaging, Physics in Medical Imaging

ETC Level: LEVEL II
Date: March 1, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Nandita Desouza; London / United Kingdom

Problems related to the DWI as a quantitative imaging biomarker (15 min)

1. To learn the potential value of quantitative DWI.

2. To appreciate the steps required to pool data for meaningful interpretation successfully.
3. To understand the limitations of quantitation to use it in AI algorithms.
QIBA Profile on DWI (15 min)

Michael Boss; Boulder / United States
1. To learn the purpose behind the QIBA Profile on DWI.

2. To appreciate the rigour behind the Profile's ADC claim statements.
3. To understand what successful implementation of the Profile can provide users.
How to conform to the QIBA Profile (15 min)

Gudrun Zahlmann; Oak Brook / United States
1. To learn how to implement the requirements of the QIBA Profile in a real setting.
2. To appreciate the challenges and solutions of DWI MR implementation as a quantitative imaging biomarker assessment.
3. To understand the necessary steps to be considered to assess ADC reliably and reproducibly.
Panel discussion: Is the QIBA DWI Profile ready for clinical use? (10 min)
594
MD 8 - Breast cancer: image-guided treatment de-escalation/optimisation - recommendations for

Categories: Breast, Multidisciplinary, Oncologic Imaging, Research

Date: March 1, 2024 | 12:45 - 13:45 CET
CME Credits: 1
Moderator:

1. To critically review novel imaging techniques for breast cancer staging and put their use in perspective to treatment developments.
2. To highlight new developments in the use of imaging for treatment optimisation in breast cancer patients.

Isabel T. Rubio; Madrid / Spain

Willemien Menke-van der Houven van Oordt; Amsterdam / Netherlands

595
CUBE 16 - Hybrid Operating Room: Experiences and Challenges

Date: March 1, 2024 | 13:00 - 13:30 CET
Hybrid Operating Room: Experiences and Challenges (30 min)

1. To outline the experiences of radiographers working in hybrid operating rooms.

2. To discuss the current challenges for radiographers when working in hybrid operating theatres.
3. To explain the benefits to patients and staff from using hybrid operating theatres.
4. To discuss the radiographer's role within the hybrid theatre.
596
RPS 1511 - Parkinson's, movement disorders, spine and nerve imaging
Categories: Artificial Intelligence & Machine Learning, Neuro, Research

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Assessment of nerve root compression and dorsal root ganglia in patients with radiculopathy using 7 Tesla MR imaging
(7 min)
Georg Constantin Feuerriegel; Zurich / Switzerland
Author Block: G. C. Feuerriegel, A. A. Marth, C. Germann, F. Wanivenhaus, D. Nanz, R. Sutter; Zurich/CH

Purpose: This study aimed to evaluate the diagnostic value of 3D dual-echo steady-state (DESS) MR imaging of the cervical spine at
7 T compared to 3 T in patients with cervical radiculopathy.
Methods or Background: Between March 2020 and January 2023, patients diagnosed with cervical radiculopathy were
prospectively recruited prior to surgical decompression and underwent 3D DESS imaging on 3 T and 7 T MRI. Two radiologists
independently assessed cervical nerve root compression and the dimensions of the dorsal root ganglion (DRG). Likert scales were
used to assess signal intensity, visibility of nerve anatomy, diagnostic confidence and image artefacts. The degree of neuroforaminal
stenosis was assessed on standard clinical 3 T scans. Statistical analysis included diagnostic accuracy and inter-reader reliability. The
Wilcoxon signed-rank test was used to assess differences between groups.
Results or Findings: 48 patients (mean age 57±12 years, 22women) were included in the study with the highest prevalence of
severe neuroforaminal stenosis observed at C6 (n=68) followed by C7 (n=43). Direct evaluation of nerve root compression showed
significantly higher diagnostic confidence and visibility of cervical nerve rootlets, roots and DRG on 7 T DESS than on 3 T DESS
(diagnostic confidence P=0.01, visibility: P<0.01). Assessment of nerve root compression using 7 T DESS allowed more sensitive
grading than standard clinical MRI (P<0.01) and improved the performance in predicting sensory or motor dysfunction (AUC
combined: 0.87).
Conclusion: 7T DESS imaging provides direct assessment of cervical nerve root compression in patients with radiculopathy,
predicting sensory or motor dysfunction better than standard clinical MRI. The diagnostic confidence and image quality of 7T DESS
were found to be superior to that of 3 T DESS.
Limitations: 7T MRI is not widely available and only patients with radiculopathy due to degenerative changes were included.
Ethics committee - additional information: This study was approved by the Cantonal Ethics Committee Zurich.
Accuracy of automated AI-driven susceptiblity map-weighted MRI analyses for neurodegenerative Parkinsonism (7 min)
Elon Wallert; Amsterdam / Netherlands
597
Author Block: E. Wallert1, E. Van De Giessen1, M. Beudel1, D. H. Shin2, T. Van Mierlo3, J. Blankevoort4, H. Berendse1, R. De Bie1, J.
Booij1; 1Amsterdam/NL, 2Seoul/KR, 3Haarlem/NL, 4Almere/NL
Purpose: This study aimed to determine the accuracy of AI-driven automated diagnostic susceptibility map-weighted imaging (SMWI)
software in patients who presented with a clinically uncertain Parkinsonian syndrome (CUPS).
Methods or Background: Parkinson’s disease (PD) is characterized by degeneration of the substantia nigra (SN), particularly in the
nigrosome-1 area, and is associated with locally increased iron deposition. The accuracy of a clinical diagnosis is estimated at 80%.
SMWI is a novel MRI sequence that combines magnitude images with quantitative susceptibility mapping to enhance the contrast of
nigrosome-1 and loss of its signal in neurodegenerative Parkinsonian patients due to iron deposition, which may aid diagnostics.
A diagnostic accuracy study was performed in patients who had a dopamine transporter (DAT) SPECT because of CUPS between
January 2019 and July 2023. The index test was the result of AI-driven diagnostic software (Heuron IPD, formerly mPDia) analysis of
the SMWI sequence between May 2022 and September 2023. The reference standard was the result of the DAT SPECT.
Results or Findings: 132 patients were enrolled in this study. Twelve were excluded from the analysis; four because of uncertain
DAT-SPECT results and eight because of severe artefacts on MRI. DAT-SPECT was abnormal in 53 and normal in 67 patients. In this
preliminary analysis, the accuracy of the diagnostic software compared with the DAT-SPECT results was 88% with a sensitivity of 87%
and specificity of 89%.
Conclusion: The diagnostic accuracy of the diagnostic software is promising in a clinically relevant population of patients who
presented with CUPS.
Limitations: A limitation of this study is the delay between the index test and the reference test of 1.6 years. DAT-SPECT is
considered to be a golden standard in the diagnostics of patients with CUPS. However, the follow-up time is limited to 2.3 years.
Funding for this study: Funding for this study was received from Heuron Co.
Ethics committee - additional information: This study was approved by the Amsterdam UMC METC.
Functional brain connectivity in patients with de novo Parkinson's disease based on rs-fMRI (7 min)
Noemi Pucci; Rome / Italy
Author Block: N. Pucci, S. Minosse, E. Picchi, V. Ferrazzoli, F. Garaci, T. Schirinzi, F. Di Giuliano; Rome/IT
Purpose: This study aimed to evaluate a putative reorganisation of brain networks between de novo Parkinson's disease patients and
healthy controls by measures based on graph theory.
Methods or Background: 31 de novo PD patients and 30 healthy controls underwent MRI examination to obtain rsf-MRI and 3D-T1-
MPRAGE data. The rs-fMRI data were pre-processed in FLS. The time series was parcelled into 116 regions using the Automated
Anatomical Labelling Atlas (AAL) per subject. Pearson correlation between all-time series was used to generate subject-based
adjacency matrices. Local nodal measures (degree and centrality), functional integration measures (global efficiency), functional
segregation measures (local efficiency, clustering coefficient, transitivity and modularity), and resilience measures (assortment
coefficient) were calculated using Brain Connectivity. The local measures were analysed by means of disruption index k. We
calculated the hub regions in the two groups using the BC of each node. All local and global variables were compared using the Mann-
Whitney U Test.
Results or Findings: We found no statistically significant differences in measures of global variables between de novo PD and
healthy controls. We did find statistically significant differences due to disaggregation index k in all local metrics. For all statistically
significant comparisons, the disruption index k was lower in the de novo PD group than in the healthy controls.
Conclusion: The alterations found by local network measurements highlight a reorganisation of brain networks in de novo PD
patients, supporting the hypothesis that the analysis of functional connectivity may facilitate a better understanding of the complexity
of PD pathophysiology from the earliest stages of the disease.
Limitations: Our study was performed in a relatively small sample size and didn't include neurocognitive tests, so we cannot
examine putative associations between neurocognitive status and MRI parameters.
Funding for this study: Funding for this study was supported by #NEXTGENERATIONEU (NGEU) and funded by the Ministry of
University and Research (MUR), National Recovery and Resilience Plan (NRRP), project MNESYS (PE0000006) - A Multiscale integrated
approach to the study of the nervous system in health and disease (DN. 1553 11.10.2022).
Ethics committee - additional information: The study protocol was approved by the local Institutional Review Board and adhered
to the tenets of the Declaration of Helsinki. All subjects provided written informed consent.
Diagnostic value of quantitative susceptibility mapping in assessing iron deposition in the basal ganglia of patients
with Parkinson's disease (7 min)
Vahid Shahmaei; Tehran / Iran
598
Author Block: V. Shahmaei, F. Sodaei, F. Faeghi, F. Ashrafi; Tehran/IR
Purpose: This study aimed to assess iron deposition in the basal ganglia using QSM, determine the diagnostic value of this method,
and explore the association between disease stage and QSM values in individual nuclei. Iron deposition in the brain is commonly
associated with Parkinson's disease. Quantitative Susceptibility Mapping (QSM) has proven to be a more sensitive imaging technique
compared to T2-weighted imaging, T2*, and R2. However, there have been limited studies utilising QSM to evaluate iron deposition in
the basal ganglia of Parkinson's disease patients.
Methods or Background: A total of 25 patients were assessed using the Hoehn and Yahr test, which classified them into three
different stages. 15 healthy subjects were included as a control group. Routine MRI sequences and QSM were performed using a
Siemens 3. Tesla scanner. Signal processing and image analysis were conducted using STI Suite software. Susceptibility measures of
all basal ganglia nuclei were extracted individually.
Results or Findings: Significant differences in susceptibility measures were observed in the Substantia Nigra, Red Nucleus,
Thalamic Nucleus, and Globus Pallidus nuclei between the patients and control groups (P-value < 0.001). There was a significant
relationship between disease stage and QSM in the Substantia Nigra, Red Nucleus, and Globus Pallidus nuclei (P-value < 0.05, R=
0.51-0.78). QSM values demonstrated a significant association with disease stage across all nuclei (P-value < 0.001). Furthermore,
QSM exhibited higher accuracy in the Substantia Nigra, Globus Pallidus, Red Nucleus, and Thalamic Nucleus, respectively (Acc=
88%-98%).
Conclusion: Utilising QSM to assess iron deposition in the Red Nucleus, Substantia Nigra, and Globus Pallidus nuclei can aid in the
diagnosis and staging of patients with Parkinson's disease. Future studies focusing on the disease stage could provide further
information.
Ethics committee - additional information: The informed consent was obtained from participants included in the study following
the supervision of the ethics committee of Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Human visual nigrosome analysis improves AI-based diagnoses in neurodegenerative disease (7 min)
Ebru Sayilir; Saint Martin D'Heres / France
Author Block: E. Sayilir1, E. Piot1, C. Bonardel1, F. Renard2, S. Grand1, A. Attye2, A. Krainik1; 1Grenoble/FR, 2La Tronche/FR
Purpose: This study aimed to evaluate the diagnostic performance of BrainGML a manifold-learning AI software (Geodaisics.com) in
neurodegenerative diseases combined with the visual analysis of nigrosome imaging. Neurodegenerative diseases are associated
with regional cerebral atrophy patterns, such as temporohippocampal and parietal atrophy in Alzheimer's disease (AD), and
frontotemporal atrophy in frontotemporal dementia (FTD). Parkinson's (PD) and dementia with Lewy bodies (DLB) have subtle
structural abnormalities such as insular atrophy and nigrosome loss. Artificial intelligence software shows potential for accurate
diagnosis of atrophy patterns, while nigrosome analysis is still unavailable.
Methods or Background: A retrospective study was conducted on patients with AD, FTD, PD, and DLB. BrainGML analysed
cerebral atrophy using 3DT1 images, assigning the highest probability for Normal, AD, FTD, or PD. Nigrosome visual analysis was
performed on susceptibility-weighted images by 4 radiologists who determined whether nigrosomes were normal or abnormal. The
primary outcome was the accuracy of the radiological diagnosis, defined as 'Right', 'Wrong', or 'Undefined' (when nigrosome imaging
was normal in PD and DLB, or when nigrosome imaging was abnormal in AD or FTD).
Results or Findings: The cohort included 79 patients (29 AD, 11 FTD, 26 PD, 13 DLB). Nigrosomes were normal in
AD (100%) and FTD (100%), and were abnormal in PD (92%) and DLB (62%). BrainGML provided 59% Right diagnoses on trained
diagnoses (all but DLB), and 49% on all patients including DLB. Combining nigrosome analysis with BrainGML decreased the ratio of
'Wrong' diagnoses from 51% to 15%, which was replaced by the increase of 'Undefined' diagnoses from 0 to 36%.
Conclusion: In conclusion, adding nigrosome visual analysis to BrainGML highest diagnosis probability turned most 'wrong'
diagnoses into 'undefined' diagnoses.
Limitations: The main limitations are small population samples, unknown DLB diagnosis by BrainGML, and no pathological
diagnoses.
Ethics committee - additional information: This study was approved by the French South-East Ethics committee.
Is pure akinesia with gait freezing associated with more ventriculomegaly in brain MRI than other subtypes of
progressive supranuclear palsy? (7 min)
Marcos Jiménez Vázquez; Pamplona / Spain
599
Author Block: M. Jiménez Vázquez, P. Del Nido Recio, M. R. López de la Torre Carretero, C. Mbongo, M. Calvo Imirizaldu, R. Garcia de
Eulate, P. Dominguez Echavarri, C. Espinoza-Vinces, M. R. I. Luquin; Pamplona/ES
Purpose: This study aimed to evaluate if patients with PAGF have more ventriculomegaly in MR imaging in comparison with other
subtypes of PSP. Freezing of gait can be seen in progressive supranuclear palsy (PSP), specifically in the pure akinesia with gait
freezing (PAGF) subtype, mimicking Normal Pressure Hydrocephalus (NPH).
Methods or Background: We retrospectively analysed a cohort of 48 patients who underwent brain MRI between 2018 and 2023
due to freezing of gait as the main clinical feature. 40 patients were posteriorly diagnosed with PSP, four with NPH and four with
Alzheimer’s disease. After excluding the last patients, we divided the PSP patients into a group posteriorly diagnosed with PAGF (27)
and another diagnosed with other PSP subtypes (13). Evans Index (EI) and Callosal Angle (CA) were measured in the two groups and
cut-off values of EI >0.3 and CA <100º were selected for defining ventriculomegaly. Midbrain-to-pons ratio and frontal and midbrain
atrophy were also evaluated. Fisher’s Exact Test and T-test were applied to studying proportions and means.
Results or Findings: The mean EI was significantly higher in the PSP-PAGF group compared to the other subtypes group (0.32 vs
0.29, p=0.041), while the mean CA was significantly lower (103.69º vs 118.35º, p=0.008). Moreover, CA cut-off <100º showed a
significant difference between both groups (p=0.015) with a specificity of 92.31%. EI cut-off >0.3 showed a non-significant sensibility
of 74.07% (p=0.155). No significant differences were found in the midbrain-to-pons ratio and frontal and midbrain atrophy between
the two groups.
Conclusion: Our results illustrate that patients with freezing of gait and ventricular enlargement in imaging will probably have a
diagnosis of PSP-PAGF, after excluding NPH. PSP-PAGF is associated with more significant ventriculomegaly than other subtypes of
PSP.
Early re-emerging tremor after MRgFUS thalamotomy: case-control analysis of procedural and imaging features (7 min)
Pierfrancesco Badini; L'Aquila / Italy
Author Block: P. Badini, C. Santobuono, F. Colarieti, A. Di Gioia, I. Antonio, F. Bruno, A. Splendiani, E. Di Cesare, A. Catalucci;
L'Aquila/IT
Purpose: This study aimed to identify possible prognostic factors determining early tremor relapse after MRgFUS thalamotomy in
patients with essential tremor (ET) and Parkinson's disease (PD).
Methods or Background: 9 patients (6 ET, 3PD) who underwent Vim MRgFUS thalamotomy in a single institution and developed
early re-emergent tremor were analysed. Demographic and clinical characteristics of patients were compared to a pairwise matched
control group of patients with stable tremor relief at the same follow-up period. In both groups, procedural parameters of target and
sonications and MR imaging findings (including lesion shape and volume in multiparametric sequences and dentatorubrothalamic
tract DTI analysis) were compared.
Results or Findings: We did not find statistically significant differences in gender and age between the two groups. Concerning MRI
analysis, we found larger thalamotomy lesions in the control group with stable outcomes, compared to patients with tremor relapse.
In the tractography evaluation, we found a more frequent eccentric position of the DRTt in patients with tremor relapse.
Conclusion: The most relevant determining factors for tremor relapse after MRgFUS thalamotomy appear to be the size of the
thalamotomy lesion and inaccurate thalamic targeting.
Limitations: Limited study sample and retrospective analysis.
Ethics committee - additional information: All subjects gave their informed consent for inclusion before they participated in the
study. The study was conducted in accordance with the Declaration of Helsinki.
Resting-state brain connectivity in patients with essential tremor undergoing vim ablation with MRgFUS: preliminary
results (7 min)
Filippo Colarieti; L'Aquila / Italy
600
Author Block: F. Colarieti, C. Santobuono, P. Badini, A. Di Gioia, I. Antonio, A. Catalucci, F. Bruno, E. Di Cesare, A. Splendiani;
L'Aquila/IT
Purpose: This study aimed to verify any changes in brain connectivity in patients with essential tremors undergoing MRgFUs
thalamotomy. Essential tremor is the result of abnormal communication between certain areas of the brain, including the thalamus,
cerebellum, and brainstem. Vim MRgFUS thalamotomy is a new minimally invasive procedure used for the treatment of disabling
tremors.
Methods or Background: 15 patients with essential tremor (TE) (mean age= 64.1, SD= 13.1), without cognitive impairments (mean
IQ= 109.0, SD= 9.9), were included in the study. These patients' resting-state brain connectivity was recorded before treatment and
after six months with 3T MRI. The MRgFUS thalamotomy treatment was effective in all patients, with a mean FTM (Fahn-Tolosa-Marin
tremor scale) reduction of the treated upper limb from 5.1 to 0.8. An analysis of ROI-to-ROI connectivity was then carried out with
Coon Toolbox studying the mains brain networks to assess the difference in pre- and post-treatment brain connectivity.
Results or Findings: The results indicated significantly greater post-treatment connectivity between the anterior cerebellum and
the left parietal lobe (default mode network) and between the left occipital cortex (visual network) and the cerebellum.
Conclusion: Although preliminary, our results indicate greater connectivity of the cerebellum following MRgFUS treatment, in
accordance with better fine motor control reported by patients undergoing surgery.
Limitations: This was a low study sample.
Ethics committee - additional information: All subjects gave informed consent for inclusion before participating in the study. The
study was conducted in accordance with the Declaration of Helsinki.
Changes to the peripheral nervous system in chronic spinal cord injury: a study on 3T MR neurography (7 min)
Johann Malte Enno Jende; Heidelberg / Germany
Author Block: J. M. E. Jende, L. Heutehaus, F. Preisner, C. Mooshage, R. Rupp, M. Bendzus, N. Weidner, F. Kurz, S. Franz;
Heidelberg/DE
Purpose: This study aimed to combine MRN and clinical assessments in individuals with chronic SCI and non-disabled controls. It is
known that structural and functional changes to the peripheral nervous system (PNS) can occur in disorders of the central nervous
system (CNS) such as multiple sclerosis. To date, it is unknown to what extent changes at the PNS level occur after spinal cord injury
(SCI) and whether such changes are relevant for functional recovery or the development of neuropathic painful symptoms below the
level of injury. 3 Tesla Magnetic Resonance Neurography (MRN) allows the detection and quantification of structural and functional
damage to peripheral nerves.
Methods or Background: 20 participants with chronic SCI and 20 controls matched for sex, age, and body-mass index underwent
t2-weighted and diffusion-weighted MRN of the sciatic nerve. The sciatic nerve’s mean cross-sectional area (CSA), fascicular lesion
load, and fractional anisotropy (FA) were calculated. Results were correlated with clinical and electrophysiologic assessments.
Results or Findings: Sciatic nerve CSA and lesion load were larger (21.29 mm 2±5.82 vs 14.08 mm 2±4.62; p<0.001, and 8.70%
±7.47 vs 3.60% ±2.45; p<0.001) in individuals with SCI compared to controls, whereas FA was lower (0.55±0.11 vs 0.63±0.08;
p=0.022). MRN parameters correlated with electrophysiological results but did not correlate with the extent of myelopathy or clinical
severity of SCI.
Conclusion: Individuals with chronic SCI are subject to a decline of structural peripheral nerve integrity that may occur
independently from the clinical severity of SCI. This may have an impact on functional recovery following SCI. The underlying cause of
this phenomenon remains yet to be determined since no association with the extent of myelopathy or the severity of SCI-related
disability could be found.
Limitations: This study is limited by its cross-sectional design and the small number of participants.
Funding for this study: Funding for this study was received by the International Foundation for Research in Paraplegia, The Else
Kröner-Fresenius Stiftung and the German Research Foundation.
Ethics committee - additional information: This study was approved by the ethics committee of Heidelberg University Hospital.
AI-based three-dimensional analysis of lumbar vertebrae micromovements in spondylolisthesis patients with weight-
bearing MRI (7 min)
Luca Alessandro Cappellini; Pieve Emanuele / Italy
601
Author Block: R. Levi1, L. A. Cappellini1, M. Battaglia1, F. Garoli1, M. De robertis1, G. Savini1, M. Fornari2, M. Grimaldi2, L. S. S. Politi2;
1
Pieve Emanuele/IT, 2Rozzano/IT
Purpose: This study aimed to automatically evaluate the micro-instability of lumbar vertebrae through 3D deep-learning analysis of
MRI in supine and standing positions in healthy subjects and patients with spondylolisthesis.
Methods or Background: The current assessment of spondylolisthesis relies on the Meyerding grading system from
morphodynamical lateral X-rays and supine MRI. These approaches measure instability only on the sagittal plane providing a partial
representation of the disease.
We retrospectively included 28 subjects (10 patients with spondylolisthesis and 18 controls) who underwent weight-bearing MRI
(wbMRI) with 3D-HYCE acquisitions in supine and upright positions. Deep-learning algorithms were employed to segment vertebrae
and dural sac and to quantify advanced micro-instability parameters: centre of mass (COM), variation of the angle passing through
the COM, and roto-translation matrix (RTM). These parameters were correlated to pain numerical rating scale (PNRS).
Results or Findings: Automatic quantification of vertebral posterior wall listhesis resulted in the upgrading of Meyerding
classification from grade I to II in 3 patients (30%). Patients with spondylolisthesis demonstrated significantly reduced lordotic
curvature in the upright position compared to controls showing reduced anterior movement of COM (p=0.0105) and narrower anterior
variation of the angle (p<0.001). The analysis of RTM reported higher rotation in controls for anteroposterior direction at L2-L3 level
(p=0.038) and cranio-caudal direction at L3-L4 level (p=0.006).
In patients, pre-operative PNRS evaluation of back pain was of 8.50 (std. 0.93) with a mean post-operative PNRS of 4.00 (std. 2.51).
Translation in the anteroposterior movement was positively associated with pre-operative PNRS (R=0.701, p=0.042). Post-operative
PNRS was positively associated with the latero-lateral movement (R=0.699, p=0.045).
Conclusion: This preliminary study highlights the clinical potentials of wbMRI and deep-learning 3D analysis for spondylolisthesis
micro-movements assessment, offering a comprehensive insight into vertebral instability, peri-procedural pain, and indication to
neurosurgical treatment.
Limitations: This study had a limited population.
Neurite orientation dispersion and density imaging quantifies microstructural impairment in the thalamus and its
connectivity in amyotrophic lateral sclerosis (7 min)
Yun-bin Cao; Fuzhou / China
Author Block: Y-b. Cao, H-J. Chen; Fuzhou/CN

Purpose: This study aimed to evaluate microstructural impairment in the thalamus and thalamocortical connectivity using neurite
orientation dispersion and density imaging (NODDI) in amyotrophic lateral sclerosis (ALS).
Methods or Background: This study included 47 healthy controls and 43 ALS patients, whose structural and diffusion-weighted data
were collected. We used state-of-the-art parallel transport tractography to identify thalamocortical pathways in individual spaces. The
thalamus was then parcellated into six subregions based on its connectivity pattern with the priori defined cortical (i.e.,
prefrontal/motor/somatosensory/temporal/posterior-parietal/occipital) regions. For each of the thalamic and cortical subregions and
thalamocortical tracts, we compared the following NODDI metrics between groups: orientation dispersion index (ODI), neurite density
index (NDI), and isotropic volume fraction (ISO). We also used these metrics to conduct receiver operating characteristic curve (ROC)
analyses and Spearman correlation.
Results or Findings: In ALS patients, we found decreased ODI and increased ISO in the thalamic subregion connecting the left
motor cortex and other extramotor (e.g., somatosensory and occipital) cortex (Bonferroni-corrected P <0.05). NDI decreased in
bilateral thalamo-motor and thalamo-somatosensory tracts and in the right thalamo-posterior-parietal and thalamo-occipital tracts
(Bonferroni-corrected P <0.05). NDI reduction in the bilateral thalamo-motor tract (P= 0.017 and 0.009) and left thalamo-
somatosensory tract (P= 0.029) was correlated with disease severity. In thalamocortical tracts, NDI yielded a higher effect size during
between-group comparisons and a greater area under ROC (P <0.05) compared with conventional diffusion tensor imaging metrics.
Conclusion: Microstructural impairment in the thalamus and thalamocortical connectivity is the hallmark of ALS. NODDI improved
the detection of disrupted thalamocortical connectivity in ALS.
Limitations: Firstly, the study had a relatively small sample size. Secondly, the study design was cross-sectional. Thirdly, the pattern
of connectivity between other subcortical areas and the cortex should also be evaluated in future ALS studies.
Funding for this study: This study was supported by grants from the Graduate Student Academic Exchange Fund of Fujian Medical
University.
Ethics committee - additional information: This study was approved by the Research Ethics Committee of Fujian Medical
University Union Hospital
602
E³ 1521 - Practical applications of radiomics
Categories: Breast, Interventional Oncologic Radiology, Oncologic Imaging

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Breast cancer (45 min)

Pascal A.T. Baltzer; Vienna / Austria
1. To understand the potential for breast cancer by considering imaging as the data source.
2. To learn about the most recent practical applications of radiomics in breast imaging and intervention.
3. To discuss how the application of radiomics could affect the management of patients with breast cancer.
Ovarian cancer (45 min)

1. To understand the potential for ovarian cancer by considering imaging as the data source.
2. To learn about the most recent practical applications of radiomics in ovarian cancer.
3. To discuss how radiomics could affect the management of patients with ovarian cancer.
603
E³ 1519 - US: beyond old limitations

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:

A tour around new US technology (25 min)

1. To explain recent innovations in US devices, including hand-held technology.

2. To present the latest developments in US probes.
3. To discuss the pros and cons of different technical solutions for elastography.
CEUS: from qualitative to quantitative approach (25 min)

Jean Michel Correas; Paris / France
1. To present different contrast agents and mechanisms of action.

2. To present different software solutions for quantitative analysis.
3. To show the advantages of quantitative analysis in different clinical settings.
The ultimate frontiers of AI in the US (25 min)

1. To summarise the current applications of AI in US examinations.

2. To describe major benefits for radiologists and the diagnostic workflow.
3. To examine the most important barriers to successful implementation.
Discussion (10 min)
604
PC 15b - Implementation and effectiveness of AI in practice
Categories: Artificial Intelligence & Machine Learning, Radiographers

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderators:

Tips and pitfalls for bringing AI into practice (15 min)

Janni Jensen; Odense / Denmark
1. To describe the practical set-up of facilitating AI solutions at a large university hospital.

2. To present examples of AI solutions in a department of radiology.
3. To identify pitfalls of AI in the clinical setting.
Methods for evaluating the effectiveness of AI (15 min)

1. To define and describe key metrics for evaluating AI performance.

2. To summarise currently available AI educational provisions for radiologists and radiographers in Europe.
3. To discuss challenges, opportunities and synergies for AI education for different medical imaging professionals.
The role of educating the medical imaging workforce for clinical adoption of AI (15 min)
Christina Malamateniou; London / United Kingdom
1. To highlight the different factors impacting AI adoption in clinical practice for medical imaging services.
2. To summarise currently available AI educational provisions for radiologists and radiographers in Europe.
3.To discuss challenges, opportunities and synergies for AI education for different medical imaging professionals.
Use of AI to help radiographers enhance patient experience (15 min)

1. To explore the forefront of AI technologies related to patient care.

2. To contextualise radiographers’ role with AI.
3.To consider ethics and consent for the use of AI in radiography practice.
Panel discussion: How can radiographers embrace AI implementation in practice? (25 min)
605
Meets 15 - The importance of the radiologist as part of a medical team: Qatar 2022 experience
Categories: Education, Imaging Methods, Interventional Radiology, Multidisciplinary, Musculoskeletal

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderators:
Eduardo Pablo Eyheremendy; Buenos Aires / Argentina
Alejandro Ulises Rolón; Buenos Aires / Argentina

Eduardo Pablo Eyheremendy; Buenos Aires / Argentina
MRI and ultrasound in sports pathology: allied methods (15 min)

Agustin Manuel Marrero; LA PLATA / Argentina
1. To demonstrate the importance of combining methods in muscle and tendon pathology.
Muscle injuries in high-performance athletes (15 min)

Tomas Pascual; Buenos aires / Argentina
1. To discuss the importance of a multidisciplinary perspective in sports injuries.

2. To describe the participation of the radiologist in recreational and elite sports.
Road to Qatar 2022: world in victory (15 min)

1. To show how Argentinian MSK radiologists started participating in sports medical teams.
2. To explain how Argentinian MSK radiologists actively contribute to decision-making in the diagnostic and therapeutical stages of
sports injuries.
606
RPS 1509 - Advanced imaging and challenging interventions in the liver
Categories: Artificial Intelligence & Machine Learning, Interventional Oncologic Radiology, Interventional Radiology, Vascular
Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Stéphanie Franchi-Abella; Le Kremlin-Bicêtre / France
Trans-splenic antegrade coil-assisted transvenous occlusion (TACATO) of gastric varices associated with gastro-renal
shunt in cirrhosis: a single-centre prospective preliminary study (7 min)
Stefano Groff; Padova / Italy
Author Block: S. Groff, G. Cicognini, G. Barbiero, M. Battistel, S. Shalaby, M. Senzolo, G. De Conti; Padova/IT
Purpose: This study aimed to report the technical success and clinical efficacy of the sole antegrade trans-splenic approach for
occlusion of gastric varices (GVs) associated with gastro-renal shunt (GRS) in cirrhotics.
Methods or Background: All patients who bled from GVs associated with GRS without severe ascites or large oesophagal varices
were selected for evaluation. From February 2020 to July 2023, 29 patients were evaluated and 19 were included for treatment as
secondary prophylaxis and prospectively followed-up (mean time of 15.3 months; range 3-43 months).
Splenoportography from trans-splenic access classified venous afferents to the GVs (Saad-Caldwell classification), which were
selectively catheterized and embolized with coils+/-N-butyl-2-cyanoacrylate+methacryloxysulfolane and ethiodized oil (NBCA+LUF).
Final splenoportography assessed occlusion of GVs (technical success). The trans-splenic tract was sealed with NBCA+LUF. Access
size, number-size of microcoils, use of NBCA+LUF and fluoroscopy time were recorded.
Postprocedural ultrasound and haemoglobin levels evaluated bleeding complications.
Follow-up included clinical evaluation, contrast-enhanced CT (CECT) and esophagogastroduodenoscopy (EGD).
Results or Findings: Splenoportography identified Saad-Caldwell type 1b GVs in two cases, type 2b in ten cases and type 3b in
seven cases. Different introducer sheath and microcoils sizes were used, with a mean number of 8.4 microcoils per patient (range
1-20). Microcoils alone were used in four patients and with NBCA+LUF in 15 patients. Mean fluoroscopy time was 20.8 min (range
7.1-43.1 min). Final splenoportography documented total occlusion of GVs in 15 patients (79%) and partial occlusion in four patients
(21%). No bleeding complications.
Clinical follow-up documented stable liver function, no re-bleeding from GVs nor development/worsening of ascites. CECT showed
three cases of splenic/portal vein thrombosis. Follow-up EGD revealed the disappearance/reduction of GVs in all patients.
Conclusion: TACATO seems safe and effective for treating GVs associated with GRS and should be further evaluated in the algorithm
of treatment.
Limitations: Preliminary study: larger number of patients required.
Ethics committee - additional information: This study was approved by the Padua University Hospital (HIC protocol #0034435)
Factors impacting survival after transarterial radioembolisation in patients with intrahepatic cholangiocarcinoma: a
combined analysis of the prospective CIRT and CIRT-FR studies (7 min)
Peter Reimer; Karlsruhe / Germany
607
Author Block: P. Reimer1, V. Vilgrain2, D. Arnold3, R. Loffroy4, B. Sangro5, M. Urdaniz6, H. Pereira7, N. de Jong6, T. K. Helmberger8;
1
Karlsruhe/DE, 2Clichy/FR, 3Hambrug/DE, 4Dijon/FR, 5Pamplona/ES, 6Vienna/AT, 7Paris/FR, 8Munich/DE
Purpose: This study aimed to discuss the treatment option of Transarterial radioembolization (TARE) with Yttrium-90 resin
microspheres for patients with intrahepatic cholangiocarcinoma (ICC). Optimizing the timing of TARE in relation to systemic therapies
and patient selection remains challenging. We report here on the effectiveness, safety, and prognostic factors associated with TARE
for ICC in an analysis of the prospective observational CIRT (CIRSE Registry for SIR-Spheres Therapy) studies (NCT02305459 and
NCT03256994).
Methods or Background: A separate analysis of all 174 ICC patients within CIRT studies was performed. Patient characteristics and
treatment-related data were collected at baseline; time-to-event data (overall survival [OS], progression-free survival [PFS] and
hepatic PFS) were collected at every follow-up visit. Log-rank tests and a multivariable Cox proportional hazard model were used to
identify prognostic factors.
Results or Findings: Patients receiving a first-line strategy of TARE, in addition to any systemic treatment, had a median OS and
PFS of 32.5 months and 11.3 months. Patients selected for first-line TARE alone showed a median OS and PFS of 16.2 months and 7.4
months, whereas TARE as 2nd or further treatment-line resulted in a median OS and PFS of 12 and 9.3 months (p=0.0028), and 5.1
and 3.5 months (p=0.0012), respectively. Partition model dosimetry was an independent predictor for better OS (HR 0.59 [95% CI
0.37-0.94], p=0.0259). No extrahepatic disease, no ascites, and <6.1 months from diagnosis to treatment were independent
predictors for longer PFS.
Conclusion: This combined analysis indicates that in unresectable ICC, TARE in combination with any systemic treatment is a
promising treatment option in the first line. Partition model dosimetry improved the effectiveness of TARE.
Limitations: Patients and physicians were not blinded within the prospective CIRSE Registry for SIR-Spheres therapy reflecting the
real-world treatment situation.
Funding for this study: This study was funded by an unconditional research grant from Sirtex Medical Europe, GmbH.
Ethics committee - additional information: This study was approved by applicable regulatory bodies and/or ethics committees in
Germany, Switzerland, Turkey, Italy, Spain, Israel, France, and Belgium as applicable for observational studies.
Simultaneous portal and hepatic vein embolisation in inducing future liver remnant hypertrophy: the impact of
steatosis in hepatic regeneration (7 min)
Domenico Santangelo; Milan / Italy
Author Block: D. Santangelo, D. Palumbo, M. Platì, C. Canevari, F. Ratti, L. Aldrighetti, F. De Cobelli; Milan/IT
Purpose: This study aimed to assess the impact of preoperative LS on complications after major liver surgery and to identify any
eventual liver steatosis modification pattern after liver venous deprivation (LVD). Liver steatosis (LS) has been widely associated with
post-surgical complications in different surgery fields.
Methods or Background: Patients who underwent, between 01/2019 and 12/2022, LVD were identified (n = 40). Those who
ultimately underwent surgery (n = 27, 67.5%) were enrolled. LS was defined as mean liver density lower than 50 HU. Pre/post-
procedural CT and scintigraphic data were collected. Data regarding post-surgical complications, in particular, post-hepatectomy liver
failure (PHLF) were collected.
Results or Findings: Of the 27 patients undergoing surgery, 12 presented with LS at a preoperative CT scan. This subgroup
demonstrated a higher chance of developing post-surgical complications (p < 0.05). Non-steatotic patients showed, after LVD, a rapid
decrease of residual liver HU values (- 5.60 HU, +/- 4.91). On the contrary, patients with baseline steatosis did not experience
significant changes (p < 0.001). Those who developed post-LVD steatosis (n = 7) did not show an increased risk of developing post-
surgical complications when compared to those who remained in their steatosis group. Patients who developed post-LVD steatosis
demonstrate a lower functional degree of hypertrophy (p < 0.05).
Conclusion: This proof-of-concept study suggests two conclusions: (1) Pre-LVD LS is a risk factor for the development of post-surgical
complications, and (2) LVD seems to function as a trigger for a modification of residual liver density; such “induced” post-LVD
steatosis differs from the metabolic one both in terms of lower post-surgical complications rate and impaired function of the liver
remnant. We are possibly looking at two different histological entities brought together by a similar radiological appearance.
Limitations: The limitations of this study are that it was retrospective and it was a small population.
Ethics committee - additional information: This study was approved by Leatum 64/INT/2021.
CT-guided high-dose-rate brachytherapy induces systemic proteins of proliferation and angiogenesis predicting
outcome in HCC (7 min)
Matthias Max Rudolf Stechele; Munich / Germany
608
Author Block: M. M. R. Stechele1, S. N. Goldberg2, M. Wildgruber1, P. M. Kazmierczak1, L. Salvermoser1, M. Alunni-Fabbroni1, J. Ricke1;
1
Munich/DE, 2Jerusalem/IL
Purpose: This study aimed to assess the potential prognostic value of proliferation and angiogenesis markers following CT-guided
high-dose-rate brachytherapy (HDR-BT) of hepatocellular carcinoma (HCC).
Methods or Background: For this prospective study, HDR-BT (1x15Gy) was performed in 24 HCC patients. Plasma was obtained at
baseline and 48 hours post-HDR-BT and analysed using an Olink proteomics Target-96 immuno-oncology-panel, including multiple
markers of angiogenesis and proliferation. Protein fold-change (FC) ratios were calculated. Patients without progression within 6m or
systemic progression within two years were classified as responders (R; n=12). Patients with recurrence within 6m and/or tumour
progression or extrahepatic disease within two years were classified as non-responders (NR; n=12).
Results or Findings: Angiopoietin-1 (median R: 0.86; NR: 1.29; p= 0.008) and PDGF-B (R: 0.89; NR: 1.25; p= 0.033) were
significantly elevated in non-responders compared to responders. FC between responders and non-responders did not differ
significantly for Angiopoietin-2, VEGF-A, and VEGFR-2. FC of EGF (R: 0.67; NR: 1.51; p= 0.028) was significantly elevated in non-
responders compared to responders with no significant difference for FGF-2, HGF, and PGF. Kaplan-Meier analyses demonstrated a
significantly shorter time to systemic progression for increased Angiopoietin-1 and EGF (p= 0.011 and 0.019 respectively), but not for
remaining proteins (all p >0.1). Pooled analysis for all 9 proteins showed significantly shorter systemic progression for FC ≥ 1.3 in ≥ 3
proteins (p= 0.022).
Conclusion: Increased plasma levels of Angiopoietin-1 and EGF after HDR-BT for HCC are associated with poor response and may
therefore function as predictive biomarkers of outcome.
Limitations: Given a small cohort, further expansion of protein panels and evaluation of potential dynamic changes of target
proteins over time are needed to achieve optimal biomarker potential.
Ethics committee - additional information: In this monocentric clinical study, we analysed prospectively from acquired
data of 24 patients from the “ESTIMATE” patient cohort (Studiennummer:
DRKS00010587, Deutsches Register Klinischer Studien). Ethical approval was
provided by the ethics committee Ethikkommission bei der LMU München:
(reference number “17-346”) on June 20, 2017, and August 26, 2020.
Transjugular intrahepatic porto-systemic shunt in paediatric patients: comparison between standard approach and
alternative techniques (7 min)
Paolo Marra; Bergamo / Italy
Author Block: P. Marra1, R. Muglia1, N. Liggeri1, C. Sallemi2, L. Dulcetta1, F. S. Carbone1, S. Sironi1; 1Bergamo/IT, 2Brescia/IT
Purpose: This study aimed to compare technical success, safety and clinical outcomes of TIPS performed with conventional
technique or assisted by percutaneous approach in paediatric patients and difficult anatomies.
Methods or Background: From January 2019, paediatric patients or young adults with native liver or split liver grafts undergoing
TIPS were retrospectively reviewed. Group A underwent TIPS through a standard transjugular approach. Group B underwent standard
DIPS or "gun-sight" TIPS assisted by transhepatic and transsplenic approaches. Technical success in terms of correct TIPS placement,
safety in terms of complications and clinical outcomes in terms of bleeding and ascites control were assessed and compared between
groups.
Results or Findings: 16 patients underwent TIPS placement due to portal hypertension and variceal bleeding (n=7), portal vein
thrombosis (n=6), Budd-Chiari syndrome (n=1) or refractory ascites (n=1). Out of six patients with portal vein thrombosis, four were
affected by chronic portal vein thrombosis and cavernous transformation, and failed percutaneous portal vein recanalization or
surgical meso-rex shunt. In two cases, TIPS was a bridge to liver transplant. In group A, eight patients (n=3 with regular anatomy;
n=3 with cavernoma; n=2 with Budd-Chiari) successfully underwent TIPS with septic shock in one case. In group B, eight patients
(n=4 with regular anatomy; n=4 with cavernoma; none with Budd-Chiari) successfully received TIPS with severe bleeding and
precipitating liver failure in one case. No technical failures were recorded. Clinical outcome was good in all patients.
Conclusion: Alternative techniques for TIPS placement have a great success rate with good clinical outcomes. In paediatric patients
and difficult anatomies DIPS and "gun-sight" TIPS might be considered.
Limitations: Small sample and potential selection bias.
Ethics committee - additional information: The study Portal_01 was approved by the Ethical Committee of Bergamo (reg.92-21).
Feasibility, safety and mid-term outcomes of non-expandable biodegradable biliary stents in paediatric liver
transplants (7 min)
Paolo Marra; Bergamo / Italy
609
Author Block: P. Marra, R. Muglia, F. S. Carbone, L. Dulcetta, S. Sironi; Bergamo/IT
Purpose: This study aimed to report the mid-term outcomes of percutaneous biliary drainage with double biodegradable stent
placement for the treatment of benign anastomotic biliary strictures in paediatric liver transplants (PLT).
Methods or Background: From March 2022, consecutive patients with PLT who underwent percutaneous transhepatic
cholangiography (PTC), bilioplasty and placement of two side-by-side 10F helical-shaped (non-expandable) biodegradable (slow
degradation profile of 11 weeks) biliary stents were prospectively enrolled. Feasibility of double stent placement, safety in terms of
procedure-related complications, freedom from stricture recurrence at imaging and absence of symptoms/signs of cholestasis were
assessed.
Results or Findings: Percutaneous transhepatic stent placement was performed in 19 patients (10 females; median age 7 years,
range 3-11 years) for the treatment of perianastomotic strictures (18 hepatico-jejunostomy; 1 duct-to-duct anastomosis). Stenting
was performed after indwelling of trans-anastomotic biliary drainage for a median time of 28 days (range 11.5-53 days). Double stent
placement in a side-by-side fashion was technically successful in 15/19 (79%) patients; in four patients a single trans-anastomotic
stent was implanted. Stent fragmentation/migration before the expected degradation time was observed in 4/19 (21%) patients. One
patient was readmitted with cholangitis and conservatively treated; no other complications occurred. Stricture recurrence was
observed during a median follow-up of 233.5 days (range 182.5-263 days) in 6/19 (32%) patients who required new PTC procedures.
Conclusion: Preliminary data suggests that the placement of biodegradable biliary stents may mitigate concerns related to the
maintenance of external biliary drainage in PLT patients affected by benign perianastomotic biliary strictures. However, the long-term
efficacy of non-expandable stents is limited and needs to be compared with that of expandable devices in the PLT population.
Limitations: This study had a limited sample and lacked case-control analysis.
Ethics committee - additional information: This study was approved by Reg. Sperim. N. 061/23
Role of different stent types in endoluminal IRE for recanalization of an occluded metal stent: mathematical model of
temperature distribution (7 min)
Peter Matkulcik; Brno / Czechia
Author Block: P. Matkulcik, M. Hemzal, T. Andrasina, T. Rohan, M. Straka; Brno/CZ

Purpose: This study aimed to analyse the temperature distribution of endoluminal IRE for stent recanalization using a mathematical
model in exact stent models.
Methods or Background: Endoluminal IRE for recanalization of different uncovered metal stents (Ella-CS, Boston Scientific, S&G
Biotech and Micro-Tech medical) using a 3-electrode IRE catheter was simulated in a mathematical model of perfused liver tissue. The
obstruction of the metal stent was simulated by 1, 2 and 3 mm layers of interposed tissue. 2D and 3D calculations by finite element
software (COMSOL Multiphysics) defined the extent of thermal distribution zones in different IRE protocols (100 100 μs pulses, 1 Hz at
300 V, 650 V, 1000 V,1300 V). Differences in temperatures at the electrode, on the stent and 1 mm outside of the stent depending on
stent type, voltage and width of interposed tissue were investigated.
Results or Findings: The biggest differences - up to 21% - were observed with 1 mm interposed tissue at voltages set to 1000 V at
the electrode and at the stent (52.606-64.433 °C and 52.643-63.687 °C respectively). The highest temperatures were observed in
simulations with Ella-CS stent, on the other hand, the lowest temperatures were observed in simulations with Microtech stent across
all simulations. At voltages set to 1300 V, temperatures at the electrode and at the stent with 1 mm interposed tissue were
72.146-83.073 °C and 72,556-81,921 °C respectively, thus IRE in those scenarios was producing also thermal effects of ablation.
Conclusion: In the mathematical model, substantial differences in temperatures were observed between the specific stents
analysed. In the highest voltages with 1 mm interposed tissue, thermal effects of ablation were present.
Limitations: This was a computer simulation.
Funding for this study: This study was funded by the Ministry of Health, Czech Republic: grant No.: NU21-08-00561.
Ethics committee - additional information: Ethics committee approval was not necessary in this study.
Multiparametric MRI-based radiomics nomogram predicts the recurrence of hepatocellular carcinoma after
postoperative adjuvant transarterial arterial chemoembolisation (PA-TACE) (7 min)
Xinyu Guo; Lishui / China
610
Author Block: X. Guo, L. Zheng, J. Ji; Lishui/CN
Purpose: This study aimed to develop and validate a multiparametric MRI-based radiomics model for predicting the recurrence in
hepatocellular carcinoma (HCC) patients after postoperative adjuvant transcatheter arterial chemoembolization (PA-TACE).
Methods or Background: In this retrospective study, 117 HCC patients (81 for training and 36 for validation) treated with PA-TACE
were included. Multiparametric radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging
(DWI) (b=800), and hepatic arterial phase (AP). Least absolute shrinkage and selection operator (LASSO)-COX regression was utilised
to select radiomics features and calculate the radiomics score (Rad-score). Optimal clinical characteristics selected by univariate and
multivariate Cox analysis were integrated with Rad-score to develop a recurrence-free survival (RFS) prediction model.
Results or Findings: 15 radiomics features were selected from three sequences of MRI. The significant difference in RFS between
the high-risk and low-risk signature groups based on the Rad-score median value of 0.434 was confirmed in two cohorts (P < 0.05).
The clinical independent predictors were neutrophil to lymphocyte ratio (NLR) (HR= 1.49, 95% CI: 1.1-2.1, P= 0.022) and tumour size
(HR= 1.28, 95% CI: 1.1-1.5, P= 0.001). The clinical-radiomics model developed by Rad-score, NLR, and tumour size demonstrated
favourable performance for predicting recurrence. The AUCs at 1-, 3- and 5-year time-dependent ROC curves based on the combined
model were 0.82-0.91, and 0.75-0.91 in the two cohorts respectively.
Conclusion: The multiparametric MRI-based radiomics nomogram can well predict recurrence in HCC patients treated with PA-TACE
and can serve as a valuable tool for clinical prognosis.
Limitations: Firstly, the study was retrospective. Secondly, all participants of this study were diagnosed with hepatitis B virus (HBV)-
related HCC in China. Thirdly, our study did not explore tissue-related biomarkers such as specific gene mutations and signatures.
Funding for this study: Funding for this study was received from the National Natural Science Foundation of China (82102162 and
82072026), Natural Science Foundation of Zhejiang Province (LGF21H180002) and the Zhejiang Medical and Health Science Project
(2022RC087).
Ethics committee - additional information: This study was approved by the Institutional Review Board and Human Ethics
Committee of Lishui Hospital of Zhejiang University. The requirement for informed consent was waived.
The challenge of percutaneous recanalisation of chronic portal vein thrombosis: from the lack of evidence to clinical
experience (7 min)
Ludovico Dulcetta; Bergamo / Italy
Author Block: L. Dulcetta, P. Marra, F. S. Carbone, R. Muglia, S. Sironi; Bergamo/IT

Purpose: This study aimed to report the experience of a tertiary referral centre for pediatric and adult liver disease and
transplantation in the IR management of chronic complete PVT. Complete chronic portal vein thrombosis (PVT) is a condition often
requiring treatment to avoid potentially serious complications.
Methods or Background: A retrospective search was performed to identify pediatric and adult patients with an imaging diagnosis of
chronic complete PVT and with clinical signs of severe portal hypertension who underwent successful percutaneous portal vein
recanalization (PVR) from January 2020 to October 2023. Clinical and imaging follow-up was analysed.
Results or Findings: 12 patients (median age=18,5 years, IQ=41 years; 9 males) with chronic complete PVT who underwent at
least one attempt of IR treatment from January 2020 to October 2023 were retrieved. Six patients were paediatrics and 7 patients
(58%) were liver transplant recipients. All patients had chronic occlusion of the main portal vein, either with (n=2) or without (n=10)
occlusion of the superior mesenteric and/or the splenic veins. Cavernomatous transformation was noted in 11/12 (92%) patients. At
the time of treatment, all patients presented clinical signs of portal hypertension. A total of 20 percutaneous transhepatic,
transsplenic and/or transjugular procedures were performed.
Overall technical success, defined as patency of the portal vein after at least one procedure, was achieved in all patients. After a
median follow-up of 12 months, clinical success in terms of the absence of clinical signs of severe portal hypertension was achieved in
all patients. No major complications occurred.
Conclusion: PVR is a feasible, safe, effective and minimally invasive procedure with excellent technical and clinical success for the
management of chronic complete PVT in both pediatric and adult patients.
Ethics committee - additional information: No information provided by submitter.
Improved identification of good candidates for the treatment of intermediate/advanced hepatocellular carcinoma by
Yttrium-90 transarterial radioembolisation (7 min)
Inès Oreistein; Angers / France
611
Author Block: I. Oreistein, C. Vitellius, C. Aubé, F. Oberti, J. Boursier, A. Bouvier, F. Lacoeuille, L. Vervueren, S. Poulard; Angers/FR
Purpose: This study aimed to validate the prognostic score for HCC treated with TARE recently proposed by Spreafico and to improve
prediction by considering also pre-operative dosimetry. Yttrium-90 transarterial radioembolisation is a treatment for
intermediate/advanced hepatocellular carcinoma, but its position in the therapeutic arsenal remains poorly defined.
Methods or Background: 86 patients with HCC treated by SIRT in our centre were included. The provisional tumoral dose of
90Yttrium was calculated during the work-up. The Spreafico prognostic score was calculated as previously described and delineated
three patient groups with “favourable”, “intermediate”, and “dismal” prognosis. The main study outcome was overall survival and the
secondary outcome was progression-free survival.
Results or Findings: Fifty-three patients (62%) were treated with Therasphere®, and 33 patients (38%) were treated with
Sirsphere®. Median OS was 12.0 months (95% CI: 9.0-15.0), and median PFS was 5.0 months (95% CI: 3.5-6.5). OS was 15, 10 and 4
months in the three prognostic groups defined by the Spreafico score (p< 0.001). Independent predictors of OS were the presence of
cirrhosis, an optimal provisional tumoral dose, and the ALBI grade. The CODAG score developed as the sum of points attributed to
these three independent predictors, identified three patient groups: good (0-1 point), intermediate (2 points) and poor (3-4 points)
candidates. The CODAG score better discriminated the prognosis with median OS in the three groups being respectively 32, 11, and 4
months (p < 0.001). Median PFS was respectively 8, 5 and 3 months in the three CODAG groups (p< 0.001). The provisional and the
received tumoral doses were very well correlated (Rs=0.814, p< 0.001).
Conclusion: The CODAG score improves the identification of good candidates for the treatment of intermediate/advanced
hepatocellular carcinoma with transarterial radioembolisation.
Limitations: This study needs to be validated with an external cohort. The CODAG score better discriminated the prognosis than the
Spreafico score.
Funding for this study: No information was provided by the submitter.
Ethics committee - additional information: The data collection is in agreement with the ethics committee.
Retrospective collection data in ANGERS.
AI-derived body composition parameters as prognostic factors in patients with HCC undergoing TACE: results from a
Lukas Müller; Mainz / Germany
Author Block: L. Müller1, T. A. Auer2, J. Haubold3, F. Nensa3, M. Eisenblaetter4, V. Steinle5, D. Pinto Dos Santos6, R. Klöckner7;
1
Mainz/DE, 2Berlin/DE, 3Essen/DE, 4Freiburg/DE, 5Heidelberg/DE, 6Frankfurt/Köln/DE, 7Lübeck/DE
Purpose: This study aimed to investigate the role of BCA parameters for prognosis prediction in patients with HCC undergoing
transarterial chemoembolization (TACE). Body composition assessment (BCA) parameters have recently been identified as relevant
prognostic factors for patients with hepatocellular carcinoma (HCC).
Methods or Background: This retrospective multicentre study included a total of 754 treatment-naïve patients with HCC who
underwent TACE at six tertiary care centres between 2010–2020. Fully automated artificial intelligence-based quantitative 3D
volumetry of abdominal cavity tissue composition was performed to assess skeletal muscle volume (SM), total adipose tissue (TAT),
intra- and intermuscular adipose tissue (IMAT), visceral adipose tissue (VAT), and subcutaneous adipose tissue (SAT) on pre-
intervention computed tomography scans. BCA parameters were normalized to the slice number of the abdominal cavity. We
assessed the influence of BCA parameters on median overall survival (OS) and performed multivariate analysis including established
estimates of survival.
Results or Findings: Univariate survival analysis revealed that impaired median OS was predicted by low SM volume (p< 0.001),
high TAT volume (p= 0.013), and high SAT volume (p= 0.006). In multivariate survival analysis, SM remained an independent
prognostic factor (p= 0.039), while TAT and SAT volumes no longer showed predictive ability.
Conclusion: Skeletal muscle volume is an independent prognostic factor for survival prediction. Thus, the integration of SM into
novel scoring systems could potentially improve survival prediction and clinical decision-making. Fully automated approaches are
needed to foster the implementation of this imaging biomarker into daily routine.
Limitations: This study used retrospective data analysis and images were acquired using different scanner types.
Ethics committee - additional information: This study protocol was approved by the Ethics Committee of the Medical Association
of Rhineland-Palatinate, Germany (permit number 15913).
TARE of "multipedicular" malignant liver tumours: can we redistribute the blood flow? (7 min)
Rachele Fruzza; Pisa / Italy
612
Author Block: R. Fruzza1, E. Bozzi1, G. Lorenzoni1, R. Cervelli1, I. Bargellini2, L. Crocetti1, R. Cioni1, E. Neri1; 1Pisa/IT, 2Candiolo/IT
Purpose: This study aimed to demonstrate that “multipedicular” liver tumours (MPLT, i.e. tumours with more than one feeding
artery) can be treated by reducing the number of injection points through redistribution of flow, achieving complete coverage and
response in the target lesion.
Methods or Background: We retrospectively analysed our database of patients submitted to TARE between 2018 to 2022 and
selected patients with MPLT. Diagnostic work-up before radioembolisation was evaluated, and all tumours with an embolised intra-
hepatic feeding artery were identified. Careful evaluation of perfusion patterns was performed by SPECT-CT imaging 1 h after 99mTc-
MAA injection and a PET-CT scan was performed after Y90 treatment. Tumour response of MPLT was evaluated after 40 days and after
3 months by CT using mRECIST criteria for primary tumours and RECIST 1.1 for liver metastases.
Results or Findings: We performed 230 cases of TARE. In 25 patients, accessory branch supplies were embolised using microcoils.
Cross-perfusion in the embolised territory was evidenced by SPECT-CT in 22 cases and by PET-CT after Y90 injection in the other 3
cases. At 40 days follow-up we observed: 18 PR, 2 SD, 3 PD, and 2 patients were lost at follow-up. At 3 months follow-up: 10 PR, 1 CR,
4 PD, and 10 patients were lost to follow-up. No complications were reported after embolisation and 90Y administration.
Conclusion: Flow redistribution after embolisation of the accessory branches in MPLT was visible and achieved in all cases. Our
results confirmed that intratumoral flow redistribution after embolisation, in terms of toxicity, median administered dose and
radiological response, is safe and effective.
613
SF 15b - Validity of explainable AI and model trustworthiness
Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:

Assessing the trustworthiness of saliency maps as a tool to increase clinical explainability of AI radiology models (25
min)
Kostas Marias; Heraklion, Crete / Greece
1. To present the most popular saliency map techniques.

2. To demonstrate current applications in radiology (e.g. highlighting regions of attention indicating pathology).
3. To describe how they can be used within the radiology workflow along with their current limitations.
Generation of visual counterfactual explanation for classification problems in (medical) imaging (25 min)
Gaspard D'Assignies; Nantes / France
1. To understand this type of explanation and the link with the radiologist's practice.
2. To learn about methods with slightly different outputs that bring complementary information.
3. To be aware of the risks of each of these methods.
Quantifying explainability: does it improve clinical acceptance of AI? (25 min)

Alberto Traverso; Milan / Italy
1. To identify the most common explainability algorithms.

2. To list the clinical applications of such algorithms.
3. To understand the impact of such algorithms in clinical decision-making.
Panel discussion: Common pitfalls in building image repositories and how to overcome them (10 min)
614
SF 15a - The future of interventional radiology (IR) is clinical
Categories: Education, Interventional Radiology, Management/Leadership, Multidisciplinary, Professional Issues

ETC Level: LEVEL II
Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:

IR as a clinical speciality (15 min)

Raman Uberoi; Oxford / United Kingdom
1. To understand what the challenges for IR are.

2. To specify how a speciality helps in dealing with these challenges.
3. To know what the barriers and challenges are in setting up a speciality.
Training future IRs (15 min)

1. To briefly describe the evolution of the IR landscape.

2. To describe the main skills that future IRs will need to acquire.
3. To describe the facilities that should train future IRs.
Cornerstones of IR practice (15 min)

Andreas H. Mahnken; Marburg / Germany
1. To know about the current role of interventional radiology in clinical medicine.

2. To know about the interventional radiology process and future trends in patient care.
3. To know about basic needs in infrastructure for a successful interventional radiology programme.
4. To know about clinical practice building in interventional radiology.
What are the expected community and social benefits of an IR team (15 min)
Roberto Lezzi; Roma / Italy
1. To highlight the advantages and benefits of IR in providing high-quality cost-effective care.

2. To discuss the pivotal role of IR in multidisciplinary patient management, building bridges with other medical disciplines.
3. To describe how cutting-edge IR treatment options can contribute to the reputation and performance of trauma centres.
Panel discussion: Different ways to implement clinical IR: experience from the speakers (25 min)
615
ESR Audit - Adding value in European radiology: developing national clinical audit infrastructure
Categories: Audit, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Roman Klöckner; Lübeck / Germany

Roman Klöckner; Lübeck / Germany
QuADRANT: definitions and proposed clinical audit models (15 min)

Alexandra Karoussou-Schreiner; Luxembourg / Luxembourg
1. To appreciate key audit definitions and models of clinical audit implementation.

2. To overview QuADRANT national professional society recommendations.
3. To update the European Commission's progress on QuADRANT implementation at the national level.
The UK model of national clinical audit (15 min)

David Howlett; Eastbourne / United Kingdom
1. To overview the UK model of external direction for clinical audit in radiology.

2. To describe the results of a national UK audit and re-audit following interventions.
3. To consider the use of enablers for clinical audit uptake.
The Finnish model of national clinical audit (15 min)

Petro Julkunen; Kuopio / Finland
1. To overview the Finnish model for national external clinical audit and the challenges encountered.
2. To describe the results and impact of previous national audit projects.
3. To highlight the importance of multi-professional involvement in clinical audit.
Swiss perspective on the clinical audit (15 min)

Manuel Bondini; Lugano / Switzerland
1. To overview the Swiss model for national clinical audits.
2. To describe the quality results in the ongoing audit system.
3. To highlight the importance and the benefits of pure peer-to-peer model.
The role of the ESR national professional societies in developing European clinical audit (15 min)
616
1. To overview the ESR national professional societies network and infrastructure.

2. To present the results of previous national professional societies audits and surveys.
3. To discuss the potential for enhanced national professional society involvement in clinical audit.
Panel discussion: Is a European radiological clinical audit network a realistic possibility? (10 min)
617
CUBE 17 - EVAR nightmares: How to stay calm!

Date: March 1, 2024 | 14:00 - 14:30 CET
EVAR nightmares: How to stay calm! (28 min)

Hicham Kobeiter; Créteil / France
1. To describe factors that lead to EVAR failure.

2. To illustrate techniques and tools of bail out.
3. To offer an overview of the obtained outcomes.
618
RPS 1507 - Prostate imaging and cancer management
Categories: Genitourinary, Multidisciplinary, Oncologic Imaging

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Vibeke Berg Logager; Herlev / Denmark
Clinically significant prostate cancer detection of MRI in-bore biopsy after negative targeted plus systematic MRI/US
fusions-guided biopsy (7 min)
Author Block: M. Boschheidgen, M. Quentin, J. P. Radtke, T. Ullrich, L. R. Drewes, B. Valentin, P. Albers, G. Antoch, L. Schimmöller;
Düsseldorf/DE
Purpose: This study aimed to discuss the challenges of patients with suspicion of clinically significant prostate cancer (csPC) on
multiparametric prostate MRI (mpMRI), but negative or inconclusive MRI/US fusion-guided biopsy (FB) in clinical practice; and to
evaluate the utility of MRI in-bore biopsy (IB) in patients with discordant imaging and histopathologic results following FB.
Methods or Background: Consecutive patients who underwent IB after FB between 01/2014 and 05/2022 with Prostate Imaging
Reporting and Data System (PI-RADS) category 4 or 5 lesions on mpMRI at 3T, without histologically confirmed csPC, were included.
The primary objective was the detection rate of csPC. Secondary objectives were to analyze these cases regarding clinical
parameters, MRI parameters, and lesion localization.
Results or Findings: In the final cohort of 51 patients, the overall PC detection rate with IB was 71% and 47% for csPC. IB resulted
in a Gleason score upgrade in 55% of cases with initial low-grade PC. The detected csPC was most commonly located apical and/or
anterior. The detection rate of PC in PI-RADS category 4 was 58% and 94% in PI-RADS 5 (csPC 39% and 61%, respectively). Patients
with csPC showed significantly smaller prostate volumes, a higher PI-RADS category, a higher prostate-specific antigen density
(PSAD), and older age.
Conclusion: In a relevant proportion of patients with PI-RADS 4 or 5 but negative or inconclusive findings on previous FB and
persistent suspicion of csPC, a csPC could be verified by subsequent IB. Therefore, IB might serve as a backup solution in cases of
uncertainty.
Limitations: The limitations of this study were that it was a single-centre and retrospective design.
Ethics committee - additional information: This study was approved by the Ethikkommission Universität Düsseldorf.
Magnetic resonance imaging characteristics and oncological follow-up of patients with high grade prostate cancer (7
min)
619
Author Block: M. Boschheidgen, L. R. Drewes, B. Valentin, J. P. Radtke, P. Albers, G. Antoch, L. Schimmöller; Düsseldorf/DE
Purpose: This study aimed to analyse multiparametric MRI characteristics of patients with ISUP 4 or 5 prostate cancer (PC) on biopsy
and to investigate the correlation with biochemical recurrence (BCR) after radical prostatectomy (RPE).
Methods or Background: In this single-centre cohort study, consecutive patients with mpMRI and ISUP 4 or 5 PC at the time of
biopsy and/or RPE were retrospectively analysed. Clinical, MR-guided biopsy and mpMRI parameters were assessed. A subcohort of
patients with RPE and follow-up was analysed separately. A univariate and multivariate analysis was performed to determine
parameters which enable the identification of patients facing an elevated risk of BCR after RPE.
Results or Findings: 145 patients (mean age 70 years, median PSA 10.9 ng/ml) were analysed. 99% had PI-RADS score of 4 or 5,
48% revealed MRI T3 stage and median tumour diameter was 15mm. Cancerous areas showed a median ADC value of 668 ×10−3
mm2/s and exhibited contrast enhancement in 94% of the cases. For patients with and without BCR after RPE (n=82), MRI parameters
were different for contact length to pseudocapsule, MRI cT3 stage, and localization of the index lesion (p<0.05). PSAD and MRI cT3
stage were independent parameters for the risk of BCR when incorporating clinical, biopsy, and MRI parameters.
Conclusion: ISUP 4 or 5 PC had distinctive MRI characteristics and were detected on MRI in all included cases. PSAD and cT3 stage
on MRI are significant predictors for BCR after RPE. These results may help clinicians, to identify patients at higher risk for recurrence
at follow-up. MRI has the ability to identify high-risk prostate cancer and may play a role in clinical risk stratification and prediction of
biochemical recurrence after radical prostatectomy.
Limitations: The limitations of this study were that it was a single-centre, retrospective design.
Ethics committee - additional information: This study was approved by the Ethikkommission der Universität Düsseldorf.
Five-year outcomes after transurethral ultrasound ablation (TULSA) of localised prostate cancer (7 min)
Author Block: J. Fütterer1, D. Bonekamp2, T. Persigehl3, S. Arora4, S. Raman5, K. J. Macura6, A. Oto7, T. Tirkes8, D. Costa9;
1
Nijmegen/NL, 2Heidelberg/DE, 3Cologne/DE, 4New Haven, CT/US, 5Los Angeles, CA/US, 6Baltimore, MD/US, 7Chicago, IL/US,
8 9
Indianapolis, IN/US, Dallas, TX/US
Purpose: This study aimed to discuss the magnetic resonance imaging-guided transurethral ultrasound ablation (TULSA) of the
prostate using ultrasound to thermally coagulate tissue in-bore. Closed-loop control matches real-time feedback from MRI
thermometry to the ablative energy administered to the prescribed volume. We report the 5-year follow-up from the TACT pivotal
study.
Methods or Background: 115 men with localised prostate cancer (PCa) were enrolled across 13 sites in 5 countries. Eligibility
included Grade Group (GG) 1-2, stage≤T2b, and PSA≤15 ng/mL. Men received a single, whole-gland TULSA treatment sparing the
prostatic urethra and urinary sphincter. Primary endpoints at 1 year were PSA reduction and adverse events. Secondary endpoints
included 10-core biopsy and mpMRI at 1 year and adverse events, quality of life, PSA, and salvage treatment to 5 years.
Results or Findings: At baseline, (median [IQR]) age was 65 (59-69) years. By 5 years, PSA decreased from 6.3 (4.6-7.9) ng/mL pre-
treatment to 0.63 (0.18-1.9) ng/mL (n=68), and 25 men (21.7%) received salvage treatment, without unexpected complications (10
prostatectomy, 11 radiotherapy, 3 ADT, 1 surgery and radiation). Median prostate volume at baseline and 1 year was 37.3 and 2.8 cc
(92% decrease). At baseline and 1 year, 72/115 (63%) and 17/111 (15%) had ≥GG2 disease. By 5 years, 61/66 (92%) recovered pad-
free continence; 80/92 (87%) preserved IIEFQ2≥2 erections. 12 men incurred Grade 3 adverse events (10%), with no Grade≥4 event
or rectal injury. Intraprocedural imaging parameters, positive MRI and rising PSA at 1 year predicted the risk of salvage therapy by 5
years.
Conclusion: Effective disease control was achieved in 78% of men at 5 years after a single TULSA procedure. Favourable safety and
functional outcomes were durable to 5 years.
Limitations: Modern protocols may translate into improved rates of local control in contemporary cases. A randomized controlled
trial comparing TULSA with radical prostatectomy is underway (NCT05027477).
Funding for this study: The funding for this study was received from Profound Medical.
Expect the unexpected: unveiling discrepancies in prostate cancer diagnosis with an MRI-pathology discordance
analysis (7 min)
Arnaldo Stanzione; Napoli / Italy
620
Author Block: A. Stanzione1, K-L. Lee2, N. Sanmugalingam2, I. Rajendran2, N. Sushentsev2, I. Caglič2, T. Barrett2; 1Naples/IT,
2
Cambridge/UK
Purpose: This study aimed to investigate the underlying explanations for inconsistent results between prostate MRI and biopsy
outcomes in biopsy-naïve patients. Additionally, it seeks to shed light on diagnostic errors and potential enhancements in the prostate
cancer (PCa) diagnostic process.
Methods or Background: A retrospective analysis was conducted on 2780 biopsy-naïve patients undergoing prostate MRI at a
tertiary referral centre from October 2015 to June 2022. Patients exhibiting discordance between MRI and biopsy results were
categorized into two groups: MRI-negative/Biopsy-positive and MRI-positive/Biopsy-negative (biopsy-positive defined as Gleason score
≥ 3+4). An expert uroradiologist reviewed cases with discrepancies, reassessing PI-RADS scores retrospectively, identifying
previously unreported MRI targets, and evaluating MRI scan quality (MRI-negative/Biopsy-positive group). Categories for discrepancies
included MRI overcalls (including known pitfalls), benign pathology findings, and errors in biopsy targeting.
Results or Findings: Patients who did not undergo biopsy (n=1258), with indeterminate MRI findings (PI-RADS 3, n=204), or with
clinically insignificant PCa (Gleason score 3+3, n=216) were excluded. Of the remaining patients, 32/1102 (3%) were classified as
MRI-negative/Biopsy-positive and 117/1102 (11%) as MRI-positive/Biopsy-negative. In the MRI-negative/Biopsy-positive group, 44% of
the scans were deemed non-diagnostic quality. Upon retrospective image review, target lesions were identified in 28% of cases. In
the MRI-positive/Biopsy-negative group, 42% were attributed to MRI overcalls, while 32% had benign pathology findings explaining
the observed abnormalities on MRI. Biopsy targeting errors accounted for 11% of cases.
Conclusion: Prostate MRI demonstrated a high diagnostic accuracy, with low occurrences of discrepant findings. To further minimize
these, it is essential to ensure optimal MRI quality and expert image assessment. Common reasons for MRI-positive/Biopsy-negative
cases included benign pathology findings and MRI overcalls.
Limitations: The limitations of this study were that it was of retrospective design and single-centre.
Ethics committee - additional information: This study was approved by the Local IRB.
Integrated diagnostics for the detection of prostate cancer through the integration of data derived from a breath
analysis device, MRI, and molecular analysis: a proof-of-concept study (7 min)
Author Block: M. Pecoraro, S. Novelli, E. Messina, L. Laschena, C. Catalano, V. Panebianco; Rome/IT

Purpose: This study aimed to deliver novel, cost-effective, evidence-based, non-invasive predictive risk-adapted diagnostic pathways
for faster, earlier, more precise, accessible, and affordable detection and screening of prostate cancer.
Methods or Background: A prospective single-centre cohort study, enrolling patients with PCa suspicion who underwent MRI, MRI-
directed fusion biopsy (MRDB), was conducted. Patients also underwent breath analysis using a non-invasive device and molecular
analysis (mir-302a-5p and 367). A network-based analysis was used to identify biomarkers drivers of clinically significant PCa.
Results or Findings: Overall, 46 men were enrolled and underwent MR for clinical suspicion of prostate cancer; during the day of
the scan or the day of the biopsy procedure, patients also underwent breath analysis and plasma sampling for microRNA expression
analysis. A total of 38 patients represented the entire cohort: 12 (31.5%) were negative for PCa, 8 (21.0%) had grade group (GG) 1
PCa, and 18 (47.3%) had GG >1 PCa. The proposed additional non-invasive analysis, including clinical data, MRI biomarkers, breath
sensing and microRNAs, provided a higher net benefit with a biopsy avoidance rate of about 23% at a low disease probability.
Conclusion: The integration of multimodal data, including results from medical devices, imaging and omics can further improve the
diagnostic pathway of patients with clinical suspicion of prostate cancer.
Limitations: The main limitation is the small sample cohort.
Funding for this study: This study was supported by the Italian National Complementary Plan - D3-4-HEALTH project - funded by
the Next Generation EU.
Ethics committee - additional information: No information was provided by submitter.
Surveillance of one-year post-focal cryotherapy for clinically significant prostate cancer using mpMRI and PIRADS v2.1:
an initial experience from a prospective phase II mandatory biopsy study (7 min)
Jyothirmayi Velaga; Singapore / Singapore
621
Author Block: J. Velaga, K. J. Tay, G. Hang, Y. G. Tan, N. Lath, J. S. Yuen, M. Chua, N. T. Ngo, Y. M. Law; Singapore/SG
Purpose: This study aimed to discuss the challenges of multiparametric magnetic resonance imaging (mpMRI) surveillance post-focal
cryotherapy (FT) of prostate cancer as post-treatment artefacts alter mpMRI findings. In this initial experience, we assessed the
diagnostic performance of mpMRI in detecting clinically significant prostate cancer (csPCa) after FT.
Methods or Background: This single-centre phase II prospective clinical trial recruited 28 men with localized csPCa for FT between
October 2019 and April 2021. 12-month post-FT mpMRI was performed prior to biopsy and sensitivity, specificity, positive predictive
value (PPV) and negative predictive value (NPV) of all mpMRI-positive subjects were analysed. Chi square goodness of fit test
correlated biopsy-positive PIRADS3 (P3) and PIRADS4/5 lesions with histology grade group. One-way ANOVA test assessed the
performance of ADC values in differentiating csPCa, non-csPCa and benign lesions.
Results or Findings: Sensitivity, specificity, PPV and NPV of mpMRI were 100%, 14.28%, 53.84% and 100% for subjects with
histologically proven cancer. Correlation of PIRADS v2.1 scores with histologically proven prostate cancer was statistically significant
(p<0.5). Correlation of P3 lesions with non-csPCa was statistically significant (p<0.02535). Higher ADC value was associated with
benign histology (adjusted odds ratio OR 0.97, 95% confidence interval: 0.94, 0.99) (p=0.008). Among the malignant lesions, higher
ADC value was associated with non-csPCa (adjusted OR: 0.97; 95% CI: 0.95, 0.99) (p=0.032).
Conclusion: mpMRI is highly sensitive in detecting residual cancer. ADC values and PIRADS scores may be of value in differentiating
csPCa from non-csPCa with a potential for risk stratification of men requiring re-biopsy versus non-invasive surveillance of remnant
prostate.
Limitations: The limitation of the study was the small sample size.
Funding for this study: Funding was received from the National Medical Research Council, Singapore.
Ethics committee - additional information: This study received IRB approval (SingHealth 2018/2482)
Diagnostic performance of PSMA-PET/CT and multiparametric MRI assessed using the PI-RR scoring system for the
detection of prostate cancer local recurrence and lymph node metastases (7 min)
Ludovica Laschena; Rome / Italy
Author Block: L. Laschena, A. Dehghanpour, E. Messina, M. Bicchetti, C. Catalano, V. Panebianco; Rome/IT

Purpose: This study aimed to discuss MRI and PSMA-PET/CT recommendations for patients at risk of developing prostate cancer
(PCa) recurrence. Recently the Prostate Imaging for Recurrence Reporting (PI-RR) system has been proposed to standardize MRI
acquisition, interpretation, and reporting in case of PCa local recurrence suspicion after therapies with curative intent. The study
aimed (1) to determine the diagnostic yield and clinical value of PET/CT and MRI using PI-RR, and (2) to stratify results according to
patient’s pre/post-operative risk.
Methods or Background: Retrospective multi-center study; 119 patients with clinical suspicion of PCa recurrence underwent MRI
and PET/CT in close succession. Exclusion criteria included lack of sufficient follow-up data, pelvic radiation therapy (previous 6
months), and history of hormone therapy. PI-RR score was assigned independently by two experienced uroradiologists. PET/CT images
were reviewed by an expert nuclear medicine physician. Reference-standard was defined using histopathologic findings, follow-up
imaging or clinical response to treatment. Sensitivity, specificity, PPV, NPV, and accuracy were calculated. The intraclass-correlation
coefficient was used to determine inter-reader agreement.
Results or Findings: MRI resulted positive (PI-RR ≥3) in 69 cases, PET/CT in 48 cases. We found a statistically significant difference
in terms of rate of PCa recurrence between MRI and PET/CT (p=0.006).
MRI (PI-RR ≥3) and PET-CT reveal respectively: sensitivity of 86% and 71%; specificity of 78% and 52%, PPV of 74% and 51%; NPV of
89% and 73%; accuracy of 81% and 60% for PCa recurrence detection. At multivariate analysis, only age (0.002) and PI-RR ≥ 3
(0.001) independently correlate with recurrence.
Conclusion: MRI proved its strong diagnostic performance in detecting PCa recurrence. Therefore, low-risk patients should be
directed to MRI to rule out local recurrence; high-risk patients should be investigated using PET/CT to rule in local and distant
recurrence.
Limitations: The limitations of this study were the high-volume tertiary referral centre and limited reproducibility.
Ethics committee - additional information: This study was approved by the Institutional Ethics Committee.
Effect of preoperative PI-RADS assessment on pathological outcomes in patients who underwent radical prostatectomy
(7 min)
Qianyu Peng; Beijing / China
622
Author Block: Q. Peng, L. Xu, G. Zhang, J. Zhang, X. Zhang, X. Bai, L. Chen, Z. Jin, H. Sun; Beijing/CN
Purpose: This study aimed to assess the effect of preoperative MRI with standardized Prostate Imaging–Reporting and Data System
(PI-RADS) assessment on pathological outcomes in prostate cancer (PCa) patients who underwent radical prostatectomy (RP).
Methods or Background: This cohort study included patients who had undergone prostate MRI and subsequent RP for PCa between
2017 and 2022. The patients were divided into a PI-RADS group and a non-PI-RADS group. The pathological outcomes included
pathological T stage (pT2 vs. pT3–4) and positive surgical margins (PSMs). Patients were further stratified according to statistically
significant preoperative variables to assess the difference in pathological outcomes. A propensity score matching based on the above
preoperative characteristics was additionally performed.
Results or Findings: A total of 380 patients were included, with 201 patients in the PI-RADS group and 179 in the non-PI-RADS
group. The two groups had similar preoperative characteristics, except for clinical T stage (cT). As for pathological outcomes, the PI-
RADS group showed a significantly lower percentage of pT3–4 (21.4% vs. 48.0%, p < 0.001), a lower percentage of PSMs (31.3% vs.
40.9%, p = 0.055), and a higher concordance between the cT and pT (79.1% vs. 64.8%, p = 0.003). The PI-RADS group also showed a
lower proportion of pT3–4 (p < 0.001) in the cT1–2 subgroup and the cohort after propensity score matching. The PSM rate of cT3
patients was reduced by 39.2% in the PI-RADS group but without statistical significance (p = 0.089).
Conclusion: Preoperative MRI with standardized PI-RADS assessment could reduce the proportion of patients with non-organ-
confined PCa undergoing PR and slightly reduce the PSM rate compared with non-PI-RADS assessment.
Limitations: This is a single-centre and retrospective cohort study, and our analysis lacks complete insight into the surgical protocol
decisions of patients.
Funding for this study: This work is supported by grants from the National High-Level Hospital Clinical Research Funding (Grant No.
2022-PUMCH-A-033, 2022-PUMCH-A-035, and 2022-PUMCH-B-069), the CAMS Innovation Fund for Medical Sciences (CIFMS) (Grant No.
2022-I2M-C&T-B-019), and the 2021 Key Clinical Specialty Program of Beijing.
Ethics committee - additional information: This study was approved by the Institutional Review Board (IRB number I-22PJ1031).
ProstaPilot: prostate cancer screening by magnetic resonance imaging with a biparametric protocol, preliminary
cohort analysis after the first screening round (7 min)
Michal Standara; Brno / Czechia
Author Block: M. Standara, M. Stanik, J. Kristek, O. Majek, M. Uher; Brno/CZ

Purpose: This study aimed to validate the accuracy of biparametric magnetic resonance (bpMRI) prostate protocol in a screening
population and its potential benefits in the screening of clinically significant prostate cancer (csPC). While PSA screening can reduce
prostate cancer mortality, the risk of overdiagnosis and overtreatment of non-significant prostate cancer is a concern.
Methods or Background: The ProstaPilot prospective cohort study enrolled 422 men aged 50-69 years between May 2022 and May
2023 who had not undergone PSA or prostate MRI in the past two years. If a patient had a PSA elevation ≥ 3 ng/ml, they were advised
to undergo a systematic biopsy. If a patient had abnormal findings on bpMRI (PI-RADS 4-5 lesion), they underwent systematic and
MRI/US targeted biopsy. The bpMRI was performed with a 15-minute PI-RADS compliant non-contrast protocol, and csPC was defined
as ISUP grade ≥ 2 prostate cancer. The data underwent statistical analyses.
Results or Findings: In the first screening round, a positive MRI result was recorded in 17 (4.0%) men, and a positive PSA test result
was found in 25 (5.9%) men, of whom seven (1.7%) had positive both MRI and PSA tests. Prostate biopsy was recommended to 35
study participants (8.3%) based on one or both positive tests. Of these, three men refused biopsy, and in the MRI-positive group, eight
(1.9%) csPC and seven (1.7%) ISUP1 carcinomas were detected, while in the PSA-positive group, four (0.9%) and four (0.9%)
carcinomas detected, respectively.
Conclusion: More clinically significant cancers were detected in the MRI-positive group than in the PSA-positive group. Due to limited
capacity, MRI cannot be performed on all screened men. Further studies are needed to determine the optimal PSA level at which men
should undergo bpMRI.
Funding for this study: Funding for this study was provided by the Ministry of Health of the Czech Republic: grant nr.
NU22-09-00539 and Ministry of Health, Czech Republic – conceptual development of research organization (MOÚ, 00209805).
Ethics committee - additional information: The study was approved by Masaryk Memorial Cancer Institute Ethics committee: Ref.
Nr. 2021/1715/MOU.
Reduction of total perfusion of prostatic adenoma following prostate artery embolisation for benign prostatic
hyperplasia (BPH) (7 min)
Ailin Dehghanpour; Rome / Italy
623
Author Block: A. Dehghanpour, S. Cipollari, E. Messina, L. Laschena, C. Catalano, V. Panebianco; Rome/IT
Purpose: This study aimed to assess changes in total prostate and central gland volumes and changes in perfusion following prostate
artery embolisation in patients with symptomatic benign prostatic hyperplasia (BPH).
Methods or Background: In this prospective study, we included patients with symptomatic BPH refractory to medical therapy with
no evidence or previous history of prostate cancer. Patients underwent multiparametric prostate MRI (mpMRI) prior to endovascular
prostate artery embolisation; a follow-up mpMRI was performed 4 weeks after the procedure. Total prostate and central gland
volumes were measured on both the baseline and follow-up mpMRI. A quantitative map of the Area Under the Curve (AUC), which
represents the integral of the contrast agent concentration over time, was generated using the OleaSphere software from the
perfusion-weighted images. The mean AUC value for all voxels within the central gland was then calculated. Embolisation procedures
were performed through a femoral trans-arterial access, a microcatheter was used to catheterize prostate arteries bilaterally, and
embolisation was achieved using non-reabsorbable microparticles (300-500 um), following contrast-enhanced cone-beam CT to
confirm the correct positioning of the microcatheter.
Results or Findings: 27 patients were included in the study. Baseline total gland volume was 62 cc, central gland volume was 37,
mean AUC of central gland perfusion was 31819.28. Technical procedural success was obtained in 25/27 patients (92%). Non-major or
minor complications were reported following the procedures. A significant reduction of BPH symptoms was achieved in 22/27 patients
(81%). There was a significant reduction in total prostate and central gland volumes (16.2%, p=0.021) and mean AUC perfusion
values (14%, p=0.008).
Conclusion: Prostate artery embolisation is a safe and effective procedure for the treatment of symptomatic BPH. Central gland
volume and perfusion were significantly reduced at follow-up mpMRI compared to baseline.
Limitations: The limitation of this study was the limited sample size.
Ethics committee - additional information: The study was approved by the Institutional Ethical Committee.
Oncological and functional outcomes following MicroUS-guided focal laser ablation of localised prostate cancer:
comparison of single and multi-laser fibers settings (7 min)
François Cornud; Paris / France
Author Block: F. Cornud, A. Lefevre, M. Galiano; Paris/FR

Purpose: This study aimed to evaluate the safety, feasibility, and short-term functional and oncological outcomes of focal laser
ablation (FLA) of tumours at low risk of progression, under micro-ultrasound guidance.
Methods or Background: 58 PI-RADS>2 lesions were treated in 55 patients between July 2020 and June 2023 with a follow-up of 12
months for 29 patients. The mean age was 69±7.4 years. The mean PSA level was 7.9±3.5 ng/ml. The large tumour axis was ≤20mm.
Gleason score was 6 in 10 lesions (10/58,17.2%) and 3+4 in 43 lesions (43/58,74.5%). The first 21 (21/58,36%) and the last 37 lesions
(37/58,64%) were treated with a single and multifiber mode, respectively, using the Echolaser® system operating at 1064nm. Image
fusion was used to cover the MRI-delineated target volume. A prostato-rectal hydrodissection was performed for posteromedial
tumours. DCE-MRI was performed 4 days after FLA to evaluate the ablated volume.
Results or Findings: MicroUS visualised 53 tumours (53/58,91.4%). The multifiber mode achieved larger ablation volumes
(19.1±11.8 vs 5.9±3.5cc, p=0.0001). A prostate-rectal fistula occurred postoperatively in a patient with a posteromedial lesion, which
was treated surgically. Urinary and erectile functions were not affected by FLA. At 12 months follow-up of 34 tumours in 29 patients,
in-field targeted biopsies (TB) diagnosed 13 recurrences (13/34,38.2%, Gleason pattern 4 in 8/13 tumours, 61.5%). Ten (10/13,77%)
occurred with a single fibre setting. Four out-field recurrences (4/34,11.7%) were diagnosed. After retreatment of 6 recurrences, TB
showed, at 12 months, an 80% negative TB-biopsy rate.
Conclusion: Multifiber microUS-guided FLA is a safe and feasible modality with promising short-term oncological and functional
outcomes. Caution is required in posteromedial lesions.
Limitations: Long-term results are required to validate our findings
Funding for this study: The authors or the institution did not receive any funding to conduct this study
Ethics committee - additional information: This study was approved after written informed consent was obtained from the
patients. This single-centre study was approved by our Institutional Review Board.
The accuracy of the MRI pathway in the detection of index prostate cancer lesion: a single-centre experience (7 min)
Selahattin Durmaz; Istanbul / Turkey
624
Author Block: S. Durmaz, B. Coşkun, M. Kılıc, S. Madendere, T. Esen, M. Vural; Istanbul/TR
Purpose: This study aimed to examine the accuracy of the MRI pathway in detecting index prostate cancer (PCa) lesions in biopsy-
naive men.
Methods or Background: A retrospective review of 379 biopsy-naive men who underwent multiparametric MRI and subsequent in-
bore prostate biopsy for the high-likelihood target (PI-RADS 4-5) was conducted. We identified 110 men with PCa who underwent
radical prostatectomy as a definitive treatment and included them in this study. An analysis of the results from multiparametric MRI,
in-bore needle biopsy, and whole-mount step-section specimens was conducted to determine the accuracy of the MRI pathway in
detecting index lesions, which are defined as lesions with a higher GG and a larger volume. We also evaluated the results of in-bore
prostate biopsy and final pathology for GG group compliance.
Results or Findings: In total, 122 high-likelihood targets (PI-RADS 4: 85 targets, PI-RADS 5: 37 targets) were biopsied. Whole-mount
specimen analysis revealed multifocal involvement in 75 (68,2%) patients. A total of 83 non-biopsied tumour foci were identified at
final pathology and more than half (44/83) of these foci were harbouring ISUP GG 1 tumour. Except for one patient, the index PCa
focus was successfully detected and sampled by the MRI pathway approach.
At final pathology, upgrade and downgrade rates were 27,3% (30/110), and 10% (11/110) respectively. The remaining 69 (62,7%)
patients had concordant ISUP GG with in-bore biopsy pathology. The majority of the upgraded group was composed of ISUP GG1
patients (20/30).
Conclusion: According to our findings, the MRI pathway is a safe and accurate approach in biopsy-naive men with high-likelihood
multiparametric MRI findings.
Limitations: The limitations of this study included;
• It's retrospective design.
• Being a single-institution study with an experienced uroradiologist and uropathologist, a factor, which may limit its generalizability.
• Selection bias due to patient recruitment over 11 years.
Ethics committee - additional information: This study was approved by our Institutional Review Board and the requirement for
informed written consent was waived.
625
PC 15a - My journey in radiology research and lessons I learned for next-generation radiology
Categories: Education, Management/Leadership, Professional Issues, Research, Students

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1
The speakers, with years of professional experience in radiology, present and share how they have pioneered and gotten involved in
various radiology research projects throughout their careers. Some projects turned out to be successful and impactful, while other
projects phased out. Speakers would provide personal critical assessments on the outcome of their research projects and take-home
lessons that could be valuable in developing roadmaps for next-generation radiology.
Moderator:
Kyongtae T. Bae; Hong Kong / Hong Kong SAR China

Trial, success, failure, persistence (15 min)

It is never too late and not always a straight line (15 min)
What I would have done differently (and what not) (15 min)
Mathias Prokop; Nijmegen / Netherlands
Panel discussion: Back to the future (10 min)
626
RPS 1501 - Insights in pancreatic imaging with current and novel techniques

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
T2-weighted image radiomics nomogram to predict pancreatic serous and mucinous cystic neoplasms (7 min)
Xu Fang; Shanghai / China
Author Block: X. Fang, Y. Bian, L. Wang, C. Shao, J-P. Lu; Shanghai/CN

Purpose: To develop and validate a T2-weighted image (T2WI) radiomics nomogram for the prediction of pancreatic serous cystic
neoplasm (SCN) and mucinous cystic neoplasm (MCN).
Methods or Background: In this retrospective study, a total of 304 patients underwent MRI and surgical resection. Radiomics
features were extracted from T2WI. A clinical model was constructed with clinical and imaging characteristics. A radiomics model
included T2WI radiomics score, clinical and imaging characteristics. Multivariable logistic regression models were developed basing on
a training cohort consisting of 159 patients with SCN and 64 patients with MCN between March 2011 and December 2018. The
performance of the nomograms was determined by its discrimination, calibration, and clinical usefulness. The models were validated
in 81 consecutive patients between January 2019 and December 2021, of which 57 and 24 patients had SCN and MCN, respectively.
Results or Findings: The clinical model achieved an area under the curve (AUC) of 0.90 (95% CI 0.86–0.94) in the training cohort
and 0.83 (95% CI 0.72–0.93) in the validation cohort. The radiomics model achieved an AUC of 0.93 (95% CI 0.90–0.96) in the training
cohort and 0.87 (95% CI 0.75–0.96) in the validation cohort. The radiomics model outperformed the clinical model. The decision curve
analysis demonstrated that the radiomics nomogram was clinically useful.
Conclusion: The T2WI radiomics nomogram could preoperatively predict SCN and MCN, and may help clinical decision-making.
Limitations: Several limitations were identified: (1) this was a single-centre retrospective study; (2) levels of serum
carcinoembryonic antigen and carbohydrate antigen 19-9 were not considered since they were not measured in all patients; (3) this
prediction model only included SCN and MCN; (4) EUS results were not included in the clinical model because many patients did not
undergo EUS.
Funding for this study: This study was funded by the following: National Natural Science Foundation of China (81871352,
82171915, 82171930); Major Clinical Research Project of Shanghai Shenkang Hospital Development Center (SHDC2020CR4073); The
Natural Science Foundation of Shanghai Science and Technology Innovation Action Plan (21ZR1478500); Shanghai Science and
Technology Innovation Action Plan Medical Innovation Research Project (21Y11910300); 234 Platform Discipline Consolidation
Foundation Project (2020YPT001).
Ethics committee - additional information: The study was approved by the Biomedical Research Ethics Committee of Shanghai
Changhai Hospital.
T1 and T2 mapping quantitative MRI in the characterisation of pancreatic cystic lesions: preliminary results (7 min)
Alessandro Beleù; Treviso / Italy
627
Author Block: A. Beleù1, R. Napoli2, E. Grosso1, C. Nistri1, M. Piccino1, M. Massani1, G. Zanus1, G. Morana1; 1Treviso/IT, 2Padua/IT
Purpose: MRI characterisation of pancreatic cystic lesions is mainly based on qualitative and morphological criteria. Quantitative
imaging techniques, such as T1 and T2 relaxation times mapping, could help in their characterisation of these lesions.
Methods or Background: Patients undergoing abdominal MRI were prospectively enrolled. The T1- and T2-values of the content of
the pancreatic cyst were sampled by a single radiologist by evaluating mean, minimum, maximum and range of T1 and T2.
Dimensional and clinical-laboratory parameters were collected. Relaxation times were compared in different lesion groups using
ANOVA and Kruskall-Wallis test.
Results or Findings: Of the 350 patients enrolled, 126 were finally included (age 68±11 years, 55% male). 106 branch-duct
intraductal papillary mucinous neoplasms (IPMN-BD), two main duct (IPMN-MD), six serous cystadenomas (SCA), two mucinous
cystadenomas (MCA), two cystic neuroendocrine tumours (cNET), two pseudocysts, two walled-off necrosis (WON), one
lymphoepithelial cyst (LEC) and one cystic chronic pancreatitis were included. Mean T1 of all cystic lesions was 1985±826 ms, while
median T2 was 96 [82-126] ms. T1 values were significantly different among all lesions (p=.02). IPMN-BD demonstrated mean
(p=.02), minimum (p=.01), and maximum (p=.02) T2 values significantly different from SCA; ROC curve test showed that mean (AUC
0.78; p=.02), minimum (AUC 0.8; p=01) and maximum (AUC 0.78; p=.02) T2 values are sensitive in the differential diagnosis of IPMN-
BD vs. SCA. A significant difference in mean T1 was observed (p=.03) compared to IPMN-BD and MCA. IPMN-BD showed significant
difference in mean (p=.01), minimum (p=.01), maximum (p=.02) T1 and T2 range (p=.01) compared to WON. SCA mean (p=.01) and
minimum T1 (p=.05) were significantly different to MCA.
Conclusion: The T1 and T2 relaxation times of pancreatic cystic lesions are different in many different lesion groups.
Limitations: The relatively low number of lesions, with different distribution in the groups, was an identified limitation.
Development and validation of an imaging-based model to predict the malignant potential of intraductal papillary
mucinous neoplasm of the pancreas: comparison with international consensus guidelines (7 min)
Junghoan Park; Seoul / Korea, Republic of
Author Block: J. Park1, J. H. Kim1, J. S. Bae1, H-J. Kang1, S-Y. Choi2; 1Seoul/KR, 2Bucheon/KR
Purpose: The study aimed to develop and validate imaging-based models for predicting the malignant risk of intraductal papillary
mucinous neoplasm (IPMN).
Methods or Background: We retrospectively analysed data from 245 surgically-confirmed IPMN patients who had preoperative CT
and MRI scans for model development. Two radiologists assessed cyst size, presence of enhancing mural nodule (EMN), EMN size,
main pancreatic duct (MPD) diameter, thickened/enhancing cyst wall, abrupt MPD calibre change with distal pancreatic atrophy, and
lymphadenopathy. Multiple logistic regression models predicting malignancy risk were created using either continuous (model C) or
dichotomised variables (model D) from the significant imaging features on univariable analysis. Validation included internal (n=55)
and external (n=43) datasets with pathologically-confirmed IPMNs. Model performance was assessed using the area under the
receiver-operating characteristic curve (AUC) and compared with the Fukuoka guideline-based model (model F).
Results or Findings: Model C identified age, EMN size, MPD diameter, and lymphadenopathy as independent predictors on CT, and
age, presence of EMN, and EMN size on MRI. In model D, independent predictors were age ≥ 68, size ≥ 31 mm, EMN ≥ 6 mm, MPD ≥
7 mm, and lymphadenopathy on CT, and age ≥ 68, size ≥ 33 mm, EMN ≥ 4.5 mm, MPD ≥ 7.5 mm, and lymphadenopathy on MRI.
Model C (AUCs for external datasets, 0.785-0.899) performed slightly better than model D (AUCs, 0.736-0.912) without statistical
significance. No significant difference was observed between models C and F (AUCs, 0.827-0.952). Combining model C with the
presence of obstructive jaundice improved diagnostic performance (AUCs, 0.883-0.941) without statistical significance.
Conclusion: Our imaging-based models effectively predicted the malignant risk of IPMN, and its combination with clinical findings
enhanced diagnostic accuracy.
Limitations: Identified limitations were the retrospective nature of this study and the limited sample size.
Ethics committee - additional information: This study was approved by the institutional review board of Seoul National University
Hospital, and the requirement for informed consent was waived because of the retrospective nature of this study.
Common imaging findings and clinical effects of high-grade pancreatic intraepithelial neoplasia in the remnant
pancreas in patients with intraductal papillary mucinous neoplasms of the pancreas (7 min)
Jung Hoon Kim; Seoul / Korea, Republic of
628
Author Block: J. H. Kim, M. C. Kim, S. K. Jeon, H-J. Kang; Seoul/KR
Purpose: The study aimed to investigate the common computed tomography (CT) findings and clinical effects of HG PanIN in the
remnant pancreas in patients with intraductal papillary mucinous neoplasm (IPMN) of the pancreas.
Methods or Background: Two hundred and fifty one patients with surgically confirmed IPMNs (118 malignant [invasive
carcinoma/high-grade dysplasia] and 133 benign [low-grade dysplasia]) were retrospectively enrolled. The grade of PanIN (233
absent/low-grade and 18 high-grade) was recorded, and all patients underwent serial CT follow-up before and after surgery. Two
radiologists analysed the CT findings of high-risk stigmata or worrisome features according to 2017 international consensus
guidelines. They also analysed tumour recurrence on serial follow-up CT after surgery. Statistical analyses were performed to identify
significant predictors and clinical impact on postoperative outcomes of HG PanIN.
Results or Findings: PanIN grade showed a significant association with IPMN grade (p=0.012). Enhancing mural nodules ≥5 mm,
abrupt main pancreatic duct (MPD) changes with distal pancreatic atrophy, increased mural nodule size and MPD diameter were
common findings in HG PanIN (P<0.05). In multivariate analysis, abrupt MPD change with distal pancreatic atrophy (odds ratio [OR]
6.59, 95% CI: 2.32-18.72, <0.001) and mural nodule size (OR 1.05, 95% CI: 1.02-1.08, 0.004) were important predictors for HG PanIN.
During postoperative follow-up, HG PanIN (OR 4.98, 95% CI: 1.22-20.33, 0.025) was significantly associated with cancer recurrence in
the remnant pancreas.
Conclusion: Although PanIN is a microscopic, non-invasive precursor of invasive carcinoma, CT can be useful for predicting HG PanIN
using common features, such as abrupt MPD changes and mural nodules. In HG PanIN, extra caution is needed to monitor
postoperative recurrence during follow-up.
Limitations: Identified limitations were the retrospective nature of the study and its limited sample size.
Ethics committee - additional information: The study was approved by Seoul National University Institutional Review Board with
approval code: 2007-183-1143.
Prediction of tumour cellularity in resectable PDAC from preoperative computed tomography imaging (7 min)
Friederike Jungmann; Munich / Germany
Author Block: F. Jungmann, G. Kaissis, S. Ziegelmayer, F. N. Harder, K. Steiger, M. Makowski, R. Braren, F. Lohöfer; Munich/DE
Purpose: Pancreatic ductal adenocarcinoma (PDAC) remains a tumour entity with a poor prognosis and a 5-year survival rate below
10%. Until today, non-invasive prediction of individual patient outcomes remains an unresolved task. We previously showed a strong
association between magnetic resonance imaging-based tumour cell estimates and patient survival. In this study, we aimed to
transfer this finding to more broadly applied computed tomography (CT) imaging for non-invasive risk stratification.
Methods or Background: Discrete cellularity regions of the PDAC resection specimen were analysed by a routine histopathological
workup. Regional tumour cellularity and CT-derived Hounsfield Units (HU), as well as iodine concentrations, were regionally matched.
One-way ANOVA and pairwise t-tests were performed to assess the relationship between different cellularity levels in conventional,
virtual monoenergetic 40 keV (monoE 40 keV) and iodine map reconstructions.
Results or Findings: A statistically significant negative correlation between regional tumour cellularity in histopathology and CT-
derived HU from corresponding image regions was identified. Radiological differentiation was best possible in monoE 40 keV CT
images. However, HU values differed significantly in conventional reconstructions as well, indicating the possibility of a broad clinical
application of this finding.
Conclusion: In this study, we establish a novel method for CT-based prediction of tumour cellularity for in-vivo tumour
characterisation in PDAC patients.
Limitations: Identified limitations were: (1) the retrospective nature of the study; (2) the limited cohort size; and (3) that this was a
single institution study.
Ethics committee - additional information: The study was approved by the local institutional review board of the Technical
University of Munich.
A nomogram based on contrast-enhanced CT radiomics to preoperatively predict perineural invasion for patients with
pancreatic adenocarcinoma (7 min)
Yan Deng; Chengdu / China
629
Author Block: Y. Deng1, H. Yu1, Z. Huang1, B. Song2; 1Chengdu/CN, 2Sanya/CN
Purpose: The study aimed to develop and validate a nomogram based on contrast-enhanced CT (CECT) radiomics features and
clinical characteristics for preoperative assessment of perineural invasion (PNI) in pancreatic adenocarcinoma (PAC).
Methods or Background: The retrospective study recruited 217 PAC patients with histopathology and randomly divided them into a
training and testing cohort, at a ratio of 7:3. Radiomics features were extracted from the artery and portal venous phase of CECT.
Univariate analysis and least absolute shrinkage and selection operator (LASSO) method were applied using the 10-fold cross-
validation for feature selection in the training cohort. The selected features were integrated into the radiomics score (Rad-score). Two
experienced radiologists evaluated the status of PNI based on CECT (CTPNI). A nomogram was constructed by Rad-score and
characteristics of statistically significant differences. Calibration and classification metrics were used to evaluate the nomogram
performance.
Results or Findings: The lymph node status determined on CT (CTLN) and CTPNI were statistically different between the PNI
positive and negative groups. The AUC for the Rad-score in the training cohort was close to that in the testing cohort (AUC 0.720 and
0.640). The AUC of CTPNI was 0.610 in the training cohort and 0.675 in the testing cohort. The Rad-score, CTLN and CTPNI were used
to construct a nomogram, and it achieves favourable discrimination of PNI status with an AUC of 0.846 and 0.741 in the training and
testing cohort, respectively. The nomogram achieved best performance both in the training cohort and testing cohort compared with
Rad-score and CTPNI (P<0.05).
Conclusion: The development of a nomogram incorporating Rad-score, CTLN, and CTPNI holds promise as a valuable non-invasive
tool for the preoperative evaluation of PNI status in PAC patients.
Limitations: It is a retrospective, single-centre study with limited sample size, and no external validation group.
Ethics committee - additional information: The retrospective study was approved by the institutional review board of West China
Hospital of Sichuan University. Each patient waived informed consent.
The value of models based on tumour-related and perivascular CT radiographic features for predicting vascular
invasion in patients with pancreatic ductal adenocarcinoma (7 min)
Yu Wen; Wuhan / China
Author Block: Y. Wen1, X. Li1, X. Jiang2, X. Yang1, T. Zhao1, S. Gui1; 1Wuhan/CN, 2Hong Kong/CN
Purpose: To investigate whether automatic segmentation of tumour-vessel spatial location based on CT images can improve the
preoperative evaluation of vascular involvement in patients with pancreatic ductal adenocarcinoma (PDAC).
Methods or Background: Patients with histopathologically confirmed PDAC who underwent a dual-energy enhanced CT scan within
four weeks prior to surgery from July 2020 to June 2022, were retrospectively included. A total of 87 patients were evaluated and 12
CT segmentation mask features were extracted. Two radiologists independently assessed the involvement of five major vessels in
enhanced CT images, including the celiac artery, common hepatic artery, superior mesenteric artery, portal vein, and superior
mesenteric vein. The classification of the relationship for each of these five vessels was based on NCCN criteria as follows: no contact,
abutment (≤180° contact), or encasement (>180° contact or vascular deformity). Pixel-wise segmentation masks for tumours and
blood vessels in 3D CT images were generated using a partially labelled multi-organ segmentation model trained on public datasets.
Multiple local radiomic features related to tumour-vascular invasion, such as tumour-vessel encasement angles and vascular
deformation, were precisely extracted to describe the spatial relationship between tumours and vessels in CT images. These features
were used to predict vascular involvement. Intraoperative observations were used as the reference standard for assessing vascular
involvement.
Results or Findings: Models that included both the tumour-vessel wrapping angle and the degree of vessel narrowing (encasement)
showed better performance compared to the radiologist's assessment. The sensitivity of the models to predict vascular involvement
was approximately 56-92%, compared with 33-75% where radiologist assessment was used. There was no significant difference in
specificity between the two methods (P＞0.05).
Conclusion: Models based on tumour-related and perivascular CT radiological features can improve the prediction of vascular
invasion in patients with PDAC.
Ethics committee - additional information: This study was approved by the medical ethics committee of Tongji Medical College of
Huazhong University of Science and Technology.
Evaluation of the efficacy of neoadjuvant chemotherapy in pancreatic ductal adenocarcinoma by three-dimensional

quantitative analysis of spectral CT (7 min)
Dan Zhang; Chongqing / China
630
Author Block: D. Zhang1, X. Zhang2, Y. Zou1; 1Chongqing/CN, 2Chengdu/CN
Purpose: To explore the value of three-dimensional quantitative analysis of spectral CT in the efficacy evaluation of neoadjuvant
chemotherapy for pancreatic ductal adenocarcinoma (PDAC).
Methods or Background: This retrospective study enrolled 49 patients with PDAC who underwent contrast-enhanced spectral CT
before one month and after two months of neoadjuvant chemotherapy from October 2021 to June 2023. Conventional polyenergetic
images (PEIs), iodine-concentration (IC) images and Z-effective (Z-eff) images were generated at portal venous phase (PVP). The
whole tumour volume, IC-total and Z-eff-total were measured using a semiautomatic segment software. The changes of the longest
diameter and whole tumour volume, IC-mean and IC-total, Z-eff-mean and Z-eff-total were compared using Wilcoxon signed rank test.
The cut-off value of changes for curative effect was obtained through a receiver operating characteristic (ROC) curve analysis.
Results or Findings: The changes of the longest diameter and whole tumour volume (P＜0.001), IC-mean and IC-total (P＜0.001), Z-
eff-mean and Z-eff-total (P＜0.001) showed a significant difference, respectively. The change of the longest diameter’s cut-off value
was 0.193 (sensitivity 57.1%, specificity 93.3%, area under the ROC curve [AUC] 0.725). The change of the whole tumour volume’s
cut-off value was 0.222 (sensitivity 85.7%, specificity 73.3%, AUC 0.746). The change of the IC-total’s cut-off value was 0.191
(sensitivity 92.9%, specificity 60%, AUC 0.795). The change of the Z-eff-total’s cut-off value was 0.167 (sensitivity 85.7%, specificity
66.7%, AUC 0.800).
Conclusion: Three-dimensional quantitative parameters of spectral CT can effectively evaluate the efficacy of neoadjuvant
chemotherapy in PDAC, and the diagnostic efficacy is higher than that of lesion’s size.
Limitations: An identified limitation was the small number of patients in the study.
Ethics committee - additional information: This study was approved by the local institutional review board and informed consent
was obtained from all patients.
Utility of conventional CT and MRI parameters in grading the severity of chronic pancreatitis: can delayed
enhancement ratio be the guiding light? (7 min)
Mohak Narang; New Delhi / India
Author Block: M. Narang, S. Sharma, K. S. Madhusudhan; New Delhi/IN

Purpose: The study aimed to evaluate the role of conventional CT and MRI parameters used in routine clinical practice for grading
the severity of chronic pancreatitis (CP).
Methods or Background: This prospective study conducted from December 2019 to December 2021 included 72 patients (mean
age: 30 years; 59 men) with suspected or confirmed CP graded as equivocal, mild, and moderate-marked using composite imaging
(Cambridge classification) and endoscopic ultrasound (Rosemont’s criteria) who underwent multiphasic CT and MRI of the abdomen.
Routine parameters such as antero-posterior pancreatic thickness (APPT) and delayed enhancement ratio (DER) were evaluated on
CT. APPT, signal intensity ratio (SIR) on T1-weighted sequence, pancreatic duct contour (PDC), apparent diffusion coefficient (ADC)
and DER were evaluated on MRI. Multiphasic CT of 20 renal donors and abdominal MRI of 20 patients with renal mass served as
controls. One way ANOVA with post-hoc Dunn-Bonferroni correction was used for analysing APPT, ADC, SIR and DER. PDC were
compared using chi-square test.
Results or Findings: The study group included patients with equivocal CP (n=20), mild CP (n=18) and moderate-marked CP (n=34).
All individual parameters were significantly different between controls and moderate-marked CP (p<0.05). Additionally, DER (on CT
and MRI) was significantly different between controls and mild CP (p<0.02, cut-off = 1.77 on CT, sensitivity = 88.89% and specificity
= 75% and 3.56 on MRI, sensitivity = 83.33% and specificity = 95%). Area under curve (AUC) for DER on CT and MRI for
differentiating controls and mild CP were 0.87 and 0.91 respectively.
Conclusion: DER on CT and MRI is a simple and accurate imaging marker for grading of CP and may be useful in the diagnosis of
mild CP before the appearance of ductal changes.
Limitations: The small sample size was an identified limitation of this study.
Ethics committee - additional information: The study was approved by the Institutional Ethics Committee.
Intra- and interpatient variability of iodine concentration in pancreatic parenchyma for individualised contrast agent
application protocols in dual-layer spectral CT (7 min)
Mathis Franz Georg Konrad; Heidelberg / Germany
631
Author Block: M. F. G. Konrad, L. M. Wünschmann, T. Mokry, H-U. Kauczor, T. F. Weber, W. Stiller; Heidelberg/DE
Purpose: Verifying the quantifiability of intra- and interpatient variability of iodine concentration (IC) in pancreatic parenchyma
measured using dual-layer spectral CT for distinct individualised contrast agent (CA) application protocols in relation to liver
parenchyma.
Methods or Background: To compare the variability of IC in pancreatic parenchyma with that known for liver parenchyma, the
mean IC of healthy pancreatic parenchyma was measured for 150 oncologic patients, who were retrospectively allocated to three
groups (G1-G3, n=50 each) in a previous study. Imaging data were acquired during portal-venous contrast-enhanced thoraco-
abdominal spectral CT staging examinations at two subsequent time points. The applied CA volume was either chosen by a
radiographer based on body-mass-index resulting in identical (G1) or different (G2) volume for both examinations, or individually
calculated (G3) using a predefined function. Intra- and interpatient coefficients of variation (CV) were calculated for the measured IC
and attenuation in virtual monoenergetic image data (VMI, 40 keV) without and with normalisation to the reference structures portal
vein, aorta, and spleen, respectively.
Results or Findings: Intra- and interpatient CV for IC were low in G3 (13.8% and 21.6%), higher in G2 (23.2% and 23.6%) and mixed
in G1 (10.6% and 27.0%). Variability of IC was lower compared to liver parenchyma in G3 (-6.0% and -3.6%). Values with the highest
variability (interpatient CV for G1) profit the most from normalisation (mean CV reduction of -5.0%). Normalisation to portal vein and
aorta leads to examination-wise differences in IC no longer being significant (p>0.05). For all patient groups all CVs are lower for VMI
compared to IC (-7.6% to -0.1%).
Conclusion: Variability of IC in pancreatic and liver parenchyma is comparable. Individualised calculation of CA volume leads to lower
variability of IC. Normalisation can stabilise high-variability measurements.
Limitations: An identified limitation was the retrospective, single-centre nature of this study.
Funding for this study: This study received research grant support from Philips Healthcare.
Ethics committee - additional information: This study was approved by the ethics committee of the Medical Faculty of the
University of Heidelberg, approval number S-348/2019
Abbreviated MRI for the detection of small (<2 cm) pancreatic ductal adenocarcinoma: a preliminary study (7 min)
Daeun Choi; Seoul / North Korea
Author Block: D. Choi, S-S. Kim, J-Y. Choi, M-S. Park, Y. E. Chung; Seoul/KP
Purpose: This study aimed to evaluate the feasibility of abbreviated MRI for the detection of small pancreatic adenocarcinoma
(PDAC).
Methods or Background: This single-centre retrospective study included 76 patients with PDAC (lesion size <2 cm, T1 stage) (age,
72.7±9.3 years; male, 50.0%) and 286 patients without PDAC (age, 56.2±14.0 years; male, 60.1%). Two abdominal radiologists
reviewed the abbreviated MRIs, which consisted of a non-contrast T1-weighted image (T1WI), T2-weighted image (T2WI), and
diffusion-weighted image (DWI), in random order. The reviewers were asked to rate the diagnostic confidence on the presence of
PDAC using a 5-point Likert scale, as well as the confidence in mass delineation on each sequence (T1WI, T2WI, DWI). The presence of
accompanying findings such as abrupt pancreatic duct (PD) cut-off, upstream PD dilatation, upstream pancreatic atrophy, associated
pancreatitis, and upstream biliary dilatation, were also recorded. In addition, reduction of image acquisition time in the abbreviated
MRI protocol compared to the conventional full-protocol MRI was estimated. For diagnostic performance, accuracy, sensitivity,
specificity, NPV, and PPV of small pancreatic cancer detection were calculated.
Results or Findings: The diagnostic performance of abbreviated MR showed a sensitivity of 86.8-89.5%, specificity of 98.6-99.7%,
negative predictive value of 94.3-97.3%, positive predictive value of 96.6-98.6% and accuracy of 96.1-97.5%. There were eight
(10.5%) and six (7.8%) missed cases per reviewer, of which five cases (6.5%) were common which all lacked accompanying findings,
such as duct cut-off or upstream dilatation. There was an estimated 74.6% reduction in image acquisition time for abbreviated MRI
compared to full-protocol MRI.
Conclusion: Abbreviated MRI may serve as an alternative to full-protocol MRI in PDAC screening of high-risk patients.
Limitations: Diagnostic performance was not compared between abbreviated MRI and full-protocol MRI due to the retrospective
design.
Ethics committee - additional information: This retrospective study was approved by our institutional review board. Informed
consent was waived due to the retrospective design.
Quantitative image analysis in pancreatic cancer: might contrast agents become redundant? (7 min)
Jennifer Gotta; Frankfurt a. Main / Germany
632
Author Block: J. Gotta, V. Koch, L. D. Grünewald, S. Mahmoudi, S. Martin, T. Vogl; Frankfurt/DE
Purpose: The evolution of computed tomography (CT) technology alongside the emerging field of radiomics has opened promising
avenues in diagnostic assessment of cancer. This study seeks to explore the potential of a machine learning classifier rooted in dual-
energy computed tomography (DECT) radiomics, specifically in its ability to differentiate between malignant and benign pancreatic
lesions in non-contrast scans.
Methods or Background: In this study, a total of 100 patients who underwent third‑generation DECT between November 2018 and
October 2022 were included. Of these, 60 patients had pancreatic cancer, while 40 had normal pancreatic tissue. 107 radiomics
features per patient were extracted from non-contrast and arterial-enhanced DECT scans. To develop and validate our models, the
dataset was divided into distinct training and test subsets. Stepwise feature reduction was conducted to identify the most important
features. Subsequently, two gradient-boosted tree models were trained.
Results or Findings: The trained machine learning classifiers achieved a diagnostic accuracy of 0.97 in the arterial-enhanced model
and 0.88 in non-contrast scans with areas under the curve of 0.97 (95% CI, 0.9178-1, p<0.001) and 0.96 (95% CI, 0.9013-1, p
<0.001), respectively. No significant differences were observed between both models (p=0.52). Meanwhile, both CT-based radiomics
models exhibited similar results in distinguishing between pancreatic tumours and normal pancreatic tissue.
Conclusion: Our study demonstrates the potential for employing radiomics on non-contrast DECT imaging to differentiate pancreatic
cancer from healthy tissue. This innovative method holds promise in enhancing the early detection of pancreatic cancer, reducing
radiation exposure and ultimately improving patient prognosis.
Limitations: Firstly, this study was conducted at a single centre. Secondly, due to the retrospective study design, inherent limitations
may affect the reliability and generalisability of the findings. Finally, clinical variables were not integrated into our radiomics
methodology, even though this might have led to a further increase in diagnostic performance.
Ethics committee - additional information: This study was approved by the ethics committee of the University Hospital Frankfurt.
633

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1


Tutors
634
RC 1503 - Imaging challenging cardiomyopathies
Categories: Cardiac, Contrast Media, Imaging Methods, Professional Issues

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

The added value of dark-blood late gadolinium enhancement (15 min)

Robert J. Holtackers; Maastricht / Netherlands
1. To distinguish between conventional bright-blood and novel dark-blood LGE techniques.

2. To integrate and apply dark-blood LGE in your daily clinical routine.
3. To select the most appropriate LGE technique given the clinical request.
Arrhythmogenic ventricular cardiomyopathy in the light of the PADUA criteria (15 min)
1. To understand the fundamentals of Arrhythmogenic Ventricular Cardiomyopathy (AVC) and its clinical significance, including the
associated risk factors and diagnostic challenges.
2. To examine the PADUA criteria, a novel classification system for diagnosing AVC based on CMR findings, and learn how to apply
these criteria in clinical practice for an accurate and standardised evaluation.
3. To discuss the clinical implications of using CMR and the PADUA criteria in diagnosing and managing AVC, including their impact on
risk stratification, treatment planning, and patient outcomes.
New horizons for imaging cardiomyopathies (15 min)

Declan O'Regan; London / United Kingdom
1. To understand the principles of CMR image analysis in cardiomyopathy.

2. To appreciate the role of new imaging techniques in diagnosing cardiomyopathy.
3. To learn the future role of AI in assessing cardiomyopathy.
Panel discussion: How to recognise and approach non-ischaemic cardiomyopathies (10 min)
635
SF 15c - Low-dose effects on human health
Categories: Breast, Chest, EuroSafe Imaging/Radiation Protection, Research

ETC Level: LEVEL II
Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
The session will provide an overview of the current knowledge on low-dose effects on human health, followed by presentations on
screening (breast and lung) to clarify the radiogenic risk in this context and on patient perspectives. A panel discussion will focus on
technological advances' impact on the current dose levels.
Moderators:

An overview of the current knowledge on the low-dose effects on human health (20 min)
Dominique Laurier; Fontenay aux Roses / France
Insights on CT lung screening and radiation dose (20 min)

Mahadevappa Mahesh; Baltimore / United States
The latest advances in breast mammography screening radiation dose (20 min)
The patient's perspectives (15 min)

Panel discussion: How technology progresses are contributing to lowering the doses? (10 min)
Mahadevappa Mahesh; Baltimore / United States
Reinhard W.R. Loose; Nürnberg / Germany
Dominique Laurier; Fontenay aux Roses / France
636
E³ 1518 - Vascular imaging: all you need to know (case-based session)
Categories: Emergency Imaging, Imaging Methods, Vascular

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
At the end of this session, participants will be able to determine which imaging modality is preferred for suspected vascular disease.
They should know how to assess the important imaging features in carotid and peripheral artery disease and how to contrast the
imaging findings of acute versus chronic aortic disease.
Moderator:
Birgitta K Velthuis; Utrecht / Netherlands

Birgitta K Velthuis; Utrecht / Netherlands
Carotid disease: how to evaluate? (28 min)

Aortic disease: acute or not? (28 min)

Peripheral artery disease: getting to the bottom (28 min)

Giles Hannibal Roditi; Glasgow / United Kingdom
637
MD 9 - Head and neck cancer: new developments in imaging and treatment - recommendations for
Categories: Head and Neck, Multidisciplinary, Oncologic Imaging, Research

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

1. To highlight controversies in the treatment of extranodal spread in head and neck cancer.
2. To critically review recent evidence in imaging of extranodal tumour spread and put this in perspective of new treatment
developments.
The head and neck surgeon's perspective (10 min)

Michiel van den Brekel; Amsterdam / Netherlands

638
SF 15d - MRI-based synthetic CT and CT-like

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Alain Blum-Moyse; Nancy / France

Alain Blum-Moyse; Nancy / France
Current strategies in constituting artificial CT images (20 min)

Pedro Augusto Gondim Teixeira; LAXOU / France
1. To acknowledge the available option for generating CT-like images in clinical practice.
2. To understand the basic principles, advantages, and disadvantages of each of these methods.
3. To propose a practical approach to CT-like MRI in clinical practice for the evaluation of musculoskeletal disorders.
Applications in large joints (20 min)

Lieve Morbee; Gent / Belgium
1. To learn about deep learning-based synthetic CT technique.

2. To assess useful applications in the sacroiliac joint, the hips and spine.
3. To identify frequently missed lesions without MRI-based synthetic CT.
CT-like MRI and synthetic CT in inflammatory disorders: opportunities and limitations (20 min)
Georg Constantin Feuerriegel; Zurich / Switzerland
1. To assess common lesions associated with inflammatory disorders using CT-like MRI and synthetic CT.
2. To describe advantages and limitations of CT-like MRI and synthetic CT for evaluating inflammatory disorders.
3. To identify common pitfalls associated with CT-like MRI and synthetic CT when assessing inflammatory disorders.
Panel discussion: MRI-based CT: what are the benefits and drawbacks? (25 min)
639
HW 15Ca - Coronary artery disease and myocardial ischaemia: imaging, diagnosis, and reporting

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:


1. To become familiar with typical and atypical imaging findings of myocardial ischaemia.
2. To become familiar with clinical data and invasive coronary angiography.
3. To learn about the limits and technical drawbacks of cardiac CT and MRI.
4. To learn how to report cardiac CT and MRI using specific templates.
640
IIQ - The Radiology Odyssey: a journey home through challenges and cases
Categories: Cardiac, General Radiology, Oncologic Imaging, Paediatric

Date: March 1, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Join us in this year's Image Interpretation Quiz in which we will help Odysseus return to Ithaca by answering different intricate and
interactive cases correctly.
Moderator:
Radiology Odyssey (80 min)

Bibi Martens; Maastricht / Netherlands
Alexandra A. Ntorkou; Paris / France
Vasileios G Xydis; Ioannina / Greece
Melvin D'Anastasi; Mosta / Malta
Closing words (5 min)

641
OF 15R - EFRS Patient-Centred Care Awards
Categories: Management/Leadership, Professional Issues, Radiographers

Date: March 1, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

Improving patient care: where are we and where should we go? (10 min)
1. To learn about the importance of patient-centred care for radiographers.

2. To appreciate the current status of patient-centred care.
3. To understand opportunities for improving patient-centred care in the future.
Role of the patient advocates in improving patient care (10 min)

Cheryl Cruwys; Haute Vienne / France
1. To learn about the role of patient representatives in supporting patient care initiatives.
2. To understand the importance of patient voice in in driving patient care forward.
EFRS 2023 PCC winner. Identification and management of patients exposed to a significant radiation dose from a
fluoroscopy-guided interventional procedure: an initiative at Beaumont Hospital to enhance patient safety (10 min)
Jose Jr Binghay; Dublin / Ireland
1. To learn about the Beaumont Hospital patient safety initiative.

2. To appreciate the impact of this initiative on patient care.
3. To understand the plans for continuing the initiative in and outside of the Beaumont Hospital.
EFRS 2023 PCC runner-up. Introduction of dog therapy (animal-assisted intervention (AAI)) in radiotherapy setting to
support and enhance patient wellbeing (10 min)
Kate Fitzgerald; Cork / Ireland
1. To learn about the dog therapy initiative.

2. To appreciate how an animal-assisted intervention can improve patient care.
3. To understand the future plans for the dog therapy initiative in radiation therapy.
Panel discussion: How can we effectively promote PCC initiatives at a European level? (15 min)
642
AI-SC 6 - Quality assurance (QA) vs Post-market surveillance (PMS) of AI tools

Date: March 1, 2024 | 15:00 - 16:00 CET
Moderators:
Mansoor Fatehi; Tehran / Iran


Leon Doorn; Amsterdam / Netherlands
1. To discuss the difference between QA and PMS of AI tools in radiology.

2. To discuss the role of industry and AI users in QA and PMS.
3. To discuss the benefits of real-time monitoring of the use of an AI tool, way beyond the need to meet the EU Medical Device
Regulation.
643
EFRS 7 - Nuclear Medicine becomes Person-Centered: An Exploration in a Group.
Categories: Nuclear Medicine, Professional Issues, Radiographers

ETC Level: LEVEL I
Date: March 1, 2024 | 15:00 - 16:00 CET
This session is a focus group on the topic of Person-Centered Care in Nuclear Medicine to validate working tool and gather
suggestions from those working in other fields.
Moderator:
Ana Geão; Montijo / Portugal

Focus Group (50 min)
Closing (5 min)
644
MD 10 - European School of Radiology (ESOR) and multidisciplinary fellowships
Categories: Education, General Radiology, Multidisciplinary, Oncologic Imaging, Students

Date: March 1, 2024 | 15:15 - 15:50 CET
CME Credits: 0.5
Interview (30 min)

Justas Žilinskas; Kaunas / Lithuania
Uli Fehrenbach; Berlin / Germany
Tatiana Plescan; Chisinau / Moldova
Wrap-up of connAction (5 min)

645
CUBE 18 - Innovations in big vessels endovascular treatment

Date: March 1, 2024 | 15:30 - 16:00 CET
1. To highlight the new tools and devices.

2. To understand how to use them.
3. To discuss the advantages and improvements in clinical practice.
Moderator:

Venous stenting: case presentations (14 min)

Velio Ascenti; Milano / Italy
Fusion imaging in aortic percutaneous interventions (14 min)

Hicham Kobeiter; Créteil / France
646

Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1


647
IF 16 - Next-generation radiology
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Management/Leadership, Multidisciplinary, Research
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Radiology is an exciting speciality, but it faces many challenges - from technological advances not translating into clinical practice as
envisioned to other technological advances threatening how radiology is done and difficulties in ever more subspecialised
multidisciplinary and multi-professional environments. The next generation of radiologists and radiographers will need to face those
challenges in order to remain relevant and responsible. This session will cover current discussions and offer potential solutions and
visions for the future.
Moderator:

The cycle of life: from nice ideas to practical application (20 min)
Radiology and AI: team member or terminator? (15 min)

Building bridges: how to stay relevant in a multidisciplinary environment? (15 min)

What does the future hold for radiographers? (20 min)

Panel discussion: How do we stay relevant and responsible? (15 min)
648
SA 16 - Imaging of lung cancer in the era of molecular medicine
Categories: Chest, Oncologic Imaging

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

Genomic testing, targeted and immunotherapy for lung cancer: past, present, and future (20 min)
Robin Cornelissen; Rotterdam / Netherlands
1. To get insight in the rapidly evolving field of genomic testing in lung cancer.
2. To get an overview of numerous forms of targeted therapy that are currently available or in development.
3. To understand the mechanisms of resistance on targeted therapy and the different subsequent treatments.
Assessing lung cancer response to immunotherapy and targeted therapies (20 min)
1. To understand the difference between RECIST and iRECIST criteria.

2. To recognise immune checkpoint inhibitor pneumonitis.
3. To understand the importance of rebiopsy following progression under targeted therapy.
Tumour assessment beyond lesion measurement (20 min)

1. To appreciate the role of contrast enhanced CTs for the evaluation of lung tumours.
2. To understand the potential of MRI for tumour classification.
3. To know the role of modern PET tracers in tumour assessment.
Panel discussion: How will lung cancer imaging change in the next five years? (25 min)
649
RC 1613 - Interventional procedures: what should we be aware of?
Categories: EuroSafe Imaging/Radiation Protection, Imaging Methods, Interventional Radiology, Multidisciplinary, Physics in Medical
Imaging, Professional Issues
Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

Clinical point of view from a radiologist: what they want to see (15 min)
Siobhan Hoare; Dublin / Ireland
1. To understand typically used imaging protocols.

2. To learn what imaging techniques allow them to achieve the appropriate image quality.
3. To appreciate how their role, alongside radiographers and physicists, is vital in the optimisation process.
A patient follow-up after high-dose interventional procedures (15 min)

Jenia Vassileva; Vienna / Austria
1. To understand which trigger levels or their combinations should be used.

2. To learn when the follow-up should be done.
3. To learn how frequent tissue effects are in clinical practice using results from SAFRAD.
Long-term risks from paediatric interventional imaging: HARMONIC Project (15 min)
Marie-Odile De Radioprotection Et De Sûreté Bernier; Fontenay Aux Roses / France
1. To get an overview of WP3 of the HARMONIC project, which aims to understand the health effects of exposure to medical ionising
radiation in children.
2. To understand risks associated with cardiac interventional imaging in paediatrics.
3. To learn how to aid clinicians in managing and implementing changes to reduce radiation risks without compromising medical
benefits.
Panel discussion: How can we aid in the optimisation of interventional procedures and communication of associated
dose risks? (10 min)
650
RC 1617 - Non-trauma acute abdomen: don't miss the diagnosis
Categories: Abdominal Viscera, Emergency Imaging

ETC Level: LEVEL I
Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

Peritonitis: diagnostic approach (15 min)

Tiina Lehtimäki; Kuopio / Finland
1. To review the imaging patterns of peritonitis.

2. To provide a search pattern that can help in the differential diagnosis of peritonitis.
3. To understand how accurate diagnosis influences clinical management.
Spontaneous haemorrhage: diagnostic approach (15 min)

Ali Devrim Karaosmanoğlu; Ankara / Turkey
1. To review different causes and imaging findings of spontaneous abdominal bleeding.

2. To understand the role of imaging in detecting the source of bleeding and the most appropriate protocol.
3. To recognise the role of CT in providing guidance for proper treatment.
Ischemia: diagnostic approach (15 min)

Maria Antonietta Mazzei; Siena / Italy
1. To define the role of different imaging techniques in the diagnosis and management of acute ischemia.
2. To name and identify the occlusive and non-occlusive forms of acute ischemia.
3. To describe the CT signs of reperfusion (effective and ineffective) and their prognostic value.
Panel discussion: From imaging findings to aetiology, not always an easy task (10 min)
651
RPS 1614 - Current insights and future horizons
Categories: Artificial Intelligence & Machine Learning, Education, Professional Issues, Radiographers
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderators:
Ljubomir Popovic; Beograd / Serbia
Green radiology: a pilot study for a sustainable radiology project (7 min)

Andrea Masperi; Abbiategrasso / Italy
Author Block: A. Masperi1, F. Pala2; 1Milan/IT, 2Legnano/IT

Purpose: The purpose of this study was to estimate the energy impact of the radiology department, implement sustainable solutions
in a testing period and evaluate the effects.
Methods or Background: The project followed the energy diagnosis phases proposed by the UNI CEI/TR 11428 certification in a
period between January and June 2023. A retrospective analysis mapped the activities and energy consumption of the functional
categories involved imaging functional categories (IFC) and complementay functional categories (CFC) and 5 IFCs and 4 CFCs were
defined as inefficient respectively. Four interventions were proposed such as optimisation of consumption, reallocation of diagnostic
activities, awareness-raising in a green imaging protocol and staff training and a energy scorecard was drawn up. The proposed
actions were deployed and the activities, energy consumption of critical IFCs and CFCs, and implementation of the green imaging
protocol were monitored.
Results or Findings: A reduction in total energy consumption of 8% was achieved for IFCs, and 31.2% for CFCs against an increase
in average diagnostic activity of 11.8%. 41/156 patients (26.2%) were retrospectively evaluated with the green imaging protocol,
recording a potential reduction in energy consumption of 378.1 kW/patient (77.3%).
Conclusion: It is possible to identify with a fair degree of accuracy inefficient processes and functional devices that do not comply
with the new guidelines proposed by the European Community for reducing energy consumption. Including radiographers in the
drafting of energy dossiers offers the possibility of studying sustainable solutions that lead to concrete results, not only in terms of
energy and money savings but also in terms of quality assistance.
Limitations: We only considered electricity consumption and not energy consumption from other sources. The main limitations in
terms of green protocol are attributable to the evaluation of simulated patients, within a limited period, limited to the abdominal
study.
Caring for tomorrow: sustainable practices in medical imaging departments implemented by radiographers and policy
makers (7 min)
Switinder Singh Ghotra; Lausanne / Switzerland
652
Author Block: S. S. Ghotra, M. Champendal, L. Flaction, C. S. D. Reis; Lausanne/CH
Purpose: Global warming is one of the main public health concerns of our time. The purpose of this study was to identify salient
approaches that can reduce the environmental impact of medical imaging departments/(MID).
Methods or Background: A review was conducted following JBI methodology on PubMed, Embase and CINAHL to include studies
published after 2013 (French, English). Combinations of keywords and MeSH terms related to environmental sustainability, recycling,
medical waste and greening were applied. Three independent reviewers screened abstracts, titles and full text. Disagreements were
solved through consensus.
Results or Findings: 4630 studies were identified; 38 articles met all criteria. Most of the studies were related to developed
countries (32/38) and 6/38 were from non-developed countries. A third of the studies included were published after 2022. Articles
focused on computed tomography (9/38), magnetic resonance imaging (6/38), interventional radiology (4/38), conventional
radiography (4/38), ultrasound (2/38), mixed modalities (9/38). Seven main categories to reduce environmental impact were
identified: 1) examination justification, 2) energy consumption, 3) waste production, 4) recycling opportunities, 5) local resources
usage, 6) environmental pollution and 7) education. The study indicates the salience of sustainability analysis within quality
assurance programmes.
Conclusion: To reduce the environmental impact of MID it is important to educate healthcare professionals and to justify adequately
examinations, to control energy consumption and to improve health outcomes. Further studies need to be conducted to identify
strategies that are most effective, supporting the decision making of managers and MI professionals.
Limitations: No experimental studies were conducted to identify the most cost-effective strategies.
Burnout and job satisfaction among radiographers internationally (7 min)

Author Block: E. Kennedy, A. England, N. Moore, R. Young, M. F. F. McEntee; Cork/IE

Purpose: Burnout and low job satisfaction in healthcare can impact patient safety and staff retention. This study aims to gain
information on the factors influencing the levels of burnout and job satisfaction among radiographers internationally, which can inform
strategies for improving workforce supply and demand imbalance.
Methods or Background: An online questionnaire was developed that included demographic questions and two validated
instruments, The Maslach Burnout Inventory (MBI) and the Job Satisfaction Survey (JSS). Statistical analysis was performed using IBM
SPSS. The questionnaire was distributed to state registered diagnostic radiographers through the EFRS Research Hub in 2023 and
online through twitter, facebook and e-mail over a six-week period.
Results or Findings: 247 participants completed the questionnaire and 207 (84.5%) were female. The questionnaire had
participants from 21 countries, with 66.5% being from Ireland. The mean values for EE (20), DP (11) and PA (36) indicate moderate
levels of burnout among responding radiographers. 44.2% of radiographers were dissatisfied, 43.7% were ambivalent and only 12.1%
had overall job satisfaction. Workload, work being underappreciated, and time pressures were ranked as the top three factors
contributing to burnout. Staff numbers, workload and poor management, were ranked as the top three factors reducing job
satisfaction.
Conclusion: Burnout levels were moderate and overall job satisfaction was very low in participating radiographers. Workload and
being underappreciated were identified among many factors that impact job satisfaction and burnout.
Limitations: Despite the international intent of the study, two thirds of participants were from the Republic of Ireland.
R-AI-diographers: exploring the changing professional role and identity of radiographers in Europe in the era of
artificial intelligence (AI) (7 min)
Gemma Walsh; Chesham / United Kingdom
653
Author Block: G. Walsh1, M. F. F. McEntee2, Y. Kyratsis3, C. A. Beardmore1, C. Malamateniou1; 1London/UK, 2Cork/IE, 3Amsterdam/NL
Purpose: This study aims to gain insights into the changing roles and identities of diagnostic and therapeutic radiographers in the
era of AI. The objective is to propose ways to better support the workforce in the face of fast technological changes.
Methods or Background: Ethics approval and written informed consent was gained prior to data collection. A Europe-wide, cross-
sectional study utilising a mixed methods online survey was designed with key stakeholder feedback and translated from English into
eight languages. Snowball sampling was used for distribution via social media. All European radiographers (including students) were
eligible to participate. The survey collected data on the following areas: a) demographics, b) the perceived short-term impact of AI on
radiographer roles, c) the potential medium-to-long-term impact of AI, d) perceived opportunities and threats of AI implementation for
radiographers' roles and careers, e) the preparedness of radiographers to work with AI and f) the potential for future AI leadership
roles for radiographers.
Results or Findings: A total of 2,258 valid responses from 38 European countries were received. Country of practice, gender,
modality expertise, and years of experience impacted the responses. Training quality and quantity influence the perceptions of AI.
Despite some concerns around job security, survey responses were collectively projecting a feeling of optimism for the future of
radiographer careers and professional identity. Knowledge, additional training, job satisfaction, better patient care, financial
compensation and career advancement were presented as key motivators to professional engagement with AI.
Conclusion: This study provides insights into radiographers' attitudes towards AI implementation on the future of their professional
identity. It proposes ways to better support the workforce in harnessing the benefits of AI.
Limitations: Survey translators were not professional translators but radiographers with expert knowledge of subject terminology.
Funding for this study: CoRIPS from the College of Radiographers and City Radiography Research Fund.
Ethics committee - additional information: This study was approved by the ethics committee of City, University of London.
What radiographers need to know about explainable artificial intelligence in medical imaging? (7 min)
Author Block: M. Champendal1, H. Müller2, J. O. Prior1, C. S. D. Reis1; 1Lausanne/CH, 2Sierre/CH

Purpose: Artificial Intelligence/(AI) is seen as a "black box" and health professionals tend to not trust it, at least not fully. This study
aimed to present to radiographers the main eXplainable Artificial Intelligence/(XAI) methods available for medical imaging/(MI)
allowing them to understand it.
Methods or Background: A review was conducted following JBI methodology, searching PubMed, Embase, Cinhal, Web of Science,
BioRxiv, MedRxiv, Google Scholar for French and English studies post 2017 on explainability and MI modalities. Two reviewers
screened abstracts, titles, full texts, resolving disagreements through consensus.
Results or Findings: 1258 results were identified; 228 studies meet all criteria. The terminology used across the articles varied:
explainable (n=207), interpretable (n=187), understandable (n=112), transparent (n=61), reliable (n=31), intelligible (n=3) being
used interchangeably.
XAI tools applied to MI are mainly intended for MRI, CT and x-ray imaging to explain lung/(Covid) (n=82) and brain/(Alzheimer's;
tumours) (n=74) pathologies.
The main formats used to explain the AI tools were visual (n=186), numerical (n=67), rule-based (n=11), textual (n=11). The
classification (n=90), prediction (n=47), diagnosis (n=39), detection (n=29), segmentation (n=13) and image quality improvement
(n=6) were the main tasks explained.
The most frequently provided explanations were local (78.1%); 5.7% were global, and 16.2% combined both local and global
approaches.
Conclusion: The number of XAI publications in MI is increasing, in aid of the classification and prediction of lung and brain
pathologies. Visual and numerical output formats are predominantly used. Terminology standardisation remains a challenge, as terms
like "explainable" and "interpretable" are used as equivalent. Future work is required to integrate all stakeholders in the construction
of XAI.
Limitations: The focus was solely on recent XAI developments, leading to the exclusion of studies published before 2018, which may
have caused other tools that were explored during that period to be excluded.
Identifying the needs of radiographers working in nuclear medicine departments for the co-construction of XAI tools
for enhanced PET image reconstruction (7 min)
654
Author Block: M. Champendal1, H. Müller2, J. O. Prior1, C. S. D. Reis1; 1Lausanne/CH, 2Sierre/CH
Purpose: AI is often viewed as a 'black box,' causing some distrust among healthcare professionals. This study aimed to pinpoint the
needs of radiographers working in a nuclear medicine department to understand better the AI algorithm used for enhancing PET/CT
images.
Methods or Background: Two Focus Groups/(FG) were conducted to identify background knowledge and perspectives about AI. An
introduction about XAI was performed, and specific needs and preferences were explored through the presentation of scenarios,
giving results corresponding or not corresponding to the ground truth. The questions that XAI should explain to support radiographers
in their practice were collected, as well as main characteristics in terms of output format, confidence levels, barriers and facilitators
for AI use. Thematic analysis was carried out following the Braun & Clark Framework.
Results or Findings: Ten radiographers (aged 31-60) from various hospital settings discussed their needs for XAI. While three
currently use AI tools in PET/CT with limited training, their main requirements for XAI tools include interactivity, adaptability, user-
friendliness, and minimal workflow disruption. They found a 'trust index,' visual comparisons, example-based, and chatbots as
valuable output formats. Facilitators for XAI use included training, support, early integration, and radiographers specialists, while
barriers included lack of understanding, organisational challenges, and system capabilities.
Conclusion: Needs for XAI tools were identified, but it is important to improve radiographers' knowledge and prepare the system to
implement it without negatively impacting the workflow and patient outcomes.
Limitations: Limitations were the focus group sample size impacting generalisation and XAI knowledge.
Ethics committee - additional information: This study did not require ethics committee approval since no personal data were
collected.
The application's significance of employing deep learning image reconstruction algorithms to enhance the quality of
whole-brain CT perfusion images (7 min)
Limin Lei; Zhengzhou City / China
Author Block: L. Lei, Y. Zhou, H. Wang, X. Guo, J. Ma, Z. Wang, W. Cao, S. W. Yue; Zhengzhou/CN
Purpose: The purpose of this study was to compare the image quality of whole-brain CT perfusion (CTP) images reconstructed
utilising deep learning image reconstruction (DLIR), adaptive statistical iterative reconstruction-Veo (ASIR-V) and filtered back
projection (FBP).
Methods or Background: 54 patients who underwent CTP scans using the GE APEX CT scanner were prospectively collected. The
scanning parameters were set as follows: tube voltage at 80 kV and tube current at 150 mA. The data sets were reconstructed using
FBP, ASIR-V (40% and 80%), and DLIR (L, M, H levels). Objective parameters, including CT values and standard deviation (SD) in gray
and white matter of the frontal, temporal, and parietal lobes on the contralateral side, as well as signal-to-noise ratio (SNR) and
contrast-to-noise ratio (CNR), were measured. Perfusion parameters such as cerebral blood flow (CBF), cerebral blood volume (CBV),
mean transit time (MTT), and time to peak (TTP) were calculated at the same regions where CT values were measured. Subjective
evaluation was using the Abels' scoring system. Statistical analysis involved one-way ANOVA and Kruskal-Wallis tests for quantitative
and subjective data.
Results or Findings: As the strength levels increased, there was a gradual reduction in SD and a gradual increase in SNR and CNR,
with DLIR-H producing the best results (all P <0.001). No statistically significant differences in CT values and perfusion parameters
were observed among the six groups (P >0.05). All images met diagnostic requirements with scores exceeding 6 points. Among them,
DLIR-M/H exhibited the highest subjective image quality scores, while FBP received the lowest.
Conclusion: In CTP examinations, DLIR-M/H significantly reduced image noise, and improved SNR, CNR, and subjective image quality
compared to FBP and ASIR-V.
Limitations: This study included a small sample size, and the value of the DILR algorithm in low-dose scanning protocols has not
been evaluated.
Computer-aided ultrasound assessment of neuro-muscular motion based on optical flow: a preliminary evaluation (7
min)
Milton Santos; Aveiro / Portugal
655
Author Block: M. Santos, A. Silva, A. Silva; Aveiro/PT
Purpose: This study aimed to develop a tool to evaluate the quantity, direction of movement, and deformation of anatomical
structures during neuro-muscular activity using ultrasound images.
Methods or Background: The tool developed uses OpenCV with Python and the Tkinter library together with Python for the frontend
component. Movement tracking was obtained using the Lucas-Kanade method. The application was assessed on images acquired
during the flexion and extension movement of the wrist (from 0º to 60º wrist flexion and from 0º to 60º wrist extension). Images were
obtained in the axial and longitudinal planes of the median nerve in the middle third of the forearm using Philips Medical Systems
CX50 equipment and a 12 MHz linear probe. All the images were exported in MPEG-4 format, and the anatomical structures'
movement and deformation were characterised using local directional motion features.
Results or Findings: The developed software i) evaluates the displacement of anatomical structures, ii) provides histograms that
characterise the contribution of the different directions assumed by the structure during the movement, iii) provides a time-velocity
graph that allows characterising the displacement of structures at each moment of the structure motion, iii) allows comparative
analysis between two structures that are part of the image acquisition plan, iv) enables the evaluation of the deformation of
structures during the structure motion, v) measures distances between points during movement. Quantitative results are exported in
tabular form for further statistical analysis or for predictive machine learning models.
Conclusion: The initial results align with what is described in the literature and encourage further development and deployment in
clinical settings. Its use in the peripheral nervous and musculoskeletal system context allows us to consider its use in other contexts.
Limitations: This experimental study requires a more robust validation using a larger sample of participants.
Funding for this study: The authors received no specific funding.
A benchmark of equality, diversity and inclusion (EDI) experiences of Irish radiography students in both clinical and
academic settings (7 min)
Marion Maher; Dublin / Ireland
Author Block: M. Maher, J. McNulty, M-L. Ryan; Dublin/IE

Purpose: The research aim was to benchmark the equality, diversity and inclusion (EDI) experiences of Irish radiography students in
both clinical and academic settings. There is limited research on the radiography student experience from an EDI perspective.
Radiography students need to be comfortable interacting with a diverse patient population, with each other and academic staff.
Research shows a lack of evidence on how EDI policies impact daily experiences of radiography students.
Methods or Background: An online survey consisting of both open and closed questions on key EDI topics was used. The population
(n=396) included all students registered to both the Bachelor of Science radiography degree programme and the graduate entry
radiography programme in the leading Irish teaching centre.
Results or Findings: A 49% (n=195) response rate was achieved. 18% of respondents have a caring responsibility, 33% have a
disability, 16% identify as neurodiverse and 43% practice a religious belief. Most respondents were cisgendered females (87%). 56%
said radiography was their first choice as a third level study option. In terms of their access route to the radiography programme: 3%
accessed through the Higher Education Access Route (HEAR), 6.5% through the disability access route (DARE) and 5% entered as
mature students. 71% of respondents considered their class diverse and 12% were unsure. 38% of participants felt they contributed
to class diversity, 29% of respondents saw evidence of diversity within the radiography curriculum. Just 27% said they wanted aspects
of EDI in the curriculum.
Conclusion: Radiography students appear diverse with a notable percentage of them having diverse backgrounds and caring
responsibilities. Various access routes to the programme further demonstrate this. The perception of diversity within the sample is
mixed, highlighting the need for an inclusive culture.
Ethics committee - additional information: Full ethical approval was received from the School of Medicine UCD, Dublin.
The role of virtual reality allowing dialogue between users and virtual patients for the preclinical education of
radiographers: a pilot study (7 min)
Kengo Kato; Tsukuba / Japan
656
Author Block: K. Kato1, D. Kon2, K. Ueda2, C. S. D. Reis3, N. Ienaga1, Y. Kuroda1; 1Tsukuba/JP, 2Narita/JP, 3Lausanne/CH
Purpose: Head-mounted display-based immersive virtual reality (VR) coaching systems (HMD-VRC) in radiographers' training often
lack adequate interaction through spoken dialogue between users and virtual patients (VP). To enable comprehensive training,
including the confirmation process to prevent risks, the study aimed to develop an HMD-VRC that allows interaction with VP to
enhance preclinical learning.
Methods or Background: HMD-VRC replicates 3D models of an x-ray room and the VP, incorporating two main features: 1) a voice
bot based on Dialogflow, Google LLC, for the VP to respond to user dialogue and 2) an on-screen hint display system for learning
support. As a pilot study, first-year students (n=10) from a Radiography/Medical Imaging (MI) school without specialised education
were given a 20-minute explanation using text and physical equipment. Subsequently, they received radiography education using
HMD-VRC for approximately one hour. Objective assessments via role-playing with actual equipment were conducted before and after
HMD-VRC education. A survey regarding HMD-VRC was administered, asking about the experience and learning process.
Results or Findings: HMD-VRC enabled all students to complete radiography training from patient entry to exit. Through HMD-VRC
training, the average score of the objective assessments with actual equipment increased from 59.2% to 88.3%, nearly 1.5 times
higher. Survey results suggested a deeper understanding of the process through HMD-VRC, highlighting the importance of
communicating/discussing with the VP in training. Additionally, HMD-VRC training was found to alleviate anxiety in objective
assessments via role-playing.
Conclusion: This pilot study suggests that education using HMD-VRC with responsive VP through spoken dialogue is effective for
preclinical education, promoting practical learning for radiographers' training. This innovative approach holds promise for improving
Radiography/MI training programs, facilitating the transition between theory and practice.
Limitations: Only chest radiography was considered, other body parts will be validated.
Funding for this study: This study was partly supported by the JSPS KAKENHI (JP23K11369).
Ethics committee - additional information: This study was approved by the ethics committee of the International University of
Health and Welfare (Approval No. 23-Im-030).
The inclusion of cross-sectional imaging within preregistration radiography training curricula (7 min)
Author Block: M. Friel, R. Young, N. Moore, M. F. F. McEntee, A. England; Cork/IE

Purpose: Clinical practice is a critical component of all radiography curricula. It provides students with essential skills and training,
enabling them to develop professional characteristics for competent radiography practice. Educational institutions are challenged with
regularly reviewing and developing strategies to ensure that curricula satisfy evolving service needs. This study aimed to gain insight
into the inclusion of cross-sectional imaging (CT/MRI) within preregistration radiography training curricula.
Methods or Background: An online questionnaire, based on previous EFRS surveys, included open and closed-ended questions, and
ascertained on level of qualification, cross-sectional imaging incorporation, and tasks and assessments within programmes. The
questionnaire was distributed through the EFRS Research Hub at the European Congress of Radiology (ECR2023), and online via social
media. Closed questions were summarised using descriptive statistics, while open-ended questions were thematically analysed.
Results or Findings: Responses were received from 64 radiography educational institutions within 29 different countries. The
inclusion of cross-sectional (CT/MRI) training was reported by 57 respondents (92%) as part of their institution's radiography
programme. An increase in the amount of time dedicated to clinical training in cross-sectional imaging was reported by 24
respondents (38%). Overall, 32 individuals (53%) stated that CT should be reserved as a specialised modality for radiographers once
they attain their basic qualification, while 36 respondents (60%) agreed that MRI should also be reserved as a specialised modality.
Conclusion: The findings indicate a lack of uniformity among preregistration radiography programmes, regarding cross-sectional
imaging incorporation within their curricula. However, there appear to be divided views on the issue.
Limitations: Predominantly, responses to the survey came from countries where English is the first/dominant language.
Radiography students' viewpoints of the clinical learning environment: a cross-sectional study (7 min)
Michelle O'Connor; Dublin / Ireland
657
Author Block: M. O'Connor, J. McNulty; Dublin/IE
Purpose: Clinical practice is fundamental to the development of students' competence and professionalism. Objective instruments
have been developed to monitor the quality of clinical learning environments (CLEs) in medicine and nursing. This study is the first to
investigate radiography students' perspectives of their CLEs using the validated Undergraduate Clinical Education Environment
Measure (UCEEM).
Methods or Background: Undergraduate (UG) (n=365) and graduate-entry (GE) radiography students (n=45) from an Irish
university were invited to complete the online, self-administered, questionnaire regarding their recent clinical placement. CLEs
consisted of 25 public and 10 private hospitals nationally. The UCEEM contains 25 items congregated into two overarching
dimensions, experiential learning and social participation, with four subscales: opportunities to learn in and through work and quality
of supervision; preparedness for student entry; workplace interaction patterns and student inclusion; and equal treatment.
Results or Findings: 215 students participated (response rate 52.4%) (n=185 UG, n=30 GE), most of whom were based in public
hospitals (87.4%). The mean total UCEEM score was 107.54±17.67 (optimal range: 90–125). Mean scores for 'experiential learning
and social participation' were 74.2±12.5 (optimal range: 60–85) and 33.37±6.29 (optimal range: 30-40), respectively. Private
hospitals ranked slightly higher than public hospitals for 'opportunities to learn in and through work and the quality of supervision' (p
<0.05). The highest ranked items related to 'equal treatment' and the 'opportunity to put theory into practice'. Few statistically
significant differences existed between cohorts except for clinical supervision-related items which were scored highest by first year
students.
Conclusion: Students across both programmes reported high levels of satisfaction with their CLEs in terms of experiential learning
and social participation. This study provides valuable baseline data for comparison of Radiography CLEs.
Limitations: The study was limited to a single Irish university.
Ethics committee - additional information: The study was approved by University College Dublin (LS-C-22-153-Oconnor).
658
TF - Highlighted lectures
Categories: Abdominal Viscera, Chest, Paediatric

ETC Level: LEVEL I
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

How to survive the night shift volume two: paediatric edition (25 min)
Ilze Apine; Riga / Latvia
1. To give an overview of the most common paediatric emergencies that every radiologist should know.
2. To learn about tips and tricks related to critical care in paediatric radiology.
3. To raise the confidence of young radiologists working with paediatric patients during shifts.
How to deal with abdominal emergencies on the night shift: do not forget mesenteric ischaemia (25 min)
Lorenzo Garzelli; Paris / France
1. To give an overview of the mesenteric ischaemia pathological changes and corresponding radiological findings.
2. To learn how to differentiate different forms of mesenteric ischaemia.
3. To understand better how to compose a clear and informative report about AMI.
Pleural pathologies: the good, the bad and the ugly (25 min)
Nadine Christina Bayerl; Erlangen / Germany
1. To give an overview of different pleural pathologies and their radiological findings.

2. To learn how to differentiate pathological and benign pleural changes.
3. To learn how pleural pathologies appear in different modalities.
Panel discussion: What are the most important features from the topics of highlighted lectures that young radiologists
should add to their portfolio of radiology wisdom? (10 min)
659
RC 1606 - Artificial intelligence is here: have you noticed?
Categories: Artificial Intelligence & Machine Learning, General Radiology, Hybrid Imaging, Multidisciplinary, Nuclear Medicine,
Physics in Medical Imaging
Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:
Thomas Küstner; Tübingen / Germany

AI: just another tool in the radiologists' toolbox (15 min)

1. To learn about the fundamental processes and technical aspects of AI.

2. To learn about current AI solutions in image reconstruction and image processing.
3. To discuss the potential challenges and risks of AI in clinical practice.
AI in imaging before it reaches the radiologist (15 min)

Flemming Littrup Andersen; Copenhagen / Denmark
1. To learn about the limitations of state-of-the-art image reconstruction.

2. To learn about how AI can be used in image reconstruction and how it can use multi-modal information (e.g., the inclusion of
anatomical data from MRI).
3. To learn about future AI directions for image reconstruction.
AI as a tool to assist the radiologist (15 min)

1. To learn about the current status of AI in image analysis and reading.

2. To learn about the current status of AI in imaging to improve patient management.
3. To discuss how AI will affect clinical radiology: positively or negatively?
Panel discussion: Will AI enhance radiologist skills and improve patient management? (10 min)
660
RPS 1605 - Exploring the frontiers of AI-enhanced radiology reporting
Categories: Artificial Intelligence & Machine Learning, Education, General Radiology

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Luís Donoso Bach; Matadepera / Spain
Utilising Chat-GPT4 for conversion of free-text head and neck cancer CT reports into structured reports (7 min)
Amit Gupta; Ansari Nagar / India
Author Block: A. Gupta, K. Rangarajan, A. Garg; New Delhi/IN

Purpose: The purpose of this study was to assess the performance of generative pre-trained transformer 4 (GPT-4) for conversion of
free-text computed tomography (CT) reports of head and neck cancer (HNCa) patients into structured reports using a predefined
template.
Methods or Background: We retrieved 50 CT reports of HNCa patients from our department. A structured CT report template for
HNCa was prepared enumerating various anatomical sites and their respective subsites. Other key imaging findings were also
included - status of cervical lymph nodes, airway compromise and involvement of other neck structures and vessels. In the chat portal
of GPT-4, the prompt with best results for structured report generation was selected after prompt engineering. Generated structured
reports were evaluated by a radiologist by recording the number of places featuring missing information, misinterpreted information
and any additional information not present in the actual report. The reporting template was then modified to explicitly incorporate the
areas of mistakes and new GPT-4 responses were recorded.
Results or Findings: GPT-4 successfully converted all 50 free-text reports into structured reports. There were ten places with
missing information: tracheostomy tube (n=3), non-inclusion of sternocleidomastoid in strap muscles (n=2), extranodal tumour
extension (n=3) and contiguous involvement of neck structures by nodal mass rather than the primary tumour (n=2). Four pieces of
information were misinterpreted: abbreviations (n=2) and non-suspicious lung nodules regarded as distant metastases (n=2). GPT-4
did not indicate any additional findings. Upon the appropriate incorporation of missing areas in the reporting template and repeating
the prompts, GPT-4 rectified all the reports with no repeated or additional mistakes.
Conclusion: The GPT-4 model can be used to structure free-text radiology reports using plain language prompts and a simple yet
comprehensive reporting template.
Limitations: Fine-tuning using the GPT-4 application programming interface (API) was not done in our study.
Ethics committee - additional information: The Institutional Ethics Committee approved this study.
Enhancing radiology reporting efficiency through structured reports: a quantitative analysis (7 min)
Paweł Pawel Paczuski; Legionowo / Poland
661
Author Block: P. Bombinski1, P. P. Paczuski1, K. Paczuski1, B. Duranc1, A. Kusak2; 1Warsaw/PL, 2Lodz/PL
Purpose: This study explores the impact of structured reporting on radiologists' efficiency, standardisation, and clinician
comprehension. We propose and analyse key metrics to quantify the acceleration of report creation using predefined templates and
trigger mechanisms.
Methods or Background: Structured reports apply checklist-driven templates for standardised radiological reporting. These
templates comprise a checklist of observations and predefined triggers, ensuring systematic reporting. Radiologists can click on
checklist items, or trigger larger report segments, such as "norm" for healthy examinations, thereby reducing the need for free text
input. Structured reports can be generated using a keyboard, mouse, or voice dictation and commands.
Results or Findings: Our results were based on 10,000 reports of various radiological examinations performed by 20 radiologists.
Our proposed metrics for evaluating the efficacy of structured reporting include: number of keystrokes (each use of computer
keyboard), number of checklist clicks (each interaction with the checklist), checklist accepted suggestions (number of checklist
suggestions included in the final document), contextual accepted suggestions (number of contextual suggestions included in the final
document), keystrokes saved, time saved, and total time spent producing the document. Our findings demonstrate that structured
reporting significantly reduces keystrokes and accelerates report generation, with an average time saving of 30% compared to
conventional keyboard use. Furthermore, 84% of the checklist suggestions were accepted, improving report standardisation and
reducing errors.
Conclusion: Structured reporting offers a promising approach to enhance radiologists' reporting efficiency. By utilising predefined
templates and triggers, radiologists can create reports more rapidly while ensuring a higher level of standardisation. Clinicians benefit
from clearer, more consistent reports, which can lead to better patient care. This study underscores the potential for structured
reporting to bring significant advancements in radiology practices, establishing a new benchmark for efficiency and standardisation.
Leveraging GPT-4 for structured radiology reporting: a multilingual proof-of-concept study (7 min)
Author Block: F. Busch1, L. C. Adams2, D. Truhn3, A. Kader1, S. Niehues1, M. Makowski2, K. K. Bressem1; 1Berlin/DE, 2Munich/DE,
3
Aachen/DE
Purpose: The purpose of this study was to examine the feasibility of automated post-hoc transformation of free-text radiology
reports into structured templates using Generative Pre-trained Transformer 4 (GPT-4), a natural language processing model by
OpenAI, to standardise reporting language across institutions and enhance data extraction.
Methods or Background: 170 fictional English CT and MRI free-text radiology reports of various body regions and examinations
(e.g. MRI of the brain, spine, joints, heart, whole body, and prostate, and CT of the head, chest, spine, thorax, abdomen, and pelvis)
were generated by two board-certified radiologists. 23 structured templates were created based on previously published templates
and the RadReport Template Library. GPT-4's performance was evaluated based on the accuracy and consistency of the generated
structured reports. In addition, GPT-4's performance in chest radiography classification was tested against the medBERT.de German
medical language benchmark on 583 German chest radiography reports. All code, JSON report templates, and CT and MRI report texts
were made openly available at: https://github.com/kbressem/gpt4-structured-reporting. The web demo application can be accessed
at: kbressem.pythonanywhere.com.
Results or Findings: GPT-4 converted all 170 free-text reports into valid JSON files for automatic reading. The model identified all
radiology report key findings without any errors or omissions and consistently chose the correct report template based on the free-
text report content. In the medBERT.de chest radiography benchmark, GPT-4 surpassed the existing leading model by detecting three
pathological findings (congestion, opacity, pneumothorax) and one therapeutic device category (venous catheter).
Conclusion: This proof-of-concept study demonstrates the potential of GPT-4 in post-hoc structured radiology report text
transformation, offering a cost-effective and scalable solution for medical database organisation.
Limitations: Restricted access to GPT-4 requires potentially sensitive data to be shared with third parties. GPT-4 is not freely
available but is comparatively inexpensive at about $0.10 per report.
Funding for this study: No funding was was received for this study.
Ethics committee - additional information: Ethics approval was not required as the study did not involve patient data.
Automated anonymisation of radiology reports: comparison of publicly available natural language processing and large
language models for HIPAA-compliant data use (7 min)
Marcel Christian Langenbach; Cologne / Germany
662
Author Block: M. C. Langenbach1, B. Foldyna2, I. L. Langenbach2, V. Raghu2, T. Neilan2, I. Hadzic2, M. T. Lu2, J. Heemelaar2;
1
Cologne/DE, 2Boston, MA/US
Purpose: The purpose of this study was to leverage publicly available offline natural language processing (NLP) methods and a large
language model (LLM) to automatically remove PHI from free-text radiology reports to allow for secondary data use compliant with
HIPAA regulations.
Methods or Background: We compared two publicly available rule-based NLP models (Google's spaCy; NLPac, accuracy-optimised;
NLPsp, speed-optimised; iteratively improved on a test set of 400 randomly selected free-text radiology chest CT reports) and one
offline LLM-model (Llama-2, Meta-AI) for PHI-anonymisation. The three models were evaluated on a test set of 100 new randomly
selected chest CT reports. Precision, recall, and F1-scores were calculated. Two investigators adjudicated anonymisation performance
based on three PHI entities (dates, medical record number (MRN), and accession numbers (ACC)) and whether relevant data was
deleted.
Results or Findings: NLPac and NLPsp successfully removed all instances of highly sensitive PHIs (dates (n=333), MRNs (n=6),
ACCs (n=92)) from the test set. The LLM-model removed all MRNs, 96% of ACCs, and only 32% of dates. NLPac was the most
consistent model, with a perfect F1-score of 1.00 for MRN, ACC, and dates, followed by NLPsp, which had lower precision (0.86) and
F1-score (0.92) for dates with non-dates classified as dates in 54 instances (28 cases). The LLM-model had perfect precision for all
PHIs but the lowest recall of 0.96 for ACC (missed 4 instances in 4 cases) and 0.52 for dates (missed 134/333 instances in 69 cases)
(F1 scores 0.98 and 0.68, respectively). Importantly, NLPac and NLPsp did not remove relevant medical information, while the LLM-
model removed relevant information in 10% (n=10).
Conclusion: Pre-trained publicly available NLP models can effectively anonymise free-text radiology reports, while anonymisation
with an LLM is more prone to remove non-PHI data.
Limitations: This was a pilot study involving only chest CTs.
Ethics committee - additional information: The study was approved by the institutional review board (IRB no. 2023P002169) with
a waiver of written informed consent.
A RAdiology Data EXtraction (RADEX) tool for fast and accurate information curation from free-text reports: case study
on thyroid ultrasound examinations (7 min)
Lewis James Howell; Leeds / United Kingdom
Author Block: L. J. Howell, A. Zarei, T. M. Wah, S. Karthik, H. H. L. Ng, J. McLaughlan; Leeds/UK

Purpose: Extracting information from 'free-text' radiology reports is important for service evaluation, audit, unbiased cohort
selection, case retrieval, and translational research including labelling medical datasets for artificial intelligence analysis. While
machine learning methods have potential for automating this task, reliance on large labelled datasets and specific computing
requirements limits their usefulness. Methods using human-defined rules offer a practical alternative, enabling better utilisation of
information-rich radiology reports.
Methods or Background: Our tool, RAdiology Data EXtraction (RADEX), leverages clinicians' domain expertise for information
extraction. It uses regular expressions (regex) for efficient and flexible text pattern-matching, including wildcard and proximity
searches, Boolean logic, and negation handling. This rule-based approach enables clinical users to define complex queries without
specialised software knowledge, giving an easy-to-understand method which allows predictions to be reviewed and rules updated in
response to changing requirements and terminology. This transparency is vital for building trust and ensuring regulatory compliance.
Results or Findings: RADEX was applied to neck and thyroid ultrasound reports performed between 2015-2019 across five different
hospitals. Nineteen sonographic observations were classified, including presence and multiplicity of thyroid nodules, British Thyroid
Association thyroid nodule grading(s), altered thyroid echotexture, thyroiditis, thyroidectomy, nodal abnormality, and parathyroid
adenomata. On an expert-labelled dataset of 400 reports, RADEX achieved >90% accuracy in all classes. Processing >10,000 reports
took less than 60 seconds on a standard laptop.
Conclusion: This free open-source tool provides a scalable approach to extracting structured data from free-text reports, prioritising
usability and explainability. It leverages regex's powerful pattern-matching without requiring knowledge of its complex syntax, suiting
research and audit tasks where free-text information is key to understanding, but manual review is time-consuming and expensive.
Limitations: The main limitation of the study is that generalisability to other datasets/languages was not evaluated.
Funding for this study: Funding was provided by the UK Research and Innovation (UKRI) Engineering and Physical Sciences
Research Council (EPSRC).
Ethics committee - additional information: REC review is not required
Large language models for structured reporting with speech recognition: a comparative feasibility study (7 min)
Benedikt Kämpgen; Würzburg / Germany
663
Author Block: B. Kämpgen1, F. Jungmann2, D. Feiler3, I. Schmittel1, J. Stöckmann3, G. Arnhold2, P. Mildenberger2, C. Düber2, T. Jorg2;
1
Würzburg/DE, 2Mainz/DE, 3Munich/DE
Purpose: Conventional structured reporting (SR) using a mouse and keyboard is too time-consuming for broad user acceptance. In
2023, a dialogue system which allows radiologists to use speech recognition to fill SR-templates instead was introduced (T Jorg et al.,
Insights Imaging DOI: https://doi.org/10.1186/s13244-023-01392-y). However, the effort of training this NLP-based system for
additional SR-templates is high, e.g., modelling of concepts, synonyms, and implicit knowledge.
Methods or Background: We extended the dialogue system with a state-of-the-art causal Large Language Model (LLM), OpenAI
GPT-4, with a suitable prompt asking to translate from text to an SR-template in JSON, and compared the performance of the original
system with the extended one.
Results or Findings: The extended LLM dialogue system showed slightly lower F1 score / precision / recall compared with (Jorg et al.
2023) on the same evaluation dataset comprising 82 fictional (-0.18 / -0.29 / -0.21) and 50 real examples (-0.09 / -0.19 / +0.03) of
urolithiasis CT reports, with LLM-based fictional (0.80 / 0.70 / 0.75) and real (0.81 / 0.77 / 0.86) versus original fictional (0.98 / 0.99 /
0.96) and real (0.90 / 0.96 / 0.83).
The LLM had difficulties with implicit information; therefore, a inconspicuous kidney did not automatically lead to the negation of
pathologies such as obstructive uropathy. Also, the LLM would hallucinate "round" calculi, or assume "no calculi" in abnormal kidneys.
Conclusion: The LLM-based dialogue system requires substantially less effort of training for new templates by only requiring a
suitable prompt and JSON representation, without substantial loss of quality. A challenge for its application is to control implicit
knowledge and hallucinations.
Limitations: The study included only one closed-source LLM; beyond speech-to-structure, the LLM's generative capabilities to
interact with users were not evaluated.
Funding for this study: Funding was provided by the Bundesministerium für Bildung und Forschung (BMBF), 2022-2025, grant
agreement number: 16SV9045, project KIPA.
Ethics committee - additional information: The study did not require professional legal advice from the Institutional Review
Board, or informed consent of patients, according to the state hospital law. All patient data were fully de-identified and retrospectively
analysed.
Using large language models to improve quality and actionability of radiology reports (7 min)
Kalyan Sivasailam; Bangalore / India
Author Block: N. Kumarasami, P. N., K. Sivasailam, B. Subramanian; Bangalore/IN

Purpose: The objective of this study was to provide radiologists with fine-tuned large language models (LLMs) to enhance the quality,
clarity, and actionability of radiology reports, with specific focus on CT Abdomen reports. Current radiology reporting methods can
lead to ambiguities or misdiagnoses, especially in a remote diagnostics/teleradiology set-up. The physician/surgeon is looking for a
detailed qualitative and quantitative description of a finding based on his/her suspicions and the patient's symptoms in order to arrive
at a narrower set of differential diagnoses, as well as the appropriate procedure(s) he/she may follow in case of surgical intervention.
Our focus was on understanding the mechanics and technical architecture behind the integration of LLMs into radiology workflows to
transform the findings of a pathology into a very detailed and actionable description that is useful and relevant for the referring
physician/surgeon.
Methods or Background: The authors fine-tuned a foundational model and built a radiology-specific large language Model, focused
on CT Abdomen, using real-life reports and templates. Initially, the LLM was fine-tuned with a data set comprising 4,500 question-
answer pairs curated by the authors using instruction fine-tuning methodology. Subsequently, a retrieval-augmented generation
method was employed, refining the models with 120,000 real-world reports. In the practical set-up, radiologist interact with a chatbot-
like interface and input the pathologies. Using patient history, an initial draft report materialises using the LLM. Radiologist continue
responding to the chatbot culminating in a comprehensive report encompassing differential diagnoses.
Results or Findings: The LLMs were deployed in a remote diagnostics setup at 5C Network, India. Productivity went up by 270%.
Queries from referring physicians dropped by 76%.
Conclusion: Incorporating LLMs into radiology workflows significantly enhances report clarity and accuracy, offering a promising
avenue for optimised patient care and streamlined diagnostic processes.
Limitations: The set-up relies on radiologists identifying the primary pathologies correctly.
Funding for this study: Funding was received from 5C Network Private Limited, India.
A natural language processing pipeline to extract relevant information from mammography reports (7 min)
Nikola Cihoric; Bern / Switzerland
664
Author Block: N. Cihoric, D. Reichenpfader, K. Nairz, R. Gaio, P. Rösslhuemer, G. Cereghetti, H. Bonel, H. Von Tengg-Kobligk, K.
Denecke; Bern/CH
Purpose: Although mammography reporting is highly standardised, it results in mostly unstructured reports that are difficult to
process automatically. Our aim is to extract relevant information from mammography reports and make it available in a structured
format.
Methods or Background: We established a framework for definition and extraction of facts from the unstructured radiology reports
adopting rules and specifications from the German version of BIRADS Atlas. We defined an annotation schema that ensures
identification of relevant phrases in a report and subsequent information extraction at a high quality through an iterative and counter
check approach. This manual annotation is supported by an automated pre-annotation to simplify handling of common phrases. The
identified phrases were mapped to a standard terminology based on common data elements (CDEs) to fill a structured form with
extracted information.
Results or Findings: BERT-based large language models were then pre-trained and fine-tuned with annotations from 210
mammography reports. Thereby we also generated a LLM based on 100,000 reports in German retrieved from our hospital. An in-
depth analysis will be presented.
Conclusion: Our annotation approach separates extraction of information from the template filling, which reduces model complexity
and permits independent improvement of both tasks. The implemented pipeline is generalisable and will allow us to structure other
types of radiology reports as well. The structured information can be used for follow-up tasks such as decision support, quality
assessment or outcome prediction.
Limitations: Our large language model is almost exclusively based on German language and it is trained on texts originating from a
single hospital.
Funding for this study: Funding was received from the Innosuisse Project "Smaragd", 59228.1.
Ethics committee - additional information: Kantonale Ethikkommission Bern approved this study.
Automatic structuring of radiology reports with on-premise open-source large language models (7 min)
Piotr Woznicki; Warszawa / Poland
Author Block: P. Woznicki1, C. Laqua1, I. Fiku1, A. Hekalo1, T. Akıncı D'Antonoli2, D. Pinto Dos Santos3, B. Baeßler1, F. C. Laqua1;
1
Würzburg/DE, 2Basel/CH, 3Cologne/DE
Purpose: Large language models (LLMs) have successfully been used to extract structured elements from plain text. However, data
protection regulations restrict the use of commercial LLMs on patient data. This study evaluated state-of-the-art, on-premise LLMs for
automatically structuring free-text radiology reports.
Methods or Background: We applied a novel approach to controlling the LLM output, ensuring the validity of nested structured
reports produced by a locally hosted Llama-2 model. We compiled a data set of chest radiographs (CXR) including 200 English reports
from a publicly available MIMIC-CXR data set and 200 de-identified German reports from a university hospital. A detailed, nested
reporting template, containing 61 fields, was prepared. Ground-truth reports were annotated by a consensus of radiologists. LLM was
compared to two human readers (a junior resident in cardiology and a radiographer). Bayesian inference (Markov Chain Monte Carlo
sampling) was used to calculate Matthew's correlation coefficient (MCC) from contingency tables, setting (-0.05;0.05) as the region of
practical equivalence (ROPE).
Results or Findings: The average MCC of the LLM was 0.87 (94% HDI: 0.83; 0.90) for English and 0.67 (0.60; 0.73) for German
reports. MCC differences were all overlapping ROPE for English: LLM-Human1 0.012 (-0.037; 0.061), LLM-Human2 -0.002 (-0.05; 0.05),
Human1-Human2 -0.01 (-0.07; 0.04), and German reports: LLM-Human1 0.001 (-0.08; 0.08), LLM-Human2 -0.065 (-0.157; 0.027),
Human1-Human2 -0.066 (-0.157; 0.026).
Conclusion: Post-hoc structuring of English CXR reports using local, open-source LLMs is feasible and on par with human readers.
However, German reports were more challenging for the model. The understanding of semantics showed variability across specialties
and languages.
Limitations: The study's small sample size as well as the fact that some reports lacked information on certain findings and were
inconclusive or ambiguous were identified as limitations.
Funding for this study: This work was funded by the German Federal Ministry of Education and Research (Project: SWAG,
01KD2215A).
Ethics committee - additional information: This study was approved by an ethics committee (nr: 20221004 02). The need for
individual informed consent was waived.
Integrating AI results into standardised structured radiology reports: feasibility and implementation (7 min)
Cyril Thouly; Sion / Switzerland
665
Author Block: C. Thouly1, B. Dufour1, B. Rizk2, D. Goyard3, P. Petetin4, H. Brat1, F. Zanca1; 1Sion/CH, 2Villars-sur-Glane/CH, 3Paris/FR,
4
Berre l'Etang/FR
Purpose: One of the main challenges the industry of radiology currently faces is the integration of AI results into clinical workflow.
Healthcare professionals navigate multiple systems and interfaces (PACS, RIS, AI report), with frequently inefficient workflows. We
aimed at demonstrating the feasibility and effectiveness of integrating AI-derived results into standardised structured reports (SSR)
for radiology, enhancing clinical workflow and reporting accuracy.
Methods or Background: A collaboration was initiated among a RIS provider, an AI platform provider, and our R&D department
within a multicentric radiology network. The structured AI results were sent to the RIS via HL7 ORU messages (TCP protocol) and one
message was generated per analysis. Each element of the AI structured result was placed in an OBX segment of the HL7 message.
We use PatientID and AccessionNumber to link images on the PACS and radiology report in the RIS. Segments were subsequently
incorporated into SSR using a beacon in the RIS, undergoing multiple iterations for layout, wording, and punctuation accuracy. The
percentage of AI pre-populated fields of SSR was estimated.
Results or Findings: AI results were promptly transmitted to the RIS as HL7 messages. On accessing the report in the RIS,
radiologists encountered prepopulated SSR subsections. Currently 40 bone age and 140 knee MRI SSR templates were successfully
integrated into clinical workflows. For bone age as well as for knee MRI, the percent of pre-populated report was 60%.
Conclusion: Seamless integration of AI results into SSRs is achievable during routine clinical workflows. The active involvement of
radiologists ensures that resultant prepopulated reports align with their requirements.
Limitations: The success of this integration hinges on AI vendors delivering structured and standardised results. Inaccurate AI results
present potential liability concerns for radiologists due to the risk of transmitting unchecked erroneous reports.
Structured reporting for efficient epidemiological and in-hospital prevalence analyses of pulmonary embolism (7 min)
Tobias Jorg; Mainz / Germany
Author Block: T. Jorg, M. C. Halfmann, D. Graafen, C. Düber, P. Mildenberger, L. Müller; Mainz/DE

Purpose: Structured Reporting (SR) not only offers advantages regarding the report quality but, as an IT-based method, also the
opportunity to aggregate and analyse large, highly structured data sets (data-mining). In this study, a data-mining algorithm was
used to calculate epidemiological data and in-hospital prevalence statistics of pulmonary embolism (PE) by analysing structured CT
reports.
Methods or Background: All structured reports for PE CT scans from the last 5 years (n = 2790) were extracted from the SR
database and analysed. The prevalence of PE was calculated for the entire cohort and stratified by referral type and clinical referrer.
Distributions of the localisations of PEs (central, lobar, segmental, subsegmental, left-sided, right-sided, bilateral) were calculated,
and the occurrence of right heart strain was correlated with the localisations.
Results or Findings: The prevalence of PE in the entire cohort was 24% (n = 678). The median age of PE patients was 71 years (IQR
58 – 80). The sex distribution was 1.2/1 (M/F). Outpatients showed a lower prevalence of 23% compared to patients from regular
wards (27%) and intensive care unit (30%). Surgically referred patients had a higher prevalence than patients from internal medicine
(34% vs 22%). Patients with central and bilateral PEs had a significantly higher occurrence of right heart strain compared to patients
with peripheral and unilateral embolisms.
Conclusion: Data-mining of structured reports is a simple method by which to obtain prevalence statistics, epidemiological data, and
the distribution of disease characteristics, as demonstrated for the use case of PE. The generated data can be helpful for multiple
purposes, such as for internal clinical quality assurance or scientific analyses. To benefit from these, consistent use of SR is required
and therefore recommended.
Limitations: The study is limited by its single-centre design.
Funding for this study: This study received no outside funding
Ethics committee - additional information: This study did not require professional legal advice from the Institutional Review
Board, or the informed consent of patients, according to state hospital law. All patient data were fully de-identified.
Extracting information from unstructured MRI reports with a local open-source GPT model (7 min)
Bastien Le Guellec; Lille / France
666
Author Block: B. Le Guellec, A. Lefevre, C. Bruge, L. Hacein-Bey, J-P. Pruvo, G. Kuchcinski; Lille/FR
Purpose: We set out to use a local open-source GPT model to automate information extraction tasks from unstructured MRI reports.
We calculated its performance on reports from emergency brain MRIs performed for patients with headaches.
Methods or Background: All consecutive radiological reports from a French quaternary centre in 2022 were retrospectively
reviewed. Two radiologists identified MRIs that were done for headaches. Four radiologists scored reports' conclusions as normal or
abnormal. Abnormalities were labelled as either headache-generating or incidental. In parallel, Vicuna, an open-source GPT large
language model, performed the same tasks. Vicuna's performances were evaluated using the radiologists' consensus as the gold
standard.
Results or Findings: A total of 2398 reports were identified, of which 595 included headache in their indication. Median patient age
was 35; 68% were female. The overall rate of causal findings in outpatients with headache was 23% (135/595). Our GPT-based
method had an accuracy of >95% for simple information extraction tasks such as indication of the exam, patient sex and age, use of
contrast medium injection and study categorisation as normal or abnormal. Vicuna's accuracy was 82% for the most complex task of
causality inference between an abnormal MRI finding and symptoms.
Conclusion: We found that an open-source GPT model can extract information from radiological reports with excellent accuracy
without further training. We hypothesise that this method could also be applied to any information extraction task relying on
unstructured medical records.
Limitations: Due to the monocentric design of our study, we could not test for variability in reporting styles or language. Further
studies will be needed to explore the adaptability of the proposed framework, even though it is expected to be high based on ability
of generative language models to handle various languages seamlessly.
Funding for this study: No specific funding was received for this study.
Ethics committee - additional information: This study was approved by the IRB of Lille University Hospital.
667

Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1


Tutors
668
Meets 16 - Education and training of Australian medical radiation professionals
Categories: Education, Multidisciplinary, Professional Issues, Radiographers, Students

Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1
The session aims to introduce the education standards for radiographers and radiation therapists in Australia. At the end of the
session, the attendees should appreciate the challenges in education delivery within Australia and become familiar with advanced
practice opportunities for Australian radiographers and radiation therapists.
Moderators:

Education, training, and professional development for medical radiation technologists in Australia (25 min)
Carolyn Louise Heyes; Taylors Lakes, VI / Australia
Formalisation of the advanced practice: challenges and opportunities, Australian experience (25 min)
Carolyn Louise Heyes; Taylors Lakes, VI / Australia
669
EFRS 5 - FORCE Project - Bring your own laptop

ETC Level: LEVEL I
Date: March 1, 2024 | 16:00 - 18:00 CET
This session aims to provide an overview on the FORCE project tools.
Moderator:

Hands-on (bring your own laptop) (100 min)
Closing (10 min)

670
RPS 1610 - Pelvis, hip, and bone lesions
Categories: Imaging Methods, Musculoskeletal, Oncologic Imaging

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Milko Charles De Jonge; Amsterdam / Netherlands
Can we differentiate acute osteomyelitis and vaso-occlusive crises in sickle cell disease patients? (7 min)
Florian Nabet; Paris / France
Author Block: F. Nabet; Paris/FR

Purpose: Vaso-occlusive crisis (VOC) is the most common complication in individuals with sickle cell disease (SCD), but distinguishing
it from acute osteomyelitis, which is also frequent, poses a daily clinical challenge. Some rare forms of VOC, localised in the
periosteum (pVOC), are also difficult to differentiate from osteomyelitis with clinico-biological features. The aim of this study was to
better characterise these two entities with imaging to improve the management of these fragile patients, particularly to avoid overly
frequent antibiotic use.
Methods or Background: From 2008 to 2023, 32 SCD patients with painful periosteal crisis and available MRI data were included
(15 women), ranging in age from 17 to 39, with a baseline haemoglobin level of 8.25 (8-8.8), 27 (85%) with SS genotype and 16
(50%) with Bantu/Bantu haplotype. The formal diagnosis (osteomyelitis vs pVOC) was confirmed by the clinician team. In imaging, two
independent radiologists blindly reviewed the MRIs and recorded simple descriptive criteria, including the presence or absence of
subperiosteal collections, their size, cortical abnormalities, soft tissue oedema, circumferential character, T1 signal with fat saturation,
T2 signal, type of enhancement, and associated bone marrow abnormality.
Results or Findings: T1 hyperintensity with fat signal saturation was exclusively found in cases of periosteal pVOC (8 cases, 42,9 %
of pVOC cases). Furthermore, circumferential oedema was never observed in cases of pVOC (3 cases, 57,1% of osteomyelitis cases).
Interobserver reproducibility was very good (Cohen's kappa = 0.82).
Conclusion: MRI appears to be a valuable tool to differentiate VOC and osteomyelitis in SCD patients with painful periosteal crises.
Limitations: This is a retrospective monocentric study with some missing data.
Within our data set, there seems to be a higher proportion of SCD patients with vaso-occlusive phenotype than in the general SCD
population.
Ethics committee - additional information: This study was approved by the HEGP ethics committee.
Hip muscle size and density are associated with trochanteric fractures of elderly women (7 min)
Pengju Huang; Beijing / China
671
Author Block: P. Huang, Y. Ge, A. Yu, L. Wang, X. Cheng; Beijing/CN
Purpose: We aimed to investigate the differences in hip muscle area and density between older patients with femoral neck (FNF) and
trochanteric fractures (TRF)
Methods or Background: A total of 554 older women patients were enrolled, including 314 FNF (77.02 ± 7.15 years) and 240 TRF
(79.70 ± 6.91 years) for the comparisons. The area and density of the gluteus medius and minimus muscle (G.Med/MinM) and the
gluteus maximus muscle (G.MaxM) were measured by CT. Total hip (TH) areal bone mineral density (aBMD) and femoral neck aBMD
(FNaBMD) were measured by quantitative CT. A cut-off of 80 years was used to stratify the cohort and to further explore the age-
specific relationship.
Results or Findings: For the total subjects, all these muscle parameters were higher in the FNF group than in the TRF group (P
<0.001). The muscle parameters except for the G.Med/MinM density were significantly correlated with hip fracture typing after
adjustment for age, BMI, and THaBMD. In the age ≧ 80 group, no statistically significant correlation was found between all hip muscle
parameters and fracture types. In contrast, in the age < 80 group, interestingly, after adjustment for age, BMI, and THaBMD, the
associations between G.MaxM density, G.MaxM area, G.Med/MinM density, and G.Med/MinM area and fracture type were all
statistically significant.
Conclusion: Our results indicate that in older women, especially under 80 years of age, gluteus muscle parameters are related to
trochanteric fractures.
Limitations: Firstly, this study features a cross-sectional design, and subsequent longitudinal cohort studies are warranted to further
investigate the relationship between gluteal muscles and fracture types. Secondly, in the measurement, we chose to measure the
healthy side to replace the data on the fractured side, which may be biased.
Funding for this study: This work is supported in part by the National Key Research and Development Program of China (No.
2020YFC2004902), National Key R&D Program of China (2021YFC2501700), National Natural Science Foundation of China (grant no.
81971617), Beijing Hospitals Authority Youth Programme (code: 20200402), and Beijing Hospitals Authority Clinical Medicine
Development of Special Funding Support (code: ZYLX202107).
Ethics committee - additional information: This study was approved by the ethics committee of Beijing Jishuitan Hospital.
Long-term active MRI-surveillance of low-grade central cartilage tumours of the long bones: towards an optimal
assessment of imaging characteristics (7 min)
Jacoba De Rooy; Nijmegen / Netherlands
Author Block: J. De Rooy, C. Deckers, D. Koopmanschap, E. Dierselhuis, B. Schreuder, I. Van der Geest, M. Prokop; Nijmegen/NL
Purpose: Determining MRI characteristics that contribute to the understanding of the natural course of low-grade central cartilage
tumours (CCTs) of the long bones during long-term active MRI-surveillance.
Methods or Background: 92 patients (31 male, 61 female) with 92 low-grade CCTs with a minimum MR imaging follow-up of 48
months (range: 48-174, mean: 74.7) and at least one baseline and one follow-up MRI were retrospectively included. Patient age at
diagnosis was 20 to 76 years (mean: 50.4 years); no aggressive MR features (cortical destruction, soft tissue mass) were present and
no biopsy had been performed. MRI characteristics (size, cortical scalloping < or ≥ 10% of circumference, intralesional fat entrapment
(FE); fatty replacement (FR) only on last MRI) were scored and CCTs were labelled as in regression (R; increasing FE and/or FR,
decreasing size (>3mm), in progression (P; loss of FE and/or increasing size (>3mm), increasing scalloping) or stable (S).
Results or Findings: The majority of low-grade CCTs were labelled S or R (81/92, 88%); 68 out of 92 patients showed initial FE
(74%) and 29/68 also showed FR. Out of 24 patients without initial FE, 13 (54%) developed FE over time and 8/13 also developed FR.
Out of 11 patients (mean 33.4 years, 20-56 years) with growth (P, 12%), 6 showed initial FE (2/6 developing FR) and 4 developed FE
(3/4 also developing FR). 1 CCT with developing FE showed new scalloping. Only one CCT showed growth without FE. None developed
aggressive MRI characteristics.
Conclusion: Active MRI-surveillance can be safely recommended for low-grade CCTs of the long bones as none developed aggressive
MRI characteristics during long-term follow-up. MRI characteristics FE and FR might predict benign behaviour of CCTs and should be
assessed on follow-up MRI.
Longitudinal changes in magnetic resonance imaging biomarkers of the gluteal muscle groups and functional ability in
Duchenne muscular dystrophy: a 12-month cohort study (7 min)
Yu Song; Chengdu / China
672
Author Block: Y. Song, H. Xu, R. Xu, H. Fu; Chengdu/CN
Purpose: The purpose of this study was to explore MRI biomarkers of the gluteal muscle groups as predictors of motor function
decline in Duchenne muscular dystrophy (DMD) by characterising longitudinal progression over 12 months.
Methods or Background: 112 participants with DMD were longitudinally enrolled and underwent MRI examination of the gluteal
muscles to determine fat fraction (FF) and longitudinal relaxation time (T1) values. The North Star Ambulatory Assessment (NSAA)
and timed functional tests (TFTs) were performed. All participants returned for follow-up at an average of 12 months and were divided
into two subgroups (functional stability/decline group) based on changes in TFTs over 12 months. Univariable and multivariable
logistic regression methods were used to explore the risk factors associated with future motor function decline in DMD.
Results or Findings: For the functional decline group, all T1 values decreased, and FF values increased significantly over 12 months
(P <0.05). For the functional stability group, only the FF of the flexors and abductors increased significantly over 12 months (P <0.05).
The baseline T1 value was positively correlated with NSAA and negatively correlated with TFTs at the 12-month follow-up (P <0.001),
while the baseline FF value was negatively correlated with NSAA and positively correlated with TFTs at the 12-month follow-up (P
<0.001). Multivariate regressions showed that increased FF of the abductors was associated with future motor function decline (OR
=1.104, 95% CI: 1.026~1.187, P =0.008), with an area under the curve of 0.874.
Conclusion: FF of the abductors is a powerful predictor of future motor function decline in DMD patients over 12 months,
underscoring the importance of focusing on it early in patients with DMD.
Limitations: We did not further explore the potential of MR biomarkers to predict future loss of ambulation, we can further explore
this in future longer-term follow-up studies.
Funding for this study: Funding was received from the National Natural Science Foundation of China (81971586); (82071874);
(81901712); (82271981).
Ethics committee - additional information: This prospective study was reviewed and approved by the Institutional Review Board
(IRB), and the clinical trial registration (registration number: ChiCTR1800018340) was completed.
Metal artifact reduction strategies for clinical photon counting computed tomography of total hip arthroplasty: a
volumetric quantitative and qualitative phantom study (7 min)
Author Block: R. P. Marcus, G. C. Feuerriegel, A. A. Marth, D. Nanz, R. Sutter; Zurich/CH

Purpose: The purpose of this study was to systematically evaluate the impact of various scanning and reconstruction modes on both
metal artifact volume and overall image quality in the context of a hip prosthesis phantom, acquired on a first generation clinical
photon-counting detector computed tomography (PCCT).
Methods or Background: A total hip prosthesis phantom was scanned on a PCCT (Naeotom Alpha, Siemens Healthineers)
employing four distinct modes at 140 kV and constant dose of 7 mGy: Quantumplus (Q+), UHR-Quantumplus (UHR-Q+), QuantumSn
(Q-Sn) and UHR-QuantumSn (UHR-Q-Sn); Sn = tin filter. Polychromatic and virtual monoenergetic images (VMI) were reconstructed
with and without iterative metal artifact reduction (iMAR, Siemens Healthineers). Artifacts were quantified using a 3D printing
software and image quality was evaluated by two radiologists.
Results or Findings: Tin filter reduced artifact volume in polychromatic reconstructions by 14% (298 ml (Q-Sn) vs 347 ml (Q+) and
310 ml (UHR-Q-Sn) vs 360 ml (UHR-Q+)). iMAR reduced the metal artifact volume by 46 – 57% with UHR-Q+ images achieving the
lowest artifact volume at 150 ml. In VMI, the smallest total artifact volume was quantified at 130 keV with Q+ (150 ml) and UHR-Q+
(164 ml), at 120 keV with Q-Sn (169 ml) and UHR-Q-Sn (172 ml). iMAR reduced artifact volume in VMI: 130 ml in Q+ (150 keV), 140 ml
in UHR-Q+ (160 keV), 134 ml in Q-Sn (150 keV) and 140 ml (UHR Q-Sn at 190 keV). Best subjective image quality was achieved for
VMI Q+ with iMAR (65 keV), polychromatic UHR-Q+ with iMAR, VMI Q-Sn with iMAR (100 keV), polychromatic Q-Sn with iMAR, VMI
UHR-Q-Sn (100 keV) and polychromatic UHR Q-Sn.
Conclusion: VMI or polychromatic images using tin filter, UHR and iMAR achieve the strongest artifact reduction and best image
quality.
Limitations: This was a single phantom size study.
Funding for this study: This study received in house funding.
Ethics committee - additional information: This was a phantom study, hence no ethics committee approval was needed.
Evaluation of lower extremity skeletal muscle fibrosis in peripheral arterial disease (PAD) using extracellular volume
fraction (ECV) from dual-layer spectral CT (7 min)
Ningning Ding; Xi'an / China
673
Author Block: N. Ding, L. Zhou, X. Zhang, Y. Han, Y. Cui, X. Huang, P. Cao, X. Zhang, J. Yang; Xi'an/CN
Purpose: Objectives Peripheral arterial disease (PAD) is characterised by arterial occlusion and fibrosis in the lower extremities.
While extracellular volume fraction (ECV) obtained from dual-layer spectral CT has been established as a biomarker for myocardial
fibrosis, its application in assessing lower extremities affected by PAD remains unexplored. This study aimed to assess the clinical
feasibility of employing ECV for quantifying calf muscle fibrosis and comparing it between normal controls (NC) and PAD patients.
Methods or Background: From October 2022 to February 2023, we recruited patients with PAD as well as sex and age-matched
diabetics without peripheral arterial disease (ankle/brachial index ＞0.9, and＜1.3 ) as a control group. All participants underwent late
iodine enhancement on lower extremity dual-layer spectral CT to determine the ECV of their calf muscles and the differences between
the NC and PAD groups.
Results or Findings: A total of 53 patients (20 in the NC group and 33 in the PAD group) were recruited. The NC group exhibited
significantly lower mean ECV compared to the PAD group (19.05% vs 28.89%, respectively, P < 0.001). Among PAD patients, those
with collateral vessels had a slightly lower mean ECV than those without (26.89% vs 30.82%, respectively, P = 0.042). Additionally,
patients with intermittent claudication in the PAD group had lower mean ECV values compared to patients with critical limb ischaemia
(26.98% vs 33.52%, respectively, P = 0.028).
Conclusion: The evaluation of skeletal muscle fibrosis in PAD using ECV obtained from dual-layer spectral CT is both feasible and
informative. ECV serves as a valuable tool in identifying the severity of PAD and assessing collateral vessel formation in affected
patients.
Ethics committee - additional information: The study was approved by the ethics committee of the First Affiliated Hospital
(ethics approval number: XJTUIAF2021LSY-223).
Multiparametric MR imaging of the sacroiliac joints in patients with axial spondyloarthritis (7 min)
Qiao Zhu; Beijing / China
Author Block: Q. Zhu, C. Ren; Beijing/CN

Purpose: One MULTIPLEX (MTP) scan offers multiple sets of images, including but not limited to: composited PD, T2∗, T1 map. The
objective of this study was to test whether MTP sequence of the sacroiliac joints (SIJs) might help in identifying patients with
spondyloarthritis (SpA).
Methods or Background: This study included 20 patients with axial SpA (15 males; mean age: 34 ± 5 years; range, 20–42) and 20
controls (15 males; mean age = 30 ± 7 years; range = 20–40) who prospectively underwent SIJs MRI using an oblique coronal MTP
sequence. PD, T1, T2∗ maps obtained from MTP sequence were used to draw regions of interests in the cartilaginous part of the SIJs.
Intra- and interobserver reproducibility of quantitative values were calculated after independent assessment by two radiologists.
Results or Findings: The PD values of cartilage in patients with SpA measured by two reviewers (Reader 1: 117.109 ±14.106 ms,
Reader 2: 119.634 ±15.275 ms) were significantly higher than those of healthy volunteers (Reader 1: 108.377 ±14.387ms, Reader 2:
107.944 ±16.720ms), P <0.05. The T2* values of cartilage in patients with SpA (Reader 1: 22.336 ±1.413 ms, Reader 2: 23.421
±1.536 ms) were also significantly higher than those of healthy volunteers (Reader 1: 20.900 ±1.917ms, Reader 2: 21. 931 ±1.612
ms), P <0.05. There was no statistical difference between SpA and healthy group in T1 values of SIJ cartilage, P >0.05.
The AUCs of the PD and T2* values for discriminating SpA from the controls by the two readers were 0.849- 0.854 and 0.816-0.847,
respectively.The PD and T2*values measurements of the SIJ cartilage showed excellent intra- and inter-observer agreements
(ICC,0.837-0.874),P<0.05.
Conclusion: The MTP sequence provides quantitative PD and T2* values that can be used to adjunctively diagnose SpA.
Limitations: The sample size was relatively small.
Funding for this study: This study was supported by grants from the Clinical Key Program of Peking University Third Hospital (No.
BYSYZD2021018).
Ethics committee - additional information: This study was approved by the Institutional Research Ethics Board and adheres to
the tenets prescribed by the Declaration of Helsinki (institutional review board M2022181).
Monostotic metachronous osteoid osteoma (7 min)

Sara Battisti; Arqua' Petrarca / Italy
674
Author Block: S. Battisti, G. Bisinella; Monselice/IT
Purpose: Osteoid osteoma (OO) is a benign bone forming tumour, which very infrequently has multifocal or multicentric
presentation.
Multiple nidi may be present close to each other in a single bone or in adjacent bones, or may be present in separate bones in
multicentric lesions. We report five unusual cases of double localisation of osteoid osteoma in the same bone after a mean of three
years interval (1-5 years). Further studies may establish a systemic etiology and associated morbidities.
Methods or Background: From 2008, we treated 83 cases of osteoid osteoma with CT-guided radiofrequency thermal ablation. We
used a 17 gauge needle; the RF was activated for 8 minutes using a temperature of 90 degrees C°.
From 2016 to 2023 we retrospectively evaluated all cases with pain relapse after treatment. We found 8 patients (9.6%) with pain
relapse out of a total of 83.
Results or Findings: The analysis of CT images showed that in 5 cases (62%) there was a second nidus close to the first one.
Conclusion: In conclusion, this is the first series of both monostotic and metachronous OO, also associated with pain relapse. Our
case highlights the importance of considering a diagnosis of double nidus.
Limitations: The limitation of our retrospective study is the small sample size. Further studies with a larger sample size are needed
to confirm our results.
profit sectors.
Ethics committee - additional information: This case report was approved by our Institutional review board and patients'
informed consent was obtained.
Hip MRI in flexion abduction external rotation (FABER) position for assessment of the ischiofemoral interval in patients
with hip pain: a feasibility study (7 min)
Florian Schmaranzer; Bern / Switzerland
Author Block: A. Heimann1, T. D. Lerch2, M. Tannast1, S. Steppacher2, M. Wagner3, E. Schmaranzer3, F. Schmaranzer4; 1Fribourg/CH,
2
Bern/CH, 3Sankt Johann in Tirol/AT, 4Zurich/CH
Purpose: The purpose of this study was to assess the feasibility of flexion-abduction-external rotation (FABER) magnetic resonance
imaging (MRI) of the hip to visualise changes in the ischiofemoral interval.
Methods or Background: This is an IRB-approved retrospective single-centre study. Patients underwent non-contrast 1.5 T hip MRI
in the neutral and FABER position. Two readers measured the ischiofemoral interval at three levels: proximal-/distal intertrochanteric
distance and ischiofemoral space. Subgroup analysis was performed for hips with (>30°) /without high femoral torsion (<30°), or
quadratus femoris muscle oedema (QFME), respectively. A receiver operating curve with calculation of the area under the curve (AUC)
for prediction of QFME was calculated.
Results or Findings: 110 patients (121 hips, mean age 34 ± 11 years, 67 females) were evaluated. FABER-MRI led to narrowing
(both P < 0.001) of the ischiofemoral interval, which decreased more at the proximal (mean decrease by 26 ± 7 mm) than at the
distal (6 ± 7 mm) intertrochanteric ridge. With high femoral torsion/QFME, the ischiofemoral interval was significantly narrower at all
three measurement locations compared to normal torsion/no QFME (P <0.05). Accuracy for predicting QFME was high with an AUC of
0.89 (95% CI 0.82 – 0.94) using a threshold of ≤ 7 mm for the proximal intertrochanteric distance.
Conclusion: Hip MRI in the FABER position is feasible, visualises narrowing of the ischiofemoral interval and may be helpful in
diagnosing ischiofemoral impingement.
Limitations: This is a feasibility study; the value of MRI in FABER position needs to be further assessed in a large group of patients
with ischiofemoral impingement.
Funding for this study: This study was partially funded by the Swiss national science foundation.
Ethics committee - additional information: This study was approved by the medical university of Innsbruck.
Are degenerative findings detected on traction MR arthrography of the hip associated with failure of arthroscopic
femoroacetabular impingement surgery? (7 min)
675
Author Block: T. D. Lerch1, A. Heimann2, M. Wagner3, P. Vavron3, S. Steppacher1, M. Tannast2, E. Schmaranzer3, F. Schmaranzer4;
1
Bern/CH, 2Fribourg/CH, 3Sankt Johann in Tirol/AT, 4Zurich/CH
Purpose: The purpose of this study was to identify preoperative degenerative features on traction MR arthrography associated with
failure after arthroscopic femoroacetabular impingement (FAI) surgery.
Methods or Background: This was a retrospective study including 102 patients (107 hips) undergoing traction magnetic resonance
arthrography (MRA) of the hip at 1.5T and subsequent hip arthroscopic FAI surgery performed from January 2016 to February 2020,
with complete follow-up. Clinical outcomes were assessed using the International Hip Outcome Tool (iHOT-12) score. Clinical endpoint
for failure was defined as an iHOT-12 of <60 points or conversion to total hip arthroplasty. MR images were assessed by two
radiologists for presence of 9 degenerative lesions including osseous, chondro-labral/ ligamentum teres lesions. Uni- and multivariate
Cox regression analysis was performed to assess the association between MRI findings and failure of FAI surgery.
Results or Findings: Of the 107 hips, 27 hips (25%) met at least one endpoint at mean 3.7 ± 0.9 years follow-up. Osteophytic
changes of femur or acetabulum (hazard ratio [HR] 2.5 - 5.0), acetabular cysts (HR 3.4) and extensive cartilage (HR 5.1) - and labral
damage (HR 5.5) >2 hours on clockface were univariate risk factors (all P <0.05) for surgery failure. Three risk factors for failure were
identified in the multivariate analysis: Acetabular cartilage damage > 2 hours on the clockface (HR 3.2, P = 0.01), central femoral
osteophyte (HR 3.1, P = 0.02), and femoral cartilage damage with ligamentum teres damage (HR 3.0, P = 0.04).
Conclusion: Evaluation of negative predictors on preoperative traction MR arthrography holds the potential to improve risk
stratification based on the already present joint degeneration ahead of FAI surgery.
Limitations: No comparison was possible to non-contrast MRI of the hip or direct MR arthrography without traction.
Funding for this study: This study was partially funded by the Swiss national science foundation.
Ethics committee - additional information: The institutional review board of the medical university of Innsbruck approved this
study.
Decentration of the femoral head: is it associated with osseous deformities predisposing to hip instability? (7 min)
1 2 3 2 3 1 2 3
Author Block: F. Schmaranzer , A. Heimann , S. Steppacher , M. Tannast , T. D. Lerch ; Zurich/CH, Fribourg/CH, Bern/CH
Purpose: To assess the prevalence of "decentration of the femoral head" on different imaging planes and compare it between hip
deformities.
Methods or Background: IRB-approved retrospective diagnostic study in 351 patients (48% men, mean age 31 ± 12 years) with hip
pain undergoing radiography and direct hip MR arthrography at 1.5T (sagittal-, axial-, and radial- PD-w TSE images). On radiographs,
lateral centre edge angle and neck-shaft angle were measured. On MRI, measurements of femoral torsion were performed at the level
of the lesser trochanter. The presence of decentration was defined as a layer of contrast agent between the posterior femoral and
acetabular cartilage layer. The prevalence of decentration sign was compared between sagittal, axial, and radial images and between
hips with/without hip dysplasia (lateral centre edge angle, LCE <25°), coxa antetorta (>39°) and valgus hips (neck-shaft angle
>139°).
Results or Findings: Decentration was detected with the highest (P <0.001) frequency on radial (28%), followed by axial (13%) and
sagittal (5%) images. Hips with decentration sign had a lower LCE angle (24 ± 9° vs 32 ± 7°; P <0.001), higher neck-shaft angle (134
± 7° vs 131 ± 6°; P <0.001) and higher femoral torsion (24 ± 15° vs 16 ± 12°; P <0.001) than hips without decentration sign. Hips
with decentration had a higher prevalence and were associated with hip dysplasia (56% vs 17%, OR of 6.4; P <0.001), coxa antetorta
(18% vs 5%, OR of 4.2; P <0.001) and valgus deformity (24% vs 8%, OR of 3.8; P <0.001) compared to hips without decentration sign
on MRI.
Conclusion: Decentration sign of the femoral head was most frequently detected on radial MR arthrography and was associated with
osseous deformities predisposing to hip instability.
Limitations: The prevalence of decentration signs is unknown in asymptomatic volunteers.
Funding for this study: This study was partially funded by the Swiss National Science Foundation.
Ethics committee - additional information: Institutional review board approval was granted by the University of Bern.
The overlooked symptom coccydynia: evaluation of sacrococcygeal morphologic and morphometric findings (7 min)
Abdullah Şükün; Antalya / Turkey
676
Author Block: A. Şükün1, H. S. Öztoprak Çubuk2, T. Cankurtaran2, B. Yavuz2, B. Yağdıran2, C. Incekas2; 1Antalya/TR, 2Ankara/TR
Purpose: Coccydynia is one of the most overlooked symptoms in daily clinical practice. Definitions for radiologic evaluation are
controversial.
We aim to compare the morphology and morphometric measurements of the sacrococcygeal region with those of a healthy
population to support radiologic decision making.
Methods or Background: 26 traumatic and 50 idiopathic patients and 74 healthy controls were compared. The morphologic type of
the coccyx, the presence of fusion, and the number of coccygeal segments were evaluated in both groups. Morphometric parameters
such as sacrococcygeal angle (SCA), sacrococcygeal joint angle (SCJA), intercoccygeal angle (ICA), sacral slope (SS), coccyx curved
length (CCL), sacrum curved length (SCL), coccyx length (CL), sacrum length (SL), and sacrococcygeal total length (SCTL) were
investigated.
Results or Findings: Significant differences were found between the Coccydynia group and the healthy control group in
morphologic parameters such as female gender, coccyx segment, coccyx morphology, presence of sacrococcygeal joint, and segment
of sacrococcygeal joint fusion (p <0.05). In morphologic measurements, sacrococcygeal joint angle, sacrum curved length, sacrum
length, coccyx, and sacrum curvature indexes were significantly increased (P <0.05). No significant difference was found in the
morphologic and morphometric parameters evaluated when compared with the duration of coccydynia (P >0.05).
Conclusion: An increase in sacrococcygeal joint angle (SCJA), sacrum curved length (SCL), sacrum length (SL), sacrum curvature
index (SCI), and coccyx curvature index (CCI) measurements predisposes to coccydynia. It would be more accurate to perform a
radiologic evaluation by becoming familiar with these morphologic and morphometric parameters.
Limitations: The limitations of our study are that it is single-centre and retrospective. In our study, CT and MR images of healthy
populations were compared and certain static parameters were evaluated. However, coccydynia is a dynamic pathology and dynamic
radiological radiographs are needed for evaluation.
Ethics committee - additional information: This study was approved by the Institutional Review Board of Baskent University
(Project no: KA23/07).
677
RPS 1603 - Biomarkers in cardiac imaging

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Pericoronary adipose tissue CT attenuation measured from CT calcium score images: a feasible study (7 min)
Didi Wen; Xi'an / China
Author Block: D. Wen, S. Li, M. Zheng, W. Yang; Xi'an/CN

Purpose: Pericoronary adipose tissue (PCAT) CT attenuation assessed by coronary CT angiography (CCTA) image has been identified
as a marker of perivascular adipose tissue inflammation, and was associated with cardiovascular risk. However, the measure of PCAT
CT attenuation in CCTA was potentially confounded by the presence of iodine.
This study aims to develop a method for measuring PCAT CT attenuation on non-contrast CT calcium score (CTCS) images and
validate its feasibility by comparing with those derived from CCTA images.
Methods or Background: Consecutive patients who had completed a CTCS and CCTA images were enrolled in this retrospective
study. Three main arteries were automatically segmented via commercial software (CoronaryDoc®, Shukun Technology Co., LTD) in
CCTA images and were registered to upsampled CTCS images with non-rigid registration methods to obtain PCAT CT attenuation in
CTCS images, while PCAT CT attenuation in CCTA images were accessed via EasyFAI® module. The Spearman correlation coefficient
and Bland-Altman analysis were used to analyse the correlation of PCAT CT attenuation between CTCS and CCTA images.
Results or Findings: PCAT CT attenuation both on CTCS and CCTA images were measured in 191 patients. The mean PCAT CT
attenuation was -79.39±10.56 on CTCS images and -80.37±8.58 on CCTA images, respectively. PCAT CT attenuation on CTCS images
showed a significant correlation with those on CCTA images (r=0.582, P≤0.001) (Figure 1). The mean difference of PCAT CT
attenuation between CTCS and CCTA images was 0.13 HU.
Conclusion: PCAT CT attenuation features can be assessed from three main coronary arteries in CTCS images with higher correlation
and consistency compared with CCTA.
Limitations: The principal limitation of this study is the limited sample size.
Funding for this study: This study received funding from the National Natural Science Foundation of China (grant numbers:
82371953; 82071917; 82202149), and the Key Research and Development Plan of Shaanxi Province (grant numbers: 2023-YBSF-512;
2019SF-182).
Ethics committee - additional information: The study was in accordance with the Declaration of Helsinki. The local ethics
committee approved the study protocol and informed consent was obtain from all patients.
A radiomic model to correlate the quality of pericoronary adipose tissue with significant coronary artery disease (7 min)
Marco De Giorgi; Salerno / Italy
678
Author Block: M. De Giorgi1, S. Dell'Aversana2, A. Ponsiglione1, A. Paludi1, L. Pinto1, A. Annunziata1, E. Cavaglià2, R. Cuocolo1, M.
Imbriaco1; 1Naples/IT, 2Pozzuoli/IT
Purpose: The study aimed to evaluate the correlation between pericoronary adipose tissue (PCAT) quality, assessed with radiomics
features, and significant coronary artery stenoses (≥70%).
Methods or Background: Four operators retrospectively analysed the images of 100 coronary CT exams (50 patients with
significant coronary artery disease [CAD] and 50 healthy controls). The course of the coronary artery on axial CT images was semi-
automatically segmented, followed by automated expansion of the VOI. A threshold was then applied to include exclusively the PCAT.
Feature extraction was done with the open source PyRadiomics package, and an Intraclass Correlation Coefficient Analysis (ICC) was
performed based on segmentations by four different readers, to assess feature stability. To explore the potential clinical value of
these stable radiomics features an exploratory supervised machine learning analysis was performed. The training set patients were
balanced using synthetic oversampling, followed by information gain ranking feature selection. Finally, using the WEKA software
platform, a simple logistic model’s performance was assessed.
Results or Findings: A total of 1,183 radiomics features from each VOI and other clinical values (age, gender hypertension,
diabetes, dyslipidaemia, familiarity) were extracted. Of these, 1,068 were found to be stable at the ICC analysis (ICC lbound ≥0.75).
After merging radiomics and clinical features, those with low variance were 108, while 777 proved to be highly intercorrelated. The
model trained on the top five features based on information gain ranking achieved an AUC of 0.88 in the training data and 0.83 in the
test set to identify significant stenoses.
Conclusion: PCAT radiomics paired with clinical data showed good discrimination between healthy and significantly stenotic coronary
vessels, with a high degree of inter-reader reproducibility. Future studies are needed to create models predictive of prognosis or
treatment response.
Limitations: This was a single-centre, retrospective study.
Ethics committee - additional information: This study was approved by local IRB, with a waiver for informed consent due to the
retrospective, observational nature of the study.
Correlation of intrathoracic fat parameters and extra coronary calcium with the presence and severity of coronary
artery disease and major adverse cardiovascular events (7 min)
Pujitha Vidiyala; New Delhi / India
Author Block: P. Vidiyala, P. Jagia, N. N. Pandey, S. Kumar; New Delhi/IN

Purpose: The current study sought to investigate the association between volume and attenuation of epicardial fat, paracardiac fat,
total intrathoracic fat and extra-coronary calcium with presence of obstructive coronary artery disease (CAD) and major adverse
cardiovascular events (MACE) on CT angiography (CTA), and whether the association, if any, is independent of conventional
cardiovascular risk factors and coronary artery calcium score (CACS).
Methods or Background: The current study is a prospective single centre study. Patients referred for coronary CTA with clinical
suspicion of CAD were included. All patients underwent prospective ECG-gated non-contrast scan (NCCT) for CACS followed by a
retrospective ECG-gated CCTA for evaluating coronary artery stenosis.
Results or Findings: A total of 799 Indian patients (median age: 53.62 years [IQR: 44.08 – 63.16] were included in the study. Details
on MACE were available for 624 out of 799 patients. The mean duration of follow-up was 15.03 ± 4.8 months. On CTA obstructive CAD
(≥50% stenosis) was present in 212 (26.5%) patients. EFV was significantly elevated in patients with obstructive CAD in comparison
to patients without an obstructive CAD (157.01 ± 48.40 versus 133.69 ± 51.01; p = <0.001). On multivariable logistic regression
analysis for predictors of obstructive CAD, presence of hypertension, history of smoking, EFV, TFV, CACS were significant independent
predictors. On multivariable logistic regression analysis for predictors of occurrence of MACE, male gender, EFV, PFV, TFV, CACS, HRPF
and presence of obstructive CAD were significant independent predictors.
Conclusion: EFV emerged as a standalone predictive factor for both the presence of obstructive CAD and MACE in patients
presenting with atypical chest pain, without known history of CAD. Addition of EFV to traditional cardiovascular risk factors and CACS
improves estimation for pretest probability of obstructive CAD and MACE.
Ethics committee - additional information: The institutional ethics board of the All India Institute of Medical Sciences, New Delhi,
approved this study.
Advancing pericoronary inflammation evaluation: dual-energy CT-derived fat fraction as a promising imaging
biomarker for cardiovascular risk stratification (7 min)
Sara Marziali; Milan / Italy
679
Author Block: S. Marziali, C. B. Monti, D. Capra, F. Rizzetto, G. Folco, F. Sardanelli, F. Secchi; Milan/IT
Purpose: The objective of this study is to assess the role of pericoronary fat fraction (FF) obtained via dual-energy CT as an
innovative imaging biomarker to detect coronary inflammation, comparing it to conventional biomarkers such as pericoronary fat
attenuation index (FAI) and the degree of coronary stenosis.
Methods or Background: In our retrospective study, we reviewed patients who had undergone cardiac CT scans at our institution
using a dual-energy calcium scoring scan. We collected both demographic and clinical data for each patient, including factors like
family history of CAD, smoking habits, type 2 diabetes, hypertension, and hypercholesterolaemia. Our main objective was to evaluate
fat fraction (FF), defined as the ratio of adipose tissue within a designated area: Each patient's FF maps were processed using the
Syngo.via software, and regions of interest (ROIs) were delineated around the coronary arteries, specifically the left anterior
descending artery (LAD), circumflex artery (CX), and right coronary artery (RCA). Concurrently, fat tissue attenuation (FAI) values
were recorded. Information regarding coronary stenoses was extracted from CT reports for comprehensive analysis.
Results or Findings: The study encompassed a sample of 99 patients, comprising 32% females with a median age of 66 years (IQR:
58-74 years). A significant negative correlation emerged between FF and FAI across coronary arteries: LAD at ρ=-0.617 (p<0.001),
LCX at ρ=-0.493 (p<0.001), and RCA at ρ=−0.506 (p<0.001). Pericoronary FF displayed a weak negative correlation with coronary
stenosis at the LAD (ρ=-0.220, p=0.035), correlations were not statistically significant for the LCX and the RCA (p=0.572).
Conclusion: Fat fraction, evaluated using dual-energy CT, presents promise as an additional imaging biomarker for coronary
inflammation. It could yield potential for cardiovascular risk stratification, underscoring the need for further, comprehensive research.
Limitations: This was single centre, retrospective study.
Ethics committee - additional information: The local ethics committee (Ethics Committee of IRCCS Ospedale San Raffaele)
approved this retrospective study (protocol code “CardioRetro”, number 122/int/2017; approved on 14th September 2017, and
amended on 19th July 2022). Informed consent was waived due to the retrospective nature of the study.
Prognostic value of changes in epicardial adipose tissue volume and density on lung cancer screening CT (7 min)
Isabel Luisa Langenbach; Erzhausen / Germany
Author Block: I. L. Langenbach1, I. Hadzic1, R. Zeleznik1, M. C. Langenbach1, D. Maintz2, T. Mayrhofer1, M. T. Lu1, H. Aerts1, B.
Foldyna1; 1Boston/US, 2Münster/DE
Purpose: To investigate how longitudinal changes in epicardial adipose tissue (EAT) volume and density relate to all-cause,
cardiovascular (CV), and lung cancer mortality in participants undergoing low-dose computed tomography (LDCT) for lung cancer
screening.
Methods or Background: EAT volume and density, known as potential risk factors for CV disease, were measured using a validated
automated deep-learning algorithm on serial, non-ECG-synchronised, non-contrast chest LDCT in heavy smokers enrolled in the
National Lung Screening Trial (NLST). EAT volumes were indexed to body surface area (cm3/m2). EAT volume and density changes
over two years (EAT; t2-years–tbaseline) were categorised into increase, no-change, and decrease categories (no-change: volume: -7
to +11%; density: -3 to +2%). EAT was associated with all-cause, CV, and lung cancer mortality (10-year follow-up) in multivariable
Cox regression models, adjusted for baseline EAT values, traditional CV risk factors, BMI, history of CV disease, and coronary artery
calcium (CAC) score.
Results or Findings: Of 20,661 people (59% men; age: 61±5 years), 3,483/20,661 (16.9%) died over a median follow-up of 10.4
(9.9–10.8) years (CV death: 816/3,483 [23.4%]; lung cancer death: 705/3,483 [20.2%]). In general, EAT volume increased, while
density slightly decreased over time (+2.5±11.0 cm3/m2 and -0.5±3.0 HU). In fully adjusted analysis, EAT volume decrease was
related to all-cause and CV mortality (HR=1.34; 95%CI: 1.23–1.46; p<0.001, HR=1.27; 95%CI: 1.06–1.51; p=0.009, respectively),
while EAT volume increase was associated with higher hazard of all-cause mortality (HR=1.15; 95%CI: 1.06–1.25; p<0.001). EAT
density increase was associated with a higher risk for all-cause (HR=1.32; 95%CI: 1.22–1.44; p<0.001), CV (HR=1.30; 95%CI:
1.09–1.55; p=0.004), and lung cancer mortality (HR=1.34; 95%CI: 1.11–1.62; p=0.002).
Conclusion: Patients eligible for lung cancer screening with changes in EAT volume and density have higher mortality risks and a
need for better risk stratification.
Limitations: Identified limitations were (1) that this was a retrospective study, and (2) that a high-risk population of heavy smokers
was used.
Prognostic value of coronary CTA-based Coronary Artery Disease Reporting and Data System, pericoronary adipose
tissue attenuation and coronary plaque burden for major adverse cardiovascular events (7 min)
Xin Yu Du; Wuhan / China
680
Author Block: X. Y. Du, Z. Huang, R. Tang, Y. Ding, X. Wang, W. Wang, Z. Li, X. Wang; WuHan/CN
Purpose: This study aimed to evaluate the prognostic value of Coronary Artery Disease Reporting and Data System (CAD-RADS),
pericoronary adipose tissue (PCAT), and coronary plaque burden for major adverse cardiovascular events (MACE) in coronary
computed tomography angiography (CTA).
Methods or Background: This retrospective study included 3,783 patients who underwent coronary CTA from 2018 to 2019. MACE
included myocardial infarction, stroke, or all-cause death. Coronary CTA data sets were analysed by artificial intelligence software to
quantify plaque burden (including calcified plaque [CP] volume, non-calcified plaque [NCP] volume, and mixed plaque [MP] volume),
PCAT attenuation, and PCAT volume. Plaque burden, PCAT CT attenuation, PCAT volume, CAD-RADS and cardiac risk score were used
to evaluate risk factors of MACE by multivariable Cox regression analysis.
Results or Findings: In 3,783 evaluable participants (60.37±10.55 years) during a median follow-up of 39 months, 243 of the 3,783
patients (6.42%) experienced MACE. From Youden index analysis, the prognostic threshold of PCAT CT attenuation was >−76.00 HU
for right coronary artery (RCA), >−82.00 HU for left anterior descending (LAD) and >−75.00 HU for left circumflex (LCX), respectively.
The prognostic threshold of plaque burden was >13 ml for CP, >11 ml for NCP and >25 ml for MP, respectively. Kaplan–Meier survival
curves showed that MACE have significant association with PCAT attenuation of the LCX, CAD-RADS, and plaque burden (all P<0.001).
Multivariable Cox regression analysis showed that plaque burden NCP>11 (hazard ratio, 1.496; P<0.005) and MP>25 (hazard ratio,
1.932; P<0.005) were independent predictors of MACE.
Conclusion: Non-calcified plaque volume and mixed plaque volume were independent predictors of MACE.
Limitations: This study lacked data on specific causes of death.
Foetal epicardial fat thickness: can it serve as a marker for gestational diabetic mellitus? (7 min)
Amandeep Singh; Amritsar / India
Author Block: A. Singh; Amritsar/IN

Purpose: To evaluate the role of foetal epicardial fat thickness as a marker to screen for GDM.
Methods or Background: The study included pregnant patients at 24 + 0/6 to 28 + 0/6 weeks of gestation scheduled for a 75 g oral
glucose tolerance test from December 2020 to March 2022. Antenatal ultrasound was performed on a Voluson E8 Expert BT12 (Wipro
GE) ultrasound machine. Out of 180 patients, 60 patients were selected, that is, 30 patients with raised 75 g OGTT results (cases of
GDM) and 30 patients with normal 75 g OGTT results. The collected data were transformed into variables, coded, and entered into
Microsoft Excel. Data were analysed using the Shapiro–Wilk normality test, student's t-test or Mann–Whitney U test, chi-square test, or
Fisher's exact test and statistically evaluated using the SPSS-PC-25 version.
Results or Findings: Foetal EFT was found to be significantly more in the GDM group in comparison to controls without GDM, and
the increased foetal EFT was positively associated with 2-hour OGTT serum glucose values. The mean foetal epicardial fat thickness
(EFT) in mothers with GDM was significantly larger, i.e., 0.17 ± 0.02 cm than in mothers without GDM, i.e., 0.12 ± 0.01 cm (p < 0.001).
The receiver operating characteristic (ROC) curve plotted from values calculated from our results shows high sensitivity (i.e., 96.67%)
and specificity (i.e., 90%) of foetal EFT as a predictor for GDM with an AUROC value of 0.96 and 95% confidence interval of 0.92 to
1.0.
Conclusion: EFT was significantly higher in foetuses of diabetic versus nondiabetic mothers, therefore concluding that epicardial fat
thickness measurement in foetuses can serve as a novel marker in GDM.
Limitations: Study can be extended to a larger sample population.
Ethics committee - additional information: Due clearance was obtained from the institutional research and ethics committee.
The value of epicardial adipose tissue characteristics of low-dose CT in predicting coronary artery stenosis, high-risk
plaque and CT-FFR (7 min)
Xueyan Ma; Zhengzhou / China
681
Author Block: X. Ma; Zhengzhou/CN
Purpose: We aimed to assess relationship between epicardial adipose tissue (EAT) radiomics from low-dose computed tomography
(LDCT) with coronary artery stenosis, high risk plaques and CT-FFR.
Methods or Background: The retrospective study included all 310 patients (168 men, 142 women, mean age 60±8 years) who
underwent both LDCT and coronary computed tomography angiography in our institution. We recorded the clinical variables including
patients’ demographics, smoking history, family history, and lipid profiles. EAT volume and density, grading of coronary stenosis, high
risk plaques and CT-FFR≤0.8 were recorded from the radiology information system. Deidentified LDCT exams were exported to a
Radiomics prototype for automatic EAT segmentation, and derivation of radiomics. Data were analyzed using multiple logistic
regression and kernel Fisher discriminant analyses.
Results or Findings: Whole EAT radiomics were better than the EAT volume and density and clinical variables for differentiating
subjects with predicting coronary stenosis (area under the curve [AUC] 0.83 vs 0.70; 0.65; 0.57). Prediction of high risk plaque and
CT-FFR≤0.8 was better on whole EAT radiomics (AUC:0.80-0.87) than with EAT volume and density and clinical variables
(AUC:0.57-0.69).
Conclusion: Whole EAT radiomics obtained from LDCT can differentiate patients with different coronary artery stenosis, high risk
plaques and myocardial ischemia for cardiovascular diseases.
Limitations: Not enough data included
Funding for this study: There is no funding source for this study
Ethics committee - additional information: The project has passed the ethics committee.
Predictive performance of machine learning-derived radiomics signature of perivascular fat attenuation index for
interventional antegrade percutaneous coronary intervention for chronic total occlusion (7 min)
Xueyan Ma; Zhengzhou / China
Author Block: X. Ma; Zhengzhou/CN

Purpose: Coronary inflammation can alter the perivascular fat phenotype. Hence, this study aimed to investigate the association
between the radiomics features of perivascular fat attenuation index (FAI) and the success of antegrade percutaneous coronary
intervention (PCI) in the treatment of chronic total occlusion (CTO).
Methods or Background: This study evaluated patients with only one CTO lesion observed on conventional coronary angiography
(CAG) who underwent coronary computed tomography angiography (CCTA) < 1 month before CAG, from 2020 to 2023. The clinical
data, CCTA-based CTO lesion morphologic characteristics, and perivascular FAI of CTO lesions were recorded and analysed. The
radiomics features were extracted from the FAI around CTO lesions, and the selected features were used to establish a radiomics
model using a machine learning algorithm, and the selected image features were combined with Radcore to form an integrated
model.
Results or Findings: In total, 173 patients with CTOs were enrolled in this study. Successful antegrade PCI (A-PCI) was achieved in
126 CTO lesions (72.8%). Ten significant radiomics features were selected. In the training group, the multivariable analysis revealed,
FAI≤−76.82 HU, severe calcification degree and occlusion length ≥ 18.73 mm, were independent predictors of A-PCI failure. The AUC
for the radiomics model and the integrated model were 0.80 and 0.75 in the training group, and 0.72 and 0.68 in the test group.
Conclusion: A CCTA-based radiomics signature of FAI may be helpful for choosing appropriate interventional strategies.
Limitations: Our study did not have external validation.
Ethics committee - additional information: Our research was approved by an ethics committee.
Is familial combined hypolipidaemia (FHBL2) due to loss of function mutation in the angiolipoprotein-like3 (ANGPTL3)
gene protective for CAD? (7 min)
Serena Paciulli; Rome / Italy
682
Author Block: S. Paciulli1, G. C. Pambianchi1, L. Marchitelli1, M. Giannetti1, M. Francone2, C. Catalano1, N. Galea1; 1Rome/IT, 2Milan/IT
Purpose: The purpose of this study was to compare the prevalence and severity of coronary atherosclerotic disease (CAD) and the
density and volume of epicardial fat (EpF) between a group of patients with FHBL2 and an age- and gender-matched control group
and to analyse the correlation between EpF and stenosis degree.
Methods or Background: A group of 111 FHBL2 subjects were prospectively recruited and underwent coronary CT (CCTA). 46
normolipidemic asymptomatic age- and gender-matched subjects, who underwent CCTA according to current guidelines, were
enrolled as a control group. CAD-RADS were used to determine stenosis severity and to divide patients into four classes (N= 0-1; L=
2; M= 3; S= 4-5). The density and volume of epicardial fat (EpF) were evaluated on CCTA images.
Results or Findings: According to CCTA images, FHLBL2 patients were classified as follows: 74 subjects were classed N (67%), 22 L
(20%), 7 M (6%) and 8 S (7%); among the controls 17 subjects were classed N (37%), 16 L (35%), 8 M (17%) and 5 S (11%). In FHLBL2
subjects there was a higher prevalence of minimal or no stenosis (N: 67% vs 37%; P =0.001), and a lower prevalence of mild stenosis
(L: 20% vs 35%; p <0.05), while there were no significant differences in the development of severe coronary atheromasia (S: 7% vs
11%; P =0.553). There were no significant differences in terms of EpF density or volume between the two groups (P >0.060 for both)
but a correlation was observed between EpF density and volume and the degree of stenosis (Rho: - 0.563; P =0.023 and Rho: 0.565; P
=0.019).
Conclusion: FHBL2 is associated with a lower prevalence and severity of CAD. An association between EpF density and volume and
the severity of stenosis was also found.
Assessing radiologic biomarkers for heart-related complications in NSCLC treatment (7 min)

Marco De Chiara; Lusciano / Italy
Author Block: V. Nardone, A. Reginelli, V. Patanè, M. De Chiara, G. Balestrucci, R. Grassi, M. P. Belfiore, S. Cappabianca; Naples/IT
Purpose: To assess the significance of radiologic biomarkers, including the Agatston score and ejection fraction, in detecting and
predicting heart-related complications arising from non-small cell lung cancer (NSCLC) treatment.
Methods or Background: This retrospective analysis centred on 173 non-small cell lung cancer(NSCLC) patients who received
treatment from 07.2018 to 07.2022. We assessed the Coronary Artery Disease (CAD) score by analyzing baseline CT scans acquired
at the time of diagnosis. To assess overall survival (OS), we calculated the duration from each patient's selected CT scan to the date
of their demise or their last follow-up appointment. To provide a comprehensive examination, we conducted a multivariate analysis
using the Cox regression method.
Results or Findings: Among our 173 patients, 120 of them (69.6%) had succumbed at the last follow-up. The median overall
survival (OS) was 28 months, with a mean of 47.2 months. In the univariate analysis, several parameters exhibited significant
correlations with OS. CAD grading (p < 0.001), disease stage (p < 0.001), a history of ischemic heart disease (p: 0.034), the use of
beta blocker drugs (p: 0.036), and cardiac ejection fraction (p: 0.005) were all significant factors. In the multivariate analysis, only a
select few parameters retained their significance. These included disease stage (p: 0.016), CAD score (p: 0.014), and cardiac ejection
fraction (p: 0.011).
Conclusion: This study underscores the significance of disease stage, CAD score, and cardiac ejection fraction in determining NCLC
patient’s OS. These findings emphasize the need for comprehensive risk assessments in all NSCLC patients before and during cancer
therapy, facilitating risk stratification for cardiovascular disease.
Limitations: This study is constrained by its retrospective design. Nevertheless, it capitalizes on a uniform cohort of patients at
different stages, employing standardized thorax CT imaging for CAD score assessments.
Funding for this study: No
Ethics committee - additional information: Prot. 2022XK8DA5
Combined use of MRI and PET imaging enhances diagnostic accuracy in cardiac sarcoidosis (7 min)
Knut Haakon Stensaeth; Trondheim / Norway
683
Author Block: K. H. Stensaeth, H. Johansen, A. C. Dale; Trondheim/NO
Purpose: Diagnosing cardiac sarcoidosis (CS) often presents challenges. This study sought to assess the efficacy of integrating
cardiac MR (CMR) with positron emission tomography (PET)-MR for diagnosing CS and guiding patient management.
Methods or Background: We conducted a retrospective review of 46 patients at St Olav's Hospital, Trondheim, Norway, between
February 2017 and October 2021. Patients underwent CS evaluation using both CMR and PET-MR. Post-CMR, an experienced
radiologist identified all patients as possibly having CS. A seasoned nuclear medicine physician then classified PET results as normal,
unspecific, or positive. The final diagnosis combined imaging and clinical findings.
Results or Findings: Participants averaged 58±11 years of age; 24% were female. 28% had a previous biopsy confirming
sarcoidosis. Predominant symptoms included dyspnea (26%), angina (15%), and syncope (13%), with 26% reporting arrhythmia
history. Late gadolinium enhancement (LGE) prominently appeared in basal segments 2-6. Abnormal F18-FDG uptake occurred in 12
(26%) patients, with 10 patients deemed positive and two unspecific. During the inclusion period, mortality stood at 9%, exclusively
among patients with positive FDG uptake. 43% exhibited compromised left ventricle (LV) function, 39% had hypertrophic
cardiomyopathy, and 17% had a dilated LV. Alternative diagnoses comprised amyloidosis, giant cell myocarditis, systemic lupus
erythematosus, Fabry's disease, arrhythmic right ventricle cardiomyopathy, and endocarditis. Additionally, 26% received an ICD or
pacemaker.
Conclusion: In patients with potential CS, the combined use of CMR and PET imaging significantly aids diagnosis and management
decisions. Integrating these modalities with clinical data can notably enhance diagnostic accuracy, especially when CMR results are
ambiguous.
Limitations: This was a retrospective single centre study.
Ethics committee - additional information: This study was approved by an ethics committee; REK 6701 South-East C Norway.
684
RPS 1602 - Functional breast imaging modalities
Categories: Artificial Intelligence & Machine Learning, Breast, Contrast Media, Imaging Methods
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
DCE MRI radiomic analysis in triple negative ductal invasive breast cancer: comparison between BRCA and non-BRCA
mutated patients (7 min)
Chiara Bozzola; Modena / Italy
Author Block: C. Bozzola, L. Crasti, E. Balboni, C. Beretta, L. Nocetti, G. Besutti, A. Toss, G. Ligabue, A. Pecchi; Modena/IT
Purpose: This study aimed to determine whether and which radiomic features from breast dynamic contrast-enhanced (DCE) MRI
predict the presence of BRCA1/BRCA2 mutations in patients with triple-negative ductal invasive (TNDI) breast cancer.
Methods or Background: This retrospective study included consecutive patients histologically diagnosed with TNDI breast cancer
who underwent breast DCE-MRI in 2010-2021. Baseline DCE-MRIs were retrospectively reviewed; two percentage maps of wash-in and
wash-out were computed and breast lesions were manually segmented, drawing a 5 mm-Region of Interest (ROI) inside the tumour
and another 5 mm-ROI inside the contralateral normal glandular tissue. Features for each map and each ROI were extracted with the
Pyradiomics extension of 3D Slicer and considered first separately (tumour and contralateral gland) and then together. In each
analysis, eight more important features for BRCA1/2 status classification were selected with Maximum Relevance Minimum
Redundancy algorithm and used to fit four different classifiers.
Results or Findings: The population included 69 patients for a total of 88 lesions (21 in BRCA1-mutated, 2 in BRCA2-mutated, and
65 in non-BRCA-carriers). The best classifier for BRCA mutation was Logistic Regression fitted with both tumour and contralateral
gland features, reaching an AUC of 0.78, a sensitivity of 0.9 (95%CI: 0.86-0.94) and a specificity of 0.78 (95%CI: 0.74-0.82). Two
features were higher in BRCA-mutated compared to non-BRCA-mutated (correlation from grey level co-occurrence matrix and root
mean square), both measured in the contralateral gland in wash-out maps. They represented the coarseness of the pattern inside the
ROI and the variability of voxel intensities, respectively.
Conclusion: This study shows the feasibility of a radiomic study with DCE maps of the breast and the potential of radiomics in
predicting BRCA mutational status.
Limitations: The limitations of the study were that it was a retrospective design and a limited sample size.
Funding for this study: This research has received funding from the European Union-NextGenerationEUthrough the Italian Ministry
of University and Research under PNRR -M4C2-I1.3 Project PE_00000019 "HEAL ITALIA" to Giulia Besutti, E93C22001860006.
Ethics committee - additional information: This study was approved by the AVEN Ethical Committee.
Feasibility and added value of contrast-enhanced tomosynthesis (7 min)

685
Author Block: P. Clauser1, N. Pötsch1, P. Kapetas1, M. Hörnig2, M. Weber1, R-I. Milos1, P. A. Baltzer1, T. H. Helbich1; 1Vienna/AT,
2
Forchheim/DE
Purpose: This study aimed to compare the diagnostic performance of a CE-DBT prototype with CEM. Contrast-enhanced
mammography (CEM) is increasingly used in clinical practice, but its value is limited by the 2D nature of the examination. Contrast-
enhanced tomosynthesis (CE-DBT) allows a quasi-3D evaluation of contrast-enhanced images and could improve lesion
characterisation.
Methods or Background: This prospective study was approved by the ethics committee and all patients gave written informed
consent. Women presenting with suspicious findings on mammography, DBT or ultrasound were invited to participate in the study.
Participants underwent CE-DBT of the breast with suspicious findings using a dedicated prototype in addition to CEM. The suspicious
findings were biopsied and only histologically verified lesions were included in the analysis. Four readers (R1 and R2 non-experienced;
R3 and R4 experienced) evaluated the images, blinded to patients’ history, previous imaging, and histology. The readers evaluated
CEM (including mammography and recombined images) and CE-DBT (including DBT and synthetic mammography) in separate
sessions and gave a BI-RADS score for each finding. Sensitivity and specificity were calculated and compared.
Results or Findings: We included 84 patients (mean age 56 years, range 39-70) with 91 histologically verified breast lesions (27
benign, 64 malignant). Sensitivity was comparable between CEM and CE-DBT for non-experienced readers, ranging from 90% to 95%.
Sensitivity improved using CE-DBT for both experience readers (from 90% and 87% with CEM to 100% and 94% with CE-DBT).
Specificity was lower for non-experienced than for experienced readers with both imaging modalities, with no significant differences
between modalities.
Conclusion: Our study showed that CE-DBT allowed an improvement in sensitivity, particularly for experienced readers, with no
significant decrease in specificity.
Limitations: This study was a single-centre and cancer-enriched dataset.
Funding for this study: Funding was received from Siemens Healthcare GmBH.
Ethics committee - additional information: This prospective study was approved by the local ethics committee.
Comparison of T1 mapping and fixed T1 method for dynamic contrast-enhanced MRI perfusion in the differential
diagnosis of benign and malignant breast lesions (7 min)
Chen Xiao Yu; Chengdu / China
Author Block: C. X. Yu; Chengdu/CN

Purpose: The objective of this study was to compare dynamic contrast-enhanced MRI (DCE-MRI) data obtained using different pre-
bolus T1 values in the differential diagnosis of benign and malignant breast lesions.
Methods or Background: We retrospectively reviewed 85 cases of breast cancer: 51 benign breast lesions and 34 malignant breast
lesions. DCE-MRI maps of plasma volume fraction (Vp), extravascular-extracellular volume fraction (Ve), and tracer transfer constant
from plasma to tissue (Ktrans) were obtained using a fixed T1 value of 2000ms and a measured T1 obtained with variable flip angle
(VFA). Tumour segmentations were performed and first-order histogram parameters were extracted from volumes of interest (VOIs)
after co-registration with the perfusion maps. The two methods were compared using the Wilcoxon matched-pairs signed-rank test
and Bland-Altman analysis. Diagnostic accuracy was obtained and compared using ROC curve analysis and DeLong’s test.
Results or Findings: Perfusion parameters obtained with the fixed T1 value were significantly higher than those obtained with the
VFA. As regards diagnostic accuracy, there were no significant differences between the two methods for the differential diagnosis of
benign and malignant breast lesions, except for a few parameters of both methods.
Conclusion: DCE-MRI data obtained with different pre-bolus T1 are not comparable and the definition of a pre-bolus T1 by T1
mapping is not mandatory since it does not improve the diagnostic accuracy of DCE-MRI.
Quantitative and qualitative evaluation of reduced dose of high-concentration of contrast media in CESM (contrast
enhancement spectral mammography) (7 min)
Dominga Pugliese; Rome / Italy
686
Author Block: D. Pugliese, C. Bernardi, D. Caruso, M. Zerunian, G. Maccagno, R. Gallo, A. Laghi, A. Speranza; Rome/IT
Purpose: This study aimed to investigate whether a reduced dose of iodinated contrast agent, at a higher concentration, maintains
adequate image quality compared with a standard dose. Contrast-enhanced spectral mammography (CESM) produces dual-energy
subtracted images that demonstrate iodine uptake in infiltrating breast tumours.
Methods or Background: One hundred patients with BI-RADS 6 who underwent CESM study were enrolled: 50 women had received
a dose of 1.5 ml/kg of Iomeprole 350mg/ml iodinated contrast agent and 50 women a dose of 1.0 ml/kg Iomeprole 400 mg/ml.
Quantitative evaluation of the images was obtained by measuring two ROIs (regions of interest): the first placed on the contrast-
enhancing lesion and the second on the background of the image. The percentage difference in signal between the enhancing lesion
and the background (%RS) and the signal-to-noise ratio (SNR) was calculated. The qualitative assessment was independently
analysed by two experienced radiologists using EUREF criteria.
Results or Findings: The quantitative evaluation showed a slightly higher average lesion-enhancing value in the group with
Iomeprole 350 mg/ml vs 400 mg/mI (mean ROI: 2104.95±37.95 vs 2072.89±35.35, P=0.12). The difference between %RS and SNR
calculated by the Wilcoxon test showed no statistically significant difference between the two groups (respectively, p-value >0.9 and
p-value >0.4). Qualitative assessment scored equally for lesions’ enhancement in both the first and second groups.
Conclusion: Our preliminary results showed that it is possible to obtain comparable image quality in CESM studies by using a
reduced dose of high-concentration iodinated contrast agent, compared to the standard dose.
Limitations: This study was a single-centre study with a small sample size.
Ethics committee - additional information: Written informed consent was acquired for all patients and approved by the
Institutional Review Board.
Machine learning for triple negative breast cancer prediction using 18F-FDG PET/CT imaging (7 min)
Yunxuan Li; Xi‘an / China
Author Block: Y. Li, Y. Liao; Xi‘an/CN

Purpose: This study aimed to discuss leveraged machine learning techniques based on 18F-FDG Positron Emission
Tomography/Computed Tomography(PET/CT) radiomics to discriminate TNBC from other molecular subtypes of breast cancer(BC).
Methods or Background: A total of 131 female BC patients who underwent 18F-FDG PET/CT imaging, between June 2016 and
December 2022, were enrolled in this study, including 22 TNBC cases. Breast tumours were delineated and 1316 PET features
radiomic features were extracted. Quantitative parameters including max and the mean value of standard uptake values (SUV),
metabolic tumour volume, and total lesion glycolysis were measured and compared with the rank sum test. The dataset was
randomly divided into training and validation sets with a ratio of 7:3. Feature selection included an initial screening using the
Maximum Redundancy Minimum Correlation (MRMR) algorithm, retaining the top 30 highly correlated features. A secondary screening
was performed using the Least Absolute Shrinkage and Selection Operator (LASSO) regression. A logistic regression algorithm was
employed for model training. The performance of the constructed model was evaluated using receiver operating characteristic (ROC)
analysis, including area under the curve (AUC), sensitivity (SEN), and specificity (SPE).
Results or Findings: The age range of included patients was 26 to 84 years (mean age ± standard deviation: 51 ± 13 years). There
were no statistically significant differences in quantitative parameters between triple-negative and other phenotypes (all P>0.05). The
final model incorporated six radiomic features. The model achieved promising results in the training cohort (AUC=0.868, SEN=0.8,
SPE=0.921) and was validated in the test cohort (AUC=0.81, SEN=1, SPE=0.697).
Conclusion: This machine learning model, leveraging PET radiomics, demonstrates strong potential for predicting TNBC. Its clinical
application may aid healthcare professionals in making informed decisions regarding preemptive systemic therapy for TNBC patients,
potentially enhancing patient outcomes.
Funding for this study: Funding was received from the Natural Science Basic Research Program of Shaanxi Province,
China(2020JZ-38).
Ethics committee - additional information: This study was approved by the Ethics Committee of the First Affiliated Hospital of Xi
'an Jiaotong University (2019LSYZD-J1-H).
Breast lesion analysis in contrast-enhanced photon-counting CT (PC-CT) reconstructions with special breast
positioning: a prospective study on iodine quantification and lesion conspicuity (7 min)
Caroline Wilpert; Freiburg im Breisgau / Germany
687
Author Block: C. Wilpert, O. Gebler, T. Stein, F. Bamberg, M. Windfuhr-Blum, C. Neubauer, J. Neubauer; Freiburg im Breisgau/DE
Purpose: The objective of this study was to describe the enhancement, iodine values and contrast of breast cancers, DCIS and
benign breast lesions in contrast-enhanced (CE) PC-CT.
Photon-counting CT (PC-CT) has the advantage of high spatial resolution, reduced noise and improved iodine contrast compared to
conventional CT. The prone breast positioning is promising for intermodal comparisons.
Methods or Background: A prospective study included 70 women (55 years ±14SD) with biopsy-confirmed breast cancers and
breast MRI for lesion comparisons (BI-RADS). A PC-CT (NAEOTOMAlpha, Siemens, Erlangen) of the thorax/abdomen was performed in
a prone position with special breast bearing after bolus injection of iodinated contrast-medium (fixed delay: 85 seconds) with
multiplanar reconstructions of both breasts (FOV: 34 cm, matrix: 1024 x 1024, slice-thickness: 2 mm) including iodine maps and 65
keV monoenergetic images. A ROI-based analysis of HU-values and spectral information was performed referenced to the ascending
aorta for quantification of iodine amount.
Results or Findings: Analysis of 70 biopsy-proven breast cancers, 26 DCIS, six fibroadenoma, eight intramammary lymph nodes and
eight cystic lesions. Eight additional cystic lesions could not be detected at all. Mean enhancement, mean iodine concentrations and
% of enhancement to the aorta reference were highest in breast cancers (112 HU, 2.7 mg/ml, 52%) and differed significantly from all
other lesion types including DCIS (91 HU,1.7 mg/ml, 33%), fibroadenoma (61 HU, 0.3 mg/ml, 5%), intramammary lymph nodes (79
HU,1.8 mg/ml, 38%) and cystic lesions (33 HU, 0 mg/ml, -1%); each p<0.001. The contrast was highest in breast cancers. Cystic
lesions presented with low contrast.
Conclusion: Compared to all other types of lesions, breast cancers displayed the highest enhancement and iodine concentrations;
these characteristics indicate that breast CT reconstruction might be suited for pre-therapeutic local breast cancer staging.
Additionally, DCIS can be discriminated against with CE PC-CT. In contrast, cystic lesions cannot be displayed efficiently.
Limitations: Only a small number of biopsy-proven benign lesions were analysed. Morphologic features were not compared to MRI.
PC-CT is not sufficient for the presentation of micro-calcifications.
Funding for this study: This study was partially funded by the Young Researchers Grant awarded by the EUSOBI.
Ethics committee - additional information: This study was approved by the Ethics Number: 21-1717, German clinical trials
register: DRKS00028997
Discrimination of axillary lymph nodes and lymph node metastasis in contrast-enhanced photon-counting CT (PC-CT)
breast reconstructions (7 min)
Caroline Wilpert; Freiburg im Breisgau / Germany
Author Block: C. Wilpert, M. Molina, T. Stein, F. Bamberg, M. Windfuhr-Blum, C. Neubauer, J. Neubauer; Freiburg im Breisgau/DE
Purpose: The objective of this study was to investigate whether lymph node metastases can be distinguished from benign lymph
nodes and breast cancers by their enhancement in contrast-enhanced (CE) PC-CT. Imaging of axillary lymph nodes continues to be a
challenge, where up to this date sonography remains the imaging reference. Photon-counting computed tomography (PC-CT) allows
for high spatial resolution and iodine quantification.
Methods or Background: Inclusion of 75 women (56 years ±14 SD) with biopsy-confirmed breast cancers and MRI available to
confirm diagnostic criteria of benign lymph nodes (BI-RADS). A PC-CT (thorax/abdomen) was performed with special prone breast
positioning (NAEOTOMAlpha, Siemens, Erlangen) after bolus injection of iodinated contrast medium (delay: 85 seconds) and
multiplanar reconstructions of both breasts (FOV: 34 cm, matrix: 1024 x 1024, slice-thickness: 2 mm) including iodine maps and 65
keV monoenergetic images. ROI-based analysis was performed using the iodine concentration of the aorta ascendens as a reference.
Results or Findings: Analysis of 70 biopsy-proven breast cancers, 18 lymph node metastasis and 57 benign lymph nodes.
Enhancement was 121 HU in metastasis, 112 HU in breast cancers and 105 HU in benign lymph nodes. Iodine concentrations and
referenced enhancement (%) was highest in breast cancers (2.7 mg/ml, 52%). Lymphnode metastasis differed significantly from
benign lymph nodes (2.5 mg/ml, 49% versus 1.9 mg/ml, 37%; enhancement: p=0.004; iodine: p=0.008) but not from breast cancers
(enhancement: p=0.251; iodine: p=0.153). In contrast, benign lymph nodes differed significantly from breast cancers (p <0.001).
Conclusion: Enhancement of lymph node metastasis was similar to breast cancers and differing from benign lymph nodes, which
might allow further differentiation. However, as ranges were rather wide we do not believe cut-off values can be drawn from the data
for lymph node evaluation with PC-CT; instead, as known from MRI, ultrasound with it’s ability to depict morphology and perfusion is
expected to outperform PC-CT providing less morphologic information.
Limitations: Morphologic features of lymph nodes were not evaluated. PC-CT was not compared to other imaging modalities.
Funding for this study: This research was partially funded by the Young Researchers Grant awarded by the European Society of
Breast Imaging (EUSOBI).
Ethics committee - additional information: This study was approved by the Ethics Number: 21-1717, German clinical trials
register: DRKS00028997
The characteristics of quantitative DCE-MRI in breast cancer according to the HER-2 expression (7 min)
Yangling Hu; Guangzhou / China
688
Author Block: Y. Hu, Z. Xiaoling; Guangzhou/CN
Purpose: This study aimed to quantitatively assess the differences in parameters of dynamic contrast-enhanced MRI (DCE-MRI) in
HER2-zero, HER2-low, or HER2-positive tumours, and to build optimal models for early prediction of HER2-low breast cancer (BC).
Methods or Background: Clinical and DCE-MRI data from 220 BC patients receiving neoadjuvant chemotherapy (NACT) were
retrospectively analysed. Quantitative and semi-quantitative DCE-MRI parameters were compared in the HER2-zero, HER2-low, or
HER2-positive groups before and after early NACT. Empirical models were developed to predict HER2-low BC using logistic regression
analysis and receiver operating characteristic (ROC) analysis.
Results or Findings: Patients of HER2: low BC have a lower pCR rate compared with HER2-zero and HER2-positive (17.9% vs. 10.4%
vs. 29.5%, p<0.001), predominantly in the HR (hormone receptor) negative group (22.2% vs. 7.7% vs. 40.5%, p<0.001). Before NACT,
HER2-low BC exhibited higher Kep, Ktrans, Washin, and lower TME intratumoral perfusion characteristics, and higher Kep and lower
TME in peritumoral than the other group. Notably, after early NACT, changes in Ktrans, Kep, and Washin in intratumoral and
peritumoral perfusion were more pronounced in the HER2-low group compared to other groups. The ROC curves (AUC) for the pre-
NACT intratumoral, peritumoral, and combined perfusion models were 0.675 (95% CI: 0.600-0.750), 0.661 (95% CI: 0.585-0.738),
0.731 (95% CI: 0.660-0.802). The combined pre-and-post-NACT perfusion model further improved predictive performance accordingly,
with AUCs of 0.764 (95% CI: 0.637-0.865), 0.795 (95% CI: 0.711-0.878), 0.850 (95% CI: 0.774-0.926).
Conclusion: The study revealed perfusion heterogeneity between different HER2 statuses and identified the best imaging model as a
non-invasive tool to predict HER2-low BC, which can help pre-treatment clinical decision-making.
Limitations: Further multi-centre validation is required.
Funding for this study: This study has received funding from the Beijing Science and Technology Innovation Medical Development
Foundation (KC2021-JX-0044-2).
Ethics committee - additional information: This study was approved by the Ethics Committee of the First Affiliated Hospital of
Sun Yat-Sen University (No.2020-536).
The effect of GD-DTPA on the apparent diffusion coefficient of breast lesions (7 min)
Lihua Qiu; Yibin / China
Author Block: L. Qiu1, Y. Ao1, H. Chen1, N. Zhang2; 1Yibin/CN, 2Shenzhen/CN

Purpose: This study aimed to explore the effect of gadolinium contrast agent on ADC of breast lesions and the value of ADC at
different time points in distinguishing benign and malignant breast lesions. Studies have shown different effects on the apparent
diffusion coefficient (ADC) of 3T magnetic resonance after gadolinium contrast injection, which we suspect may be related to the
scanning time point after contrast injection.
Methods or Background: The ADC values of 272 lesions confirmed by biopsy or surgical pathology before and after enhancement
were retrospectively analysed. The differences in ADC values of invasive ductal carcinoma, ductal carcinoma in situ and benign
tumours at 3 minutes, 13 minutes after enhancement and before enhancement were compared, and the diagnostic efficacy of ADC
values at different time points before and after enhancement was analysed.
Results or Findings: The ADC value of the invasive ductal carcinoma group decreased significantly at 3 minutes after enhancement,
and the difference was statistically significant compared with that before enhancement (P <0.001). There was no significant
difference in ADC value between the other pathological types before and after enhancement. The AUC of ADC values before and 3
minutes after enhancement, before and 13 minutes after enhancement in the diagnosis of benign and malignant breast lesions were
0.974, 0.966, 0.978 and 0.967 respectively.
Conclusion: The ADC value of invasive ductal carcinoma group decreased significantly at 3 minutes after enhancement. These
changes have negligible effects in distinguishing between benign and malignant breast tumours and do not affect the ability to
differentiate different types of lesions. Thus, it is recommended to collect the DWI sequence after enhancement.
Limitations: Not all patients were studied for ADC values at different times after GD-DTPA enhancement.
Funding for this study: Funding for this study was received from the Key Research and Development Plan of Sichuan Province
(2023YFQ0011). The Project of Health Medical Research of Yibin City in 2023 (No. 2023YW008).
Ethics committee - additional information: The ethics committee notification can be found under the number 2023-109-01.
Proportion of suboptimal baseline 18F-FDG PET/CT exams in oestrogen positive breast cancer patients according to
18F-FDG uptake in the primary tumour: a single-centre retrospective analysis (7 min)
Melissa Lenaerts; Peer / Belgium
689
Author Block: M. Lenaerts1, L. Van der Voort1, M. Smidt1, C. Van De Weijer1, V. Tjan-Heijnen1, F. A. Gallagher2, L. Aloj2, T. van
Nijnatten1; 1Maastricht/NL, 2Cambridge/UK
Purpose: This study aimed to assess the proportion of ER-positive tumours with suboptimal 18F-FDG uptake. The extent of breast
cancer is an important prognostic factor in locally advanced breast cancer (LABC). Therefore, staging using 18F-FDG PET/CT is
recommended. However, previous studies have confirmed a significantly lower degree of 18F-FDG uptake in primary breast tumours
of the oestrogen receptor (ER)-positive subtype compared to other subtypes. Consequently, suboptimal 18F-FDG uptake in ER+ LABC
might lead to suboptimal staging.
Methods or Background: Baseline 18F-FDG PET/CT scans of female patients diagnosed with ER+ LABC in the Maastricht University
Medical Centre between 2011-2022 were retrospectively collected. The maximum standardised uptake value (SUVmax) of the primary
tumour was measured. Different SUVmax cut-off values were applied to determine the proportion of suboptimal 18F-FDG PET/CT
exams. Multivariable logistic regression was performed to determine the possible correlation between clinicopathological predictors
and the SUVmax of the primary tumour.
Results or Findings: 74 patients were included. SUVmax cut-off values of 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, and 5.0 correspond with a
proportion of women with a primary tumour SUVmax below the cut-off value of 6.8%, 16.2%, 24.3%, 31.1%, 39.2%, 40.5%, 45.9%,
respectively. When considering 3.0 as an arbitrary cut-off value for SUVmax of the primary tumour, multivariable logistic regression of
both ER-percentage (10-100%) and tumour grade (1-2 vs. 3) showed a lower tumour grade to be significantly correlated with a lower
SUVmax (0.07 [0.008–0.562]; p=0.013).
Conclusion: A considerable proportion of ER+ LABC patients have a relatively low SUVmax value of the primary tumour, indicating a
potential suboptimal staging on the baseline 18F-FDG PET/CT exam. When considering SUVmax 3.0 as an arbitrary cut-off value of the
primary tumour, 24.3% of the patients might have a suboptimal baseline 18F-FDG PET/CT exam.
Ethics committee - additional information: Due to the retrospective design of the study, approval was waived by the local ethics
committee
Breast imaging reporting and data system in contrast-enhanced mammography (CEM): "Lesion Conspicuity”
correlation with malignancy and tumour receptor pattern (7 min)
Davide Pupo; Naples / Italy
Author Block: D. Pupo1, L. Nicosia2, A. C. Bozzini2, L. Mariano3, G. Signorelli2, F. Abbate2, F. Priolo2, E. Cassano2; 1Naples/IT, 2Milan/IT,
3
Turin/IT
Purpose: This study aimed to evaluate the diagnostic performance and relationship with the receptor profile of this new descriptor.
The new version of the Breast Imaging Reporting And Data System (BIRADS) related to contrast-enhanced mammography (CEM)
encourages investigations on a new enhancement descriptor: "Lesion Conspicuity" (LC): defined as the intensity of enhancement of a
lesion in relation to the background parenchymal enhancement (BPE).
Methods or Background: 325 patients with 381 breast lesions undergoing CEM before histological evaluation were selected. Four
radiologists, in a blind study, classified LC into the following levels: absent, low, moderate and high. Considering "moderate" and
"high" as predictive of malignancy, the diagnostic performance of CEM was calculated using histological biopsy results as the gold
standard. The association between LC values and receptor profile of neoplasms was also evaluated.
Results or Findings: The mean age at CEM examination was 50 years (range: 45-59). Considering the LC value of the most
experienced radiologist in Low Energy (LE) image interpretation, we obtained a sensitivity (SE) of 91.9% (95% CI: 88.6%-95.2%) and
specificity (SP) of 67.2% (95% CI: 58.9%-75.5%). An association was observed between "high" lesion conspicuity with unexpressed
ER/PgR (p=0.025), with Ki-67 > 20% (p=0.033) and with G3 grading (p=0.020).
Conclusion: The new descriptor, "Lesion Conspicuity," demonstrated more than satisfactory performance in predicting lesion
malignancy and significant correlation with the receptor profile of malignant breast neoplasms.
Limitations: This was a monocentric study.
Ethics committee - additional information: The retrospective study was approved by The Institutional Review Board.
690
OF 16T - Back to the basics: classic signs in radiology
Categories: Education, General Radiology, Genitourinary, Head and Neck, Paediatric, Students
ETC Level: LEVEL I
Date: March 1, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

Classic signs in head and neck radiology (15 min)

1. To illustrate the most common classic signs related to head and neck radiology.
2. To discuss their implications in daily practice and potential pitfalls of interpretations.
Classic signs in paediatric radiology (15 min)

Maria Raissaki; HERAKLION / Greece
1. To illustrate the most common classic signs related to paediatric radiology.

2. To discuss their implications in daily practice and potential pitfalls of interpretations.
Classic signs in uroradiology (15 min)

Aart J. Van Der Molen; Leiden / Netherlands
1. To illustrate the most common classic signs related to uroradiology.

2. To explore how these will impact radiology practice through case examples.
Panel discussion: What is the role of classic signs in radiology today? (10 min)
691
MS 16 - Inflammatory bowel disease
Categories: GI Tract, Imaging Methods

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Stuart Andrew Taylor; Prestwood / United Kingdom
What the gastroenterologist wants (15 min)

Sara McCartney; LONDON / United Kingdom
1. To understand how a diagnosis of IBD is made.

2. To appreciate state of the art medical therapy in IBD.
3. To learn the clinical questions the gastroenterologist needs imaging to answer.
What the surgeon wants (15 min)

Christopher Wood; Henley-on-Thames / United Kingdom
1. To understand the role of surgery in IBD.

2. To learn the commonly performed surgical procedures including stricturoplasty techniques.
3. To appreciate the clinical questions the surgeon needs imaging to answer before and after operating.
The role of the histopathologist (15 min)

Ruma Saraswati; London / United Kingdom
1. To understand the basic histopathological findings in IBD.

2. To appreciate the overall strengths and limitations of histopathology in IBD.
3. To learn the histopathological composition of IBD-related strictures.
What can radiology offer (15 min)

1. To understand the strengths and limitations of the available imaging modalities in IBD.
2. To learn how imaging can help differentiate between active and inactive disease.
3. To appreciate the most efficient use of imaging in the immediate post-operative period.
Case discussion: Complex IBD, the power of the team (25 min)
692
iViolin - Objective image quality determination in oncological patient images
Categories: EuroSafe Imaging/Radiation Protection, Oncologic Imaging, Physics in Medical Imaging, Research
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Workshop description:
In this workshop you can test and evaluate a web-based tool for image quality analysis of thorax CT or abdomen CT images. In this
initial version, the quality is assessed in an area surrounding the major fissure of the left lung or in some parts of the abdomen CT
images. To work with your own data in this workshop, please prepare the following:
• CT DICOM images of the thorax / lung where the major fissure is visible or CT DICOM images of the abdomen where the psoas
muscle is visible.
• [x, y] coordinate of a point on the major fissure in thorax images, i.e. the major fissure should cross these coordinates or [x, y]
coordinates of a point on the psoas muscle in abdomen images.
The workshop targets radiologists, radiographers, and medical physicists interested in image quality analysis.
Moderator:

Determine image quality in patient images based on physical parameters (15 min)
1. To learn about the assessment of physics-based metrics for image quality in patient images.
2. To appreciate the relevant input from radiologists and radiographers.
3. To understand the potential of image quality assessment in patient images for optimisation.
Image quality assessment based on physics (15 min)

1. To learn about physics-based metrics for image quality assessment.

2. To appreciate the different approaches like Fourier-based and task-based measures.
3. To understand the limitations of phantom-based measurements.
Requirements on clinical data for setting up sufficient databases (10 min)

1. To learn about data structures needed for this approach.

2. To appreciate the necessary input from radiologists, medical physicists and radiographers.
3. To understand the potential implementation into the workflow.
Hands-on workshop: Evaluate your own images or use existing data sets to test the methods provided by the i-Violin
project for image quality assessment in patient images (45 min)
693
Zahra Passand; Magdeburg / Germany
Dimitar Petrov; Leuven / Belgium
694
E³ 1623 - Paediatric
Categories: Abdominal Viscera, Chest, GI Tract, Neuro, Paediatric

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

Paediatric neuroimaging (28 min)

1. To become familiar with the normal development of the brain.

2. To learn about the most common congenital disorders of the brain.
3. To learn about the most common brain tumours in children.
Paediatric chest imaging (28 min)

1. To describe the normal development of the lung and mediastinum.

2. To explain the imaging features of congenital disorders of the lung and mediastinum.
3. To understand the imaging manifestations of respiratory distress and bronchopulmonary dysplasia in infants.
4. To describe the most common tumours of the chest in children.
Paediatric abdominal imaging (28 min)

Giulia Perucca; London / United Kingdom
1. To understand the imaging features of congenital disorders of the abdomen.

2. To describe the diagnostic evaluation and imaging presentation of the most common emergencies in children according to age.
3. To understand the imaging presentation of the most common oncologic disorders of the abdomen in children.
695
SF 16 - Patient-centred care
Categories: Education, Management/Leadership, Professional Issues

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

Knowledge is power: appropriate communication with our patients (15 min)

1. To empathetically convey the diagnosis of cancer in a manner the patient can understand.
2. To effectively communicate the next steps of the management.
3. To understand how much information the patient can understand at different points in their journey and provide links to additional
support.
Educating the healthcare providers (15 min)

1. To show the importance of educating radiologists on patient-centred care.

2. To highlight the role of adequate communication.
3. To describe ESOR activities in optimising patient-centred care.
The radiographer's perspective (15 min)

1. To outline the radiographers' role in promoting high-quality patient-centred care.

2. To discuss contemporary opportunities for improving care within radiographic practice.
3. To highlight the role of professional bodies in optimising patient-centred care.
What does a patient need? (15 min)

1. To understand the responsibility of a radiologist towards a patient.

2. To learn about the opportunities of patient care on a multidisciplinary level.
3. To appreciate the benefits and tools of patient-centred care with a 360 degree approach.
The ethical perspective (15 min)

Ilenia Rapisarda; Aci Sant'Antonio, Catania, Sicilia / Italy
1. To understand the relationship between ethics and artificial intelligence.

2. To highlight the instrumental nature of healthcare robotics.
3. To confirm the personal and irreplaceable nature of the care relationship between healthcare professional and patient.
696
Panel discussion: How do we make sure that our patients are in the centre? (13 min)
697
E³ 1620 - Access to interventional radiology training
Categories: Interventional Radiology, Professional Issues, Students

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
The session summarises the expected training for an interventionalist in the main regions worldwide. The audience will also
familiarise themselves with the relevance of undertaking an international certification.
Moderator:

IR training in Europe (18 min)

Rok Dežman; Ljubljana / Slovenia
IR training in Northern America (18 min)

Monica M Matsumoto; Philadelphia / United States
IR training in Africa and the Middle East (18 min)

Mohamed Elsayed Elbadawy Mostafa Fouad; Frankfurt am main / Germany
The EBIR exam: the advantages of an international certification (18 min)

Panel discussion: Which actions should be taken to harmonise IR training worldwide? (13 min)
698
E³ 1621 - Sport injuries
Categories: Emergency Imaging, General Radiology, Imaging Methods, Musculoskeletal

Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Wrist (45 min)

Marco Zanetti; Baden / Switzerland
1. To become familiar with the most common clinical features and mechanisms of injury to the wrist in athletes.
2. To learn about the use of imaging to detect sport-related injuries to the wrist.
Hand and fingers (45 min)

Gerd Bodner; Vienna / Austria
1. To become familiar with the most common clinical features and mechanisms of injury to the hand and fingers in athletes.
2. To learn about the use of imaging to detect sport-related injuries to the hands and fingers.
699
NH 16 - The brain's waste facility: the glymphatic system
Categories: Imaging Methods, Neuro

ETC Level: LEVEL II
Date: March 1, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
E. Turgut Tali; Ankara / Turkey

Tchoyoson Lim; Singapore / Singapore
Current concepts in intracranial interstitial fluid transport and the glymphatic system (20 min)
Shinji Naganawa; Nagoya / Japan
1. To name and identify the two major concepts of the waste clearance system of the brain.
2. To list three of the causes for the impaired function of the glymphatic system.
3. To describe and reflect on how impaired function of the glymphatic system relates to neurodegenerative diseases.
Imaging techniques for the glymphatic system (20 min)

Geir Andrè Ringstad; Oslo / Norway
1. To name and identify advantages and disadvantages of intrathecal vs intravenous contrast-enhanced MRI for imaging the
glymphatic system.
2. To describe how MRI can be employed to study driving forces of the glymphatic system.
3. To describe how ultra-high field MRI can be used to measure CSF-mobility in perivascular spaces.
Glymphatic waste clearance and neurological diseases (20 min)

Toshiaki Taoka; Nagoya / Japan
1. To summarise the dynamics of the cerebrospinal fluid (CSF) and interstitial fluid (ISF) dynamics.
2. To understand the mechanism in the various pathological conditions in which the CSF/ISF dynamics are impaired.
3. To describe the imaging technique to evaluate the CSF/ISF dynamics.
Panel discussion: Cerebrospinal, interstitial fluids and glymphatic system (25 min)
700
ST 12 - European Radiology Experimental - Novel approaches moving radiology forward

Date: March 1, 2024 | 16:30 - 17:00 CET
The Editor-in-Chief of European Radiology Experimental, Prof. Francesco Sardanelli (Milan/IT) explains the scope of the journal, the
recent receipt of the Impact Factor and thematic series, and his vision for the future.
Moderators:
Interview (30 min)

701
ST 13 - How to start or boost your academic career with an ESOR Exchange Programme
Date: March 1, 2024 | 17:15 - 17:45 CET

The European School of Radiology (ESOR) is an institution, fulfilling the mission of the European Society of Radiology (ESR) in the field
of education. One of its main goals is to assist in harmonising radiological education in Europe. With its wide range of activities ESOR
additionally aims to raise standards in the field of scientific radiology, to extend and coordinate teaching resources worldwide and to
help young radiologists to achieve the knowledge and skills to fulfil tomorrow’s requirements.
In her function as ESOR Educational/Scientific Director, Prof. Valérie Vilgrain will give an overview of the broad spectrum of short-term
and long-term training programmes ESOR has on offer for radiology residents as well as recently board-certified radiologists. In
addition, she will also talk about her personal experience as mentor of many ESOR scholars and fellows over the past years and the
win-win situation that ESOR training programmes represents also for European training centers that enrol as host institutions.
Dr. Fulvio Renoldi Bracco, CEO of Bracco Imaging SpA, will inform about the value of education in partnership and the long-standing
and fruitful cooperation between ESOR and Bracco over nearly two decades. Young radiologists, thanks to the many projects
organised in partnership between Bracco Imaging and ESOR, can learn and grow and contribute to innovation in imaging for the
benefit of patients.
Dr. Angela Ammirable, Dr. Filipa Coelho, Dr. Petr Matkulcik and Dr. Irina Stavarache are young radiologists who have only recently
been awarded an ESR Scholar- or Fellowship grant. They will give an insight into the impact the ESOR training abroad had on their
professional and personal lives.
Moderator:
Interview (15 min)

Fulvio Renoldi Bracco; Milano / Italy
Interview (15 min)

Filipa Coelho; Porto / Portugal
Peter Matkulcik; Brno / Czechia
Irina Stavarache; Bucuresti / Romania
702

Date: March 1, 2024 | 18:00 - 18:15 CET
Moderators:
Interview (30 min)
703

Date: March 2, 2024 | 07:50 - 08:00 CET
Moderators:
Interview (30 min)

704
RPS 1714 - Enhancing patient experience and safety in medical imaging and radiotherapy
Categories: Evidence-Based Imaging, Professional Issues, Radiographers

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderators:
Ana Blanco Barrio; Murcia / Spain
Radiographers’ current experiences interacting with patients with dementia and cognitive impairment during magnetic
resonance imaging (7 min)
Author Block: S. Hanley, P. C. Murphy, R. Young, N. Moore, A. England, M. F. F. McEntee; Cork/IE

Purpose: Dementia is the umbrella term used to describe several conditions, causing damage to the brain. This study aims to
investigate radiographers’ current experiences interacting with patients with dementia and cognitive impairment during MRI.
Methods or Background: This is a prospective quantitative study consisting of a questionnaire completed by MRI radiographers.
The study location was in a single public hospital in the Republic of Ireland, performed over a six week period. The inclusion criteria
required "dementia", "cognitive impairment", or "confusion" to be documented in the request form, detailed by the ward staff, or if
any one of these were discovered by the MRI radiographers when the patient arrived.
Results or Findings: There were twenty questionnaires completed. The point when it was discovered the patient had dementia,
cognitive impairment or confusion was predominantly through contact from the ward, accounting for 50%. Not one out of the twenty
patients required someone being present during the scan. 30% of patients were agitated or distressed during the MRI examination.
15% of examinations needed to be aborted and rebooked and 25% were incomplete.
Conclusion: There were issues in discovering the patient had dementia, cognitive impairment, or confusion. There were implications
regarding examinations needing to be aborted and rebooked. This is a concern as MRI is a very sought after modality. This is an area
which needs to be addressed as these examination slots are vital for assisting patients along their clinical pathway.
Limitations: Identified limitation were (1) the single-centre nature of the study and (2) the relatively small sample size.
Radiographers’ experience during medical imaging of patients with dementia in Norway (7 min)
Christine Eikefet; Borre / Norway
705
Author Block: C. C. Chilanga1, C. Eikefet2, E. Kjelle2; 1Svelvik/NO, 2Drammen/NO
Purpose: This study aimed to explore radiographers’ experiences during imaging examinations of people living with dementia in
Norway.
Methods or Background: Semi-structured qualitative interviews were conducted with eight radiographers, four working in MRI or
general radiography and four working in nuclear medicine, at three different hospital trusts in Norway. The interview guide included
the following topics: radiographers’ experience of working with patients living with dementia, challenges faced when conducting
examinations for these patients, knowledge about dementia and initiatives in the department. All interviews were verbatim
transcribed and inductive content analysis as described by Elo and Kyngäs was used to analyse the data.
Results or Findings: The analysis resulted in three main categories, each with two to five subcategories. The main categories were
"radiographers experience", "measures taken to accommodate for the patient" and "competencies". The radiographers frequently
encountered patients with dementia in the department. The challenges they faced included a lack of information before receiving
patients with dementia, communicating with the patients, and ensuring their stability during the procedure. MRI safety was of
particular concern when communication and information sharing were problematic. None of the departments had any overarching
procedures or training related to patients with dementia. Creating a calm environment, collaborating with carers, scheduling
adequate time for examinations, and possessing good communication skills were viewed as facilitators for conducting examinations
successfully.
Conclusion: Radiographers experienced managing and imaging patients living with dementia to generally be uncomplicated. The
knowledge on examining patients with dementia was mostly acquired through clinical practice. However, in some cases the
department’s environment, and communication problems caused stress and restlessness in patients.
Limitations: The limitation of this study is that findings from interviews with a few radiographers in Norway may not be easily
generalised to other settings.
Ethics committee - additional information: The Norwegian Agency for Shared Services in Education and Research approved the
treatment of personal information (project reference: 155338)..
An investigation into the perceived challenges and facilitators Parkinson’s disease patients undergo when accessing
radiology examinations (7 min)
Author Block: B. Treanor, M. F. F. McEntee, N. Moore, R. Young, R. O'Regan, A. England; Cork/IE

Purpose: Parkinson's disease (PD) is a progressive neurological disorder characterised by a wide range of symptoms. Understanding
the facilitators and barriers for patients with PD accessing radiology is paramount, as this patient group would require frequent access
to imaging and the nature of their condition can pose additional challenges. The aim of this study is to investigate the perceived
barriers or facilitators to radiology examinations that improve or degrade the experiences for patients with PD by conducting an
experience-based focus group.
Methods or Background: This was a qualitative study in which the experiences of patients with PD were studied by conducting an
audio-recorded experience-based focus group. Recruitment and data collection took place from March 2023 until April 2023. The
primary researcher then personally validated the transcription. Following this, the primary reviewer and secondary reviewer each
performed a thematic analysis of the transcription with no discrepancies found between both findings.
Results or Findings: There were eight participants in this study. There were three key themes identified for the facilitators which
included "supportive hospital staff", "communication" and "knowledge". There were also three main themes identified for the barriers
which included "anxiety", "knowledge", and "communication".
Conclusion: By addressing these facilitators and barriers, we can work towards creating a more supportive and patient-centred
radiology environment for individuals living with PD. Further research is warranted to explore the experiences of patients with PD in
radiology departments across different healthcare settings and geographical regions.
Limitations: Responses were received from a single service user group in Ireland. Involving participants from more groups or
undertaking an international study would be advantageous.
Ethics committee - additional information: This study was approved by Medical School SREC, University College Cork.
The effect of inhalation aromatherapy on patient anxiety in nuclear medicine (7 min)

Karen Borg Grima; Naxxar / Malta
706
Author Block: J. Sciberras, K. Borg Grima; Msida/MT
Purpose: The aim of this study was to evaluate if chamomile intervention has an impact on the anxiety levels of patients undergoing
a bone or thyroid scan in nuclear medicine.
Methods or Background: A quantitative, experimental, prospective and cross-sectional design was employed for this study. Fifty
participants who complied with this study’s inclusion and exclusion criteria were enlisted. Using convenient sampling, the participants
were equally distributed between the experimental and control group. In both groups, routine protocol was followed when scanning
the patients, apart from those in the experimental group, who were additionally introduced to the chamomile smell. Following this,
the State-Trait Anxiety Inventory Questionnaire was used to measure the participants’ anxiety level before and after the scan, in both
groups.
Results or Findings: Participants in both the experimental and control groups experienced higher pre-scan state anxiety (37.84,
40.20) scores when compared to post-scan state anxiety scores (26.24, 37.12). Furthermore, the post-scan state anxiety score of the
experimental group was significantly lower compared to the control group’s post-scan state anxiety score (P<0.001). In addition,
when analysing the difference in anxiety levels between genders a statistical significance was noted. The results indicated that
females were more vulnerable to anxiety reactions.
Conclusion: From this study’s findings, there was statistical significance which indicated that chamomile intervention was effective
at lowering the patients’ anxiety levels in nuclear medicine. Additionally, the results of this study suggested that chamomile
intervention should be used in clinical practice to reduce patients' stress levels since it is non-invasive and cheap.
Limitations: The small sample size used in this study could have influenced the accuracy of the results. For future research, a larger
sample size should be applied and data should be collected over a longer period of time.
Ethics committee - additional information: This study was approved by the University of Malta Research Ethics Committee
(UREC), and by the Data Protection officer of a state general hospital where this research was conducted.
Evaluation of the influence of music in women undergoing radiation therapy (7 min)

Magda Cruz Ramos; Olhao / Portugal
Author Block: M. C. Ramos1, A. Rosa2, F. Serra1, A. F. Abrantes2, L. P. V. Ribeiro2, S. I. Rodrigues2; 1Olhao/PT, 2Faro/PT
Purpose: Breast cancer is the most commonly diagnosed cancer among women worldwide. Patients often experience anxiety,
particularly during radiotherapy, due to the disruptions in their daily routines and the associated side effects. Music therapy has
emerged as a promising psychosocial intervention, offering potential benefits such as reduced anxiety, pain relief, and an improved
quality of life for cancer patients. The primary aim of this research was to conduct a comprehensive examination of how these factors
collectively impact anxiety levels in breast cancer patients undergoing radiotherapy.
Methods or Background: This study assume a comparative analysis between two groups of breast cancer patients. The assessment
tools utilised consisted of a 'yes' system, a concise interview, the STAI (State-Trait Anxiety Inventory) questionnaire, and a visual
analog comfort scale. To assess anxiety levels, the STAI questionnaire was administered twice: at the initiation of the treatment
regimen and upon its completion.
Results or Findings: This study assessed anxiety levels, heart rate, education status, and comfort levels, along with examining the
influence of music on anxiety during radiotherapy. The findings indicated no noteworthy age-related differences, but they did reveal a
moderate correlation between anxiety and heart rate in the experimental group. No significant correlations were identified between
anxiety and education status or comfort levels. Regarding the impact of music on anxiety between the groups, lower mean values of
final anxiety were observed in the experimental group (64.00 ± 2.94) compared to the control group (68.65 ± 3.69).
Conclusion: While this study did not yield statistically significant differences in anxiety reduction through music during radiotherapy,
its findings remain valuable and instructive. The feedback transmitted by the patients of the experimental group was shown to be
favourable for the use of music, regardless of whether the results were significant or not.
Ethics committee - additional information: This study received authorisation from the institutional board.
Using the e-Delphi method to develop informative nutritional leaflets for patients undergoing radiotherapy to the head
and neck (7 min)
Duncan Munro; Haz-Zebbug / Malta
707
Author Block: D. Munro, G. V. Dijk, J. G. Couto; Msida/MT
Purpose: Patients diagnosed with head and neck (HN) cancer undergoing radiotherapy (RT) frequently encounter distressing side
effects that significantly impede their ability to consume food. This often results in deteriorating nutritional status and subsequent
weight loss. The primary objective of this study was to evaluate the efficacy of a modified e-Delphi approach in creating informative
dietary advice pamphlets aimed at addressing these side effects, while garnering consensus among healthcare professionals.
Methods or Background: An e-Delphi methodology was employed, involving six participants representing various healthcare
professions specialising in HN patient care (radiographers, nurses, oncologists). These participants were tasked with providing
feedback on four dietary leaflets tailored to specific symptoms, all meticulously crafted based on an extensive prior literature review.
Following each round of feedback, the participants' recommended modifications were incorporated. Significant alterations to the
leaflets were subjected to participant voting before implementation.
Results or Findings: After three rounds of deliberation, unanimous consensus was achieved, as all participants expressed a "highly
likely" inclination to incorporate the leaflets into their clinical practice. The majority of the participants' suggestions were consistent
with existing literature. The only change that deviated from the literature, and was accepted through voting, pertained to sugar
consumption.
Conclusion: The participants successfully attained consensus and developed leaflets aligned with literature recommendations, which
they deemed suitable for clinical application. This variation of the e-Delphi method demonstrated its efficiency in establishing
consensus among healthcare professionals concerning patient information resources.
Limitations: The participant pool could have been larger. However, all the radiotherapy professions were represented in the sample.
Including a dentist could have been beneficial.
Ethics committee - additional information: The researchers submitted the research for records to the University of Malta’s
University Research Ethics Committee (reference number: FHS-2022-00039) following a self-assessment that showed that the
research was low risk and did not require formal ethical approval.
Radiographers’ knowledge, clinical expertise and application of pain management strategies in the radiology
department (7 min)
Author Block: R. O'Regan, N. Moore, B. Treanor, R. Young, M. F. F. McEntee, A. England; Cork/IE

Purpose: Pain is a widely encountered medical concern that necessitates effective pain management interventions. Pain
management challenges are reported within the literature as having a significant impact on medical practice, healthcare
professionals, radiology departments, and the healthcare system. Radiographers frequently image patients who are in pain and thus
require a heightened awareness and focus on this crucial issue. This study aims to evaluate radiographers' knowledge, clinical
expertise, and challenges associated with pain management in the radiology department.
Methods or Background: A qualitative research study consisting of a single focus group was conducted with six participants. The
research questions focused on pain management challenges in the radiology department, with a particular emphasis on identifying
radiographers' current practices and strategies to address these obstacles. The focus group study was moderated and video-recorded
to facilitate comprehensive analysis. The data obtained from the study was transcribed and thematically analysed.
Results or Findings: Data analysed from the focus group identified four main themes: consequences of pain management,
communication, professional experience, and barriers. In addition, the study identified there are no current protocols, policies, and
guidelines in practice for effectively managing the challenges associated with imaging patients in pain within the radiology
department.
Conclusion: The focus group study revealed an array of concerns related to the challenges associated with imaging patients in pain
within the radiology department. The primary consequences identified were stress experienced by both staff and patients, the
potential for suboptimal images, patient safety concerns, and negative effects on image quality. The results of this research study
demonstrate how patient discomfort negatively affects examinations and identifies the potential consequences for radiographers
aspiring to achieve optimal gold standard imaging.
Limitations: Identified limitations were: (1) the small sample size and (2) that participants were from a single hospital group in
Ireland.
Ethics committee - additional information: This study was approved by Medical School SREC, University College Cork.
Deaf-friendly MRI: radiographers' communication toolkit (7 min)

Eleonora Stefani; Treviso / Italy
708
Author Block: E. Stefani, M. Centenaro, G. Cappellina; Treviso/IT
Purpose: Equality is vital in healthcare and communication is part of It. Most systems are adequately prepared to communicate with
certain type of patients but they don’t represent the majority of them. It is necessary to adjust communication to all types of patients.
To tailor radiographers' communication to all types of patients is essential to ensure patient safety, compliance and image quality.
The study aims to identify tools to improve radiographers' effective communication with deaf patients who undergo MRI examination.
Methods or Background: This was a qualitative study. One focus group has been created including radiographers from three
radiology departments in a hub and spoke organisation. A total of 25 radiographers participated in the data collection. Data was
analysed using ATLAS.ti software.
Results or Findings: 85% of radiographers feel a sense of discomfort when dealing with deaf patients and only 20% of them know
about policies supporting deaf patients in the radiological divisions they work in.
Communication failures may occur due to: lack of knowledge about how to interact with deaf patients (90% of radiographers);
difficulty in calling deaf patients in the waiting room (35%); and scheduling LIS interpreters (60%).
Three tools have been identified in support of radiographers: (1) a visual system to call patients; (2) guidelines with essential points
for interacting with the deaf patient; (3)
a video in LIS concerning safety in MRI has been realised.
Conclusion: Communication is a big part of the success of MRI radiological examination. Radiographers are the main interface to
patients and it is important to tailor communication approaches to their needs and to get feedback from them. Tools supporting
inclusion for deaf patients optimise image quality and enhance patient experience.
Limitations: It would be worth assessing radiographers both within the public and private sectors.
709
RC 1707 - Overview of "RADS" in genitourinary

ETC Level: LEVEL II
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Jelle O. Barentsz; Arnhem / Netherlands

Jelle O. Barentsz; Arnhem / Netherlands
Kidney's score (15 min)

Refky Nicola; Syracuse / United States
1. To understand the potential benefits of developing a KI-RADS scoring system.

2. To learn the complexities of moving from risk of cancer to risk of aggressive cancer.
3. To learn how a scoring system might join cystic and solid masses into a single paradigm.
O-RADS (15 min)

2. To learn about the validation of the score in the literature.
3. To understand the clinical implications of the score.
VI-RADS (15 min)

2. To learn about the validation of the score in the literature.
3. To understand the clinical implications of the score.
Panel discussion: Genitourinary-RADS as living documents: what's next? (10 min)
710
RPS 1705 - Radiomics applications in MDCT
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Physics in Medical Imaging
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Valentina Giannini; Torino / Italy
A reference framework for standardisation and harmonisation of CT radiomics features: the "CadAIver" analysis (7 min)
Riccardo Levi; Pieve Emanuele / Italy
Author Block: R. Levi1, M. Mollura2, G. Savini1, F. Garoli1, M. Battaglia1, A. Ammirabile1, L. A. Cappellini1, R. Barbieri2, L. S. Politi1;
1
Pieve Emanuele/IT, 2Milan/IT
Purpose: The purpose of this study was to quantify the effects of different dose protocols, reconstruction algorithms, fields of view
(FOVs) and CT scanners on radiomics features (RFs) of the lumbar vertebrae in a cadaveric trunk and to develop a normalisation
algorithm to harmonise radiomics analyses.
Methods or Background: We performed a total of 112 CT acquisitions of a cadaveric trunk on 3 different CT scanners from 2
different vendors, using varying kV (80-140) and mA (250-400). Each acquisition was performed using 2 FOVs (Abdomen/Spine) and
two reconstruction kernels (Standard/Bone). Lumbar vertebrae were segmented using a convolutional neural network and RFs were
extracted using pyradiomics. Intra/Inter-scanner analyses were assessed and each RF was tested using a generalised linear model
(GLM) to assess the effects of all the above-mentioned CT acquisition parameters. GLM model was employed to standardise RFs
across different acquisitions, and was compared to the ComBat algorithm, using 10-folds cross-validation (CV) R2.
Results or Findings: KV variation showed the highest feature modification in intra/inter-scanner analyses, with the FirstOrder
features showing high variability (94.4%, Scanner 1). Little effects were evident upon mA variation. 100% of Shape features on all
scanners were found significantly dependent on FOV, and 83% of GLSZM were statistically different between reconstruction kernels.
The proposed GLM algorithm obtained a mean R2 across CV higher than 0.90 in 21 Radiomics features (19.6%), whereas ComBat
obtained an high R2 value in 1 Radiomics feature (0.90%). GLM was statistically superior in 39 Radiomics features in respect to
ComBat.
Conclusion: This study is the first attempt in describing the effects of CT acquisition parameters on RFs from cadaveric donor.
Current and Voltage affect the RFs in different ways. GLM was superior to ComBat in normalising RFs across all different CT
acquisitions.
Limitations: This study only featured vertebral analysis.
Ethics committee - additional information: Ethics committee approval is not required for the use of cadaveric body parts. The
cadaver was donated willingly to science and obtained from medcure.org (Medcure, Orlando, FL, USA).
Contrast-enhanced CT-based radiomics model predicts chemotherapy efficacy in ovarian metastatic colorectal cancer
patients: a preliminary study (7 min)
Jinghan Yu; Chengdu, Sichuan / China
711
Author Block: J. Yu, H. Zeng, B. Wu; Chengdu/CN
Purpose: Ovarian metastasis (OM) from colorectal cancer (CRC) are uncommon and carry a poor prognosis. Our study aims to
investigate the value of contrast-enhanced CT-based radiomics model in predicting CRCOM outcomes after systemic chemotherapy.
Methods or Background: A total of 52 ovarian metastatic CRC patients who received 1 to 3 first-line systemic chemotherapy were
retrospectively included in this study. All patients were categorised into chemo-benifit (C+) and no chemo-benifit (C-) groups, using
response criteria in solid tumours (v1.1) as the gold standard. The radiomics features were extracted from baseline CT images and
were further selected. Multiple radiomics models were constructed by using different machine learning classifiers such as naïve bayes
(NBB), random forest (RF) and support vector machine (SVM) through leave-one-out cross-validation. The diagnostic performance of
these models was evaluated by receiver operating characteristics (ROC) analysis, and the clinical utilities were assessed by
calibration analysis and decision curve analysis (DCA).
Results or Findings: 25 patients (48.1%) evaluated as partial response and stable disease were classified as C+ group, 27 patients
obtaining progressive disease were classified as C- group. Five radiomics features were selected for model development. Among
those machine learning-based radiomics models, the SVM model showed the best performance in the validation data set, with the
AUC, accuracy, sensitivity and specificity achieving 0.903 (95% CI, 0.788-0.967), 88.5%, 95.7% and 82.8%, respectively. All radiomics
models showed good calibration, and the DCA demonstrated that the SVM model had higher net benefit than other models across the
majority range of threshold probabilities.
Conclusion: The contrast-enhanced CT-based radiomics model using SVM classifier has good discriminatory power in predicting the
outcome of patients with ovarian metastatic colorectal cancer receiving chemotherapy and has the potential to be used as a non-
invasive tool for clinical practice.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Department of Radiology,
West China Hospital, Sichuan University, China.
Development of a radiomic model to predict the risk of hepatocellular carcinoma in cirrhotic patients (7 min)
Ludovica Leo; Parma / Italy
Author Block: L. Leo, S. Schirò, O. Bazzini, F. Cazzato, D. Stefanelli, M. V. Bazzocchi, G. Milanese, G. Missale, N. Sverzellati; Parma/IT
Purpose: Patients with cirrhosis are at increased risk of hepatocellular carcinoma (HCC); clinical and radiological scores for predicting
the risk of developing HCC are currently unknown. The aim of the study is to assess if radiomics can identify cirrhotic patients at risk
of HCC.
Methods or Background: 98 subjects (M:F = 64:34; mean age 67 years ±9.81) were included in this retrospective monocentric
study. Two groups were identified: group (a) with 49 patients (M:F=31:18; mean age 67.4 years ±9.12) who had a baseline CT with
radiological signs of cirrhosis and follow-up CTs without evidence of HCC (LI-RADS 1-3); group (b) with 49 patients (M:F = 33:16; mean
age 67.5 years ±10.5) who had a baseline CT with radiological signs of cirrhosis and evidence of HCC in one follow-up CT scan (LI-
RADS 4/5). Four radiologists (three years of experience) provided complete liver segmentations by manually drawing volumes of
interest (VOI) on non-enhanced baseline CT scans, extracting 851 radiomic features (RF). 100 train:test (0.7:0.3) splits were created
and recursive feature elimination with a 5-fold cross-validation was performed on train partitions using the random forest classifier
(RFC). Subsequently, RFC was trained by selecting iteratively an increasing number of features sorted by their occurrences to
evaluate the minimum number of informative features. Finally, means and 95% confidence intervals of accuracy, sensitivity,
specificity, precision, area under the receiver operating characteristic curve (ROC-AUC) were calculated on the test partitions.
Results or Findings: The best model exploited two features (wavelet-LLH_glcm_DifferenceAverage, wavelet-
HLH_gldm_DependenceVariance), reaching accuracy, sensitivity, specificity, precision and ROC-AUC of 0.72 [0.70-0.73], 0.76
[0.74-0.79], 0.68 [0.65-0.70], 0.71 [0.69-0.73] and 0.78 [0.76-0.79], respectively.
Conclusion: Our radiomic model, evaluated in cross-validation, predicted the development of HCC with satisfactory performances
and was more sensitive than specific.
Limitations: There was a monocentric retrospective cohort and no external validation.
Ethics committee - additional information: This study was approved by an ethics committee; 930/2022/OS/UNIPR (Protocol 5853
07/02/2023).
Predicting progression free survival (PFS) of pancreatic ductal adenocarcinoma (PDAC) after curative surgery using CT
radiomics analysis (7 min)
Ming He; Guangzhou / China
712
Author Block: M. He, X. Dong; Guangzhou/CN
Purpose: The aim of this study was to construct a prediction model for predicting progression free survival (PFS) of pancreatic ductal
adenocarcinoma (PDAC) after curative surgery using CT radiomics analysis.
Methods or Background: A retrospective analysis was performed on 162 PDAC patients who had undergone curative surgery and
enhanced CT before operation. The enrolled patients were followed up to disease progression or for at least 2 years. The tumour was
segmented on three phase images (arterial, portal and delayed phase) to extract the radiomics features. Feature correlation analysis
and Bayesian model averaging were used to select radiomics features. Five-fold cross validation was used for internal validation.
Kaplan Meier curve was used for univariate analysis to select predictive parameters. Random survival forest (RSF) was used to
establish clinical-radiological, radiomics and the combination model. The integrated area under the time-dependent ROC curve (iAUC)
was used to calculate and compare the prediction performance.
Results or Findings: The independent prognostic clinical parameters were CA19-9 level (cut-off value 37U/ml) and tumour
differentiation (poor differentiation vs medium-well differentiation). Portal phase radiomics features (HR = 2.945, 95% CI: 1.980,
4.380, P <0.001), postoperative CA19-9 (HR = 1.596, 95% CI: 1.087, 2.344, P =0.017), and poor differentiation (HR = 1.525, 95% CI:
1.021, 2.276, P =0.039) were included in the combined model and the iAUC was 0.788 (95%CI: 0.745, 0.832), which was statistically
higher than that of the clinical model (0.696 (95%CI: 0.632, 0.735)) (P <0.001).
Conclusion: The combination of the portal phase radiomics features, CA19-9 (cut-off value 37U/ml) and poor tumour differentiation
provide better PFS prediction in resected PDAC patients, but its clinical value still needs further external validation.
Limitations: This was a retrospective study conducted at a single centre, with a relatively small sample size, so selection bias is
inevitable. We did not perform external validation.
Funding for this study: This study received funding from the National Natural Science Youth Fund (No. 82101996).
Ethics committee - additional information: The Guangdong Provincial people's Hospital's review board approved this
retrospective study and written informed consent was waived.
Influence of CT scanners on radiomics features in abdominal CT: a multicentre phantom study (7 min)
Markus Obmann; Basel / Switzerland
Author Block: C. Aberle1, M. Bach1, O. Jimenez-del-Toro2, R. Schaer2, K. Flouris3, E. Konukoglu3, H. Müller2, A. Depeursinge2, M.
Obmann1; 1Basel/CH, 2Sierre/CH, 3Zurich/CH
Purpose: The aim of this study was to investigate the influence of different CT scanners on the stability and discriminative power of
radiomics features using an anatomically accurate 3D-printed abdominal phantom.
Methods or Background: Based on a patient's CT scan with multiple hepatic lesions, an anatomically and texturally realistic
phantom was commercially 3D-printed using potassium-iodide ink on paper. The phantom was scanned on 13 CT scanners by 4
different manufacturers at 8 institutions with 10 scan repetitions each. A harmonised clinical oncologic CT acquisition protocol was
used on all scanners. Images were reconstructed using iterative reconstruction algorithms. 86 radiomics features were assessed for
six different ROIs (metastasis, haemangioma, 2 cysts, 2 normal liver parenchyma regions) using principal component analyses (PCA)
and Kruskal-Wallis tests.
Results or Findings: For all ROIs, PCA analyses clearly showed clustering by scanners and manufacturers, with the same scanner
models overlapping. Kruskal-Wallis tests for each ROI and radiomics feature showed significant differences between scanners in 511
of 516 tests (P <0.05). Pairwise ROI comparison in the PCA showed both separation of the 13 different CT scanners and of the ROIs,
while the separation between ROIs was stronger than between scanners.
Conclusion: In this multicentre study, radiomics features are impacted by CT scanner models in varying degrees, despite the use of
matched acquisition and reconstruction parameters. When performing multicentre studies, an a priori phantom analysis and feature
harmonisation techniques may be used to account for these influences and select more stable radiomics features.
Limitations: As the phantom includes 1-2 ROIs per tissue type, variability of the same tissue type was not studied and results of ROI
separation may not be fully generalisable to tissue type classification. Patient motion cannot be assessed with this phantom and may
aggravate interscanner variations.
Funding for this study: This work was partly supported by the Swiss Personalised Health Network with the QA4IQI Quality
assessment for interoperable quantitative computed tomography imaging project DMS2445 and the IMAGINE project. It was also
partially supported by the Swiss National Science Foundation (grant 205320_179069).
Ethics committee - additional information: This was a phantom study.
Radiomics and machine learning for the assessment of renal tumour histological subtypes on multiphase computed
tomography: a multicentre trial with independent testing (7 min)
Sophie Bachanek; Göttingen / Germany
713
Author Block: S. Bachanek1, A. Leha1, P. Zeuschner2, A. Massmann3, L. Trojan1, J. Uhlig1, J. Lotz1, A. Uhlig1; 1Göttingen/DE,
2
Homburg/DE, 3Stuttgart/DE
Purpose: The purpose of this study was to distinguish histological subtypes of renal tumours identified on multiphase computed
tomography (CT) using radiomic features and machine learning in a multicentre setting.
Methods or Background: Patients undergoing surgical resection and histopathological assessment of renal tumours at two tertiary
urological centres between 2012 and 2022 were retrospectively included. Preoperative arterial and venous phase CTs from multiple
referring imaging centres were segmented and standardised radiomic features extracted. After preprocessing and class imbalance, an
extreme gradient boosting tree-based (XGB) machine learning (ML) algorithm was used to predict renal tumour subtypes using 10-
fold cross-validation, assessed as multiclass AUC. ML algorithms were trained on data from one centre, and independently tested on
data from the other centre.
Results or Findings: The training cohort comprised n=297 patients (n=191 ccRCC, n=40 pRCC, n=22 chRCC, n=28 oncocytomas,
and n=16 AML), and the testing cohort n=121 patients (n=68/ n=20/ n=4/ n=26/ n=3). The XGB algorithm demonstrated a diagnostic
performance of AUC=0.81/ 0.64/ 0.8 for venous / arterial / combined contrast phase CT in the training cohort, and AUC=0.75/ 0.67/
0.75 in the independent validation cohort. In pairwise comparisons, the lowest diagnostic accuracy was evident for identification of
oncocytomas (AUC=0.57-0.69), and the highest for identification of AMLs (AUC=0.9-0.94).
Conclusion: Radiomic feature analyses acquired from clinical routine CT yield robust results for renal tumour assessment. For renal
tumour subtype discrimination, venous phase CT yields the most pertinent imaging information, without evident diagnostic benefit of
an added arterial contrast phase. Among all renal tumours, oncocytomas are hardest to differentiate using CT.
Limitations: Limitations of the study mainly include the heterogeneity of renal tumour subtypes in the training and testing cohort
with occasional resulting low case numbers in the subgroups.
Ethics committee - additional information: The study received ethical approval by the ethics committee at Goettingen University
(No. 2/4/17) and at Saarland University (No. 67/19).
Overcoming data scarcity in radiomics/radiogenomics using synthetic radiomic features (7 min)

Milad Ahmadian; Amsterdam / Netherlands
Author Block: M. Ahmadian, Z. Bodalal, H. J. van der Hulst, C. Vens, L. Karssemakers, R. G. H. Beets-Tan, M. van den Brekel, J. A.
Castelijns; Amsterdam/NL
Purpose: This study evaluates the potential of synthetic radiomic data generation to augment the performance of
radiomics/radiogenomics prediction models.
Methods or Background: This study was conducted on a retrospectively collected cohort of 386 colorectal cancer patients for
whom matched contrast-enhanced CT images and TP53 mutational status were available. Five different tabular synthetic data
generation models were used to generate synthetic radiomic data based on real-world radiomics extracted from our cohort. The
quality and reproducibility of the generated synthetic radiomic data were assessed. Synthetic radiomics were then combined with
real-world radiomic training data to evaluate their impact on the predictive model's performance.
Results or Findings: Using only real-world radiomic data, increasing training samples (n=200, 400, 1000, 2055 lesions) improved
the model's predictive performance on the unseen test set (average AUC=0.52, 0.53, 0.56, and 0.64). Synthetic tabular data
generation models created reproducible synthetic radiomic data with properties highly similar to real-world data (for n =1000 lesions,
average Chi-square =0.932, average basic statistical correlation =0.844). Augmenting predictive models with synthetic radiomics
enhanced their performance by 9.61%, 11.32%, 16.07%, and 3.22% for the outlined training sets, respectively. Synthetic radiomics
derived from randomised/noisy radiomic data failed to enhance predictive performance, while true signal data was effectively
amplified.
Conclusion: Synthetic radiomic data, when combined with real radiomics, can enhance the performance of predictive models.
Tabular synthetic data generation might help overcome limitations in medical AI stemming from data scarcity.
Limitations: Our study was limited to a retrospective monocentric cohort. It would be beneficial to validate our findings on external
cohorts, helping to ensure that the conclusions were not specific to the centre where the data was gathered.
Funding for this study: This work was made possible via a grant from the Hanarth Foundation. The computational infrastructure for
the analysis was funded via the Maurits en Anna de Kock Stichting and the NVIDIA Academic GPU programme.
Ethics committee - additional information: This study was approved by the Institutional Review Board (IRB): IRBd19-147.
Impact of population size and validation method on the performance of radiomics models: application to COVID lung
lesions (7 min)
Antoine Decoux; Vigneux sur Seine / France
714
Author Block: A. Decoux, L. Duron, A. Arnoux, L. S. Fournier; Paris/FR
Purpose: The purpose of this study was to explore the impact of population size and validation strategy on the estimated
performance and reproducibility of radiomics studies.
Methods or Background: Radiomics parameters were extracted from lung lesions segmented by experts on CT in 3,737 COVID-19
patients (STOIC cohort). 1121 (33%) patients were set aside to simulate an external validation population ("generalisation set").
Among the remaining patients, subpopulations of varying sizes were generated to simulate the training/test population for the
radiomics study. Prediction models were trained on 100 bootstrapped samples to estimate variance of the AUC, i.e. model stability.
Three validation strategies were tested: one time split, cross-validation and nested cross-validation. The mean and variance of AUCs
of each model was calculated on the subpopulation as well as on the "generalisation" set, and the difference was the generalisation
gap.
Results or Findings: Increasing the size of the training data sets improved model performance on both internal validation and
generalisation sets, decreased the variance of performance on the validation set and decreased the generalisation bias, thereby
increasing overall confidence in the model, with a plateau at 400 patients.
Cross-validation helped reduce variance and generalisation bias compared to one time split. Nested cross-validation reduced variance
but at the expense of increased generalisation bias.
Conclusion: As expected, population size has a strong impact on model performance, particularly on the estimated performance
variance (stability) of models. This study is the first to estimate the minimum population size needed to improve generalisability of
radiomics studies. However, as it is applied to a single data set, results are expected to vary according to predictive power of imaging
for a given clinical question.
Limitations: Our generalisation set serves as a surrogate for an external validation set, it doesn't constitute a true external
validation set.
Funding for this study: This work was funded by the French government under management of the Agence Nationale de la
Recherche as part of the "Investissements d'avenir" programme, reference ANR19-P3IA-0001 (PRAIRIE 3IA Institute).
715
RPS 1711 - CSF, glymphatic system and MR contrast agents
Categories: Imaging Informatics, Imaging Methods, Multidisciplinary, Neuro, Research, Translational Imaging
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Milos A. Lucic; Sremska Kamenica / Novi Sad / Serbia
In vivo CSF mobility mapping at ultra-high field MRI in patients with cerebral amyloid angiopathy (7 min)
Katerina Deike; Bonn / Germany
Author Block: K. Deike, A. Radbruch, G. Petzold; Bonn/DE

Purpose: Impaired perivascular clearance is a critical step in the pathogenesis of cerebral amyloid angiopathy (CAA), which is
characterized by amyloid-beta accumulation in cerebral vessels. However, only few brain clearance imaging techniques are available
that can be applied non-invasively in humans and the majority of assessed parameters is of indirect and/or semi-quantitative nature.
Therefore, this study aimed to quantifiy the cerebrospinal fluid (CSF) mobility in the PVS of CAA patients and healthy controls (HC).
Methods or Background: To assess perivascular CSF mobility, we applied a specifically for this task developed 7-Tesla MRI
sequence for the first time in patients with a diagnosed brain clearance disorder. The CSF mobility maps of 8 CAA patients (according
to Boston criteria 2.0) and 9 HCs were used to semi-automatically segment the PVS in the centrum semiovale (CSO) and to calculate
the CSF flow within the PVS.
Results or Findings: The CAA and HC group did not differ significantly in terms of age and gender. However, CAA patients depicted
a significantly higher PVS volume in the CSO compared to HC (p < 0.01). CSF mobility correlated negatively with PVS size in both
groups (p < 0.01), which is in accordance with expectations from fluid dynamics. While absolute CSF mobility did not differ
significantly between both groups, CAA patients revealed a significantly lower decrease in CSF mobility with increasing PVS size
compared to HC (p < 0.05).
Conclusion: This study is the first providing proof of concept for clinical usage of brain clearance imaging with non-invasive CSF
mobility mapping in humans and revealed altered brain clearance in enlarged PVS of CAA patients.
Limitations: CSF mobility measurements do not allow to draw conclusions on net CSF flow, flow direction or overall clearance
capacity.
Funding for this study: This work was supported by the EU Joint Programme for Neurodegenerative Disease Research (JPND) and
the Fondation Leducq (Transatlantic Network of Excellence 23CVD03).
Ethics committee - additional information: Number 437/21
DTI-ALPS sequences for the study of the glymphatic system in idiopathic normal-pressure hydrocephalus (7 min)
716
Author Block: F. M. Doniselli, A. Gans, M. A. Broggi, V. Redaelli, M. Moscatelli, M. Verri, D. Aquino, R. Pascuzzo, M. Grisoli; Milan/IT
Purpose: Idiopathic normal-pressure hydrocephalus (iNPH) is a neurodegenerative disorder present in more than 5% of the
population over 80 years of age, and the treatment of choice is ventriculoperitoneal (VP) shunt surgery. Assessment of high-volume
lumbar tap test (LTT) has sufficiently high positive predictive values to predict VP response. Radiologic signs of iNPH (Evans index,
transcallosal angle, and disproportionately dilated subarachnoid space (DESH)) provide diagnostic morphologic features but are not
predictive of response to surgery. More recent etiologic hypotheses of iNPH include a dysfunction of the glymphatic system (GS). DTI
imaging targeting perivascular spaces (DTI-ALPS) has shown the presence of reduced perivascular diffusivity in iNPH patients, an
indirect index of GS dysfunction. The purpose of our work is to evaluate DTI-ALPS as a predictor of VP response.
Methods or Background: Between 2021 and 2023, 89 patients diagnosed with probable iNPH underwent LTT testing. Of these, 22
were enrolled in the study with pre- and post-LTT DTI-ALPS sequences. Fifteen patients responded positively to LTT test and 14 of
them underwent VP. Among the 14 patients who underwent VP, 9 underwent 3-month MR follow-up. 7 healthy patients were included
as a control group.
Results or Findings: 68% (15/22) of patients responded positively to LTT; of these 93% (14/15) underwent VP surgery with clinical
benefit in 86% (12/14). The ALPS index in controls was higher than in iNPH patients (P <0.01). The ALPS index pre-LTT (T0) was
significantly higher in LTT-responsive patients (P <0.01). The ALPS index (T0) correlated with clinical indices (P <0.01), and these
were significantly improved at T3 (P <0.05).
Conclusion: ALPS index provides a noninvasive, reproducible, and reliable diagnostic and prognostic tool in the evaluation of the
iNPH patient that can significantly improve surgical selection.
Limitations: There was only a small cohort of patients and this was a single-centre study.
Funding for this study: This study received no funding.
Ethics committee - additional information: This study was approved by the internal ethics committee.
The exploration of glymphatic system alteration in narcolepsy type 1 using the DTI-ALPS index (7 min)
Peng xin Hu; Nanchang / China
Author Block: P. x. Hu, X. P. Tang, Y. Zou; Nanchang/CN

Purpose: This study aimed to evaluate glymphatic system function alteration in patients with narcolepsy type 1 (NT1) compared with
healthy controls using diffusion tensor imaging (DTI) with the perivascular space (DTI-ALPS) index.
Methods or Background: The study enrolled patients with NT1 who diagnosed through polysomnography (PSG) and genetic
evaluation and healthy controls. All participants underwent brain scans with diffusion tensor imaging (DTI) using the same 3T MRI
scanner. We calculated the DTI-ALPS index to assess glymphatic function and compared the DTI-ALPS index of patients with NT1 and
healthy controls. Additionally, we conducted a correlation analysis between the DTI-ALPS index and clinical characteristics.
Results or Findings: We enrolled a total of 27 NT1 patients and 27 healthy controls in the study. The mean ALPS index of the
patients with NT1 was significantly lower than that of the healthy controls (1.464 vs 1.603, P = 0.023). In patients with NT1, DTI-ALPS
showed a positive connection with Wake after Sleep Onset (WASO) (r = 0.510, P = 0.018) and N3 Sleep Percentage (r = 0.641, P =
0.007).
Conclusion: Patients with NT1 had a lower DTI-ALPS index compared to the healthy controls might indicate glymphatic system
dysfunction in patients with NT1. The results also show that glymphatic system index is negatively correlated with WASO and N3
sleep percentage in patients with NT1. This demonstrates that a decline in glymphatic function in patients with NT1 results in
increased WASO and percentage of N3 sleep. The ALPS index may be a potential biomarker for detecting the function of the
glymphatic system in patients with NT1 and the severity of the disease in NT1 patients.
Glymphatic system dysfunction in middle-aged and elderly chronic insomnia patients: relation to cognition (7 min)
Yu Jin; Chengdu / China
717
Author Block: Y. Jin, X. Zhang, G. Chen, X. Ding; Chengdu/CN
Purpose: Chronic insomnia has been associated with risk of cognitive impairment and dementia in the elderly. The glymphatic
system may have a bridge role in the relationship. Using diffusion-tensor imaging analysis along the perivascular space (DTI-ALPS),
we aimed to explore whether glymphatic function is impaired in middle-aged and elderly chronic insomnia patients, and to identify
the relationship between glymphatic dysfunction and cognitive impairment.
Methods or Background: A total of 33 chronic insomnia patients (57.36±5.44 years, 30 females) and 20 age- and sex-matched
healthy controls (57.95±5.78 years, 16 females) were prospectively enrolled between May 2022 and January 2023. All the
participants completed 3.0-T brain MRI scan, cognition and sleep assessments at the time of participation. The DTI-ALPS index was
calculated according to the DTI images.
Results or Findings: Our findings revealed that the DTI-ALPS index was significantly lower in both the chronic insomnia patients
with impaired cognition group (1.32±0.14) and normal cognition group (1.46±0.09) compared to controls (1.61±0.16) (P <0.0001, P
=0.0008, respectively). MMSE scores of chronic insomnia patients with cognitive impairment were positively correlated with the DTI-
ALPS index (Partial correlation analyses after correction for age: r = 0.74, p = 0.001). DTI-ALPS index had moderate accuracy (AUC:
0.81) and high sensitivity (82.4%), but low specificity (68.7%) in distinguishing impaired cognition from normal cognition in chronic
insomnia patients.
Conclusion: Our study provides evidence of glymphatic system dysfunction in middle-aged and elderly chronic insomnia patients. In
addition, glymphatic system dysfunction is well correlated with cognitive decline. Future studies are needed to explore the diagnostic
and therapeutic value of glymphatic function in individuals with cognitive impairment in chronic insomnia.
Limitations: The sample size was relatively small and restricted to patients from a single hospital.
Ethics committee - additional information: This study was approved by the ethics Committee of Chengdu Second People's
Hospital.
Intraindividual comparison between gadopiclenol (0.05 mmol/kg)- and gadobutrol (0.1 mmol/kg)-enhanced MRI in
terms of brain metastases visualisation (7 min)
Lisa Maria Farina; Pavia / Italy
Author Block: A. Pichiecchio1, L. Loevner2, G. Hutóczki3, K. Dziadziuszko4, C. Groden5, C. Oppenheim6, L. M. Farina7; 1Borgarello/IT,
2
Philadelphia, PA/US, 3Debrecen/HU, 4Gdańsk/PL, 5Mannheim/DE, 6Paris/FR, 7Pavia/IT
Purpose: Gadopiclenol (Elucirem™, Guerbet) is a high relaxivity macrocyclic gadolinium-based contrast agent (GBCA), approved by
the FDA and currently under review by the EMA. This study aimed to compare contrast-enhanced MRI with gadopiclenol at 0.05
mmol/kg and gadobutrol at 0.1 mmol/kg in terms of visualisation of brain metastases.
Methods or Background: This is a post hoc analysis of the phase III PICTURE (gadoPIClenol for cenTral nervoUs system magnetic
REsonance) study. A subpopulation of patients with brain metastases (N=46) who underwent two separate MRIs with gadopiclenol
and gadobutrol was analysed. Lesion visualisation parameters (border delineation, internal morphology and contrast enhancement)
were assessed by three off-site blinded readers. Percentage of enhancement (E%), lesion to background ratio (LBR) and contrast to
noise ratio (CNR) were measured. Overall diagnostic preference was assessed in a global matched pairs fashion by three additional
blinded readers.
Results or Findings: For all readers, and all visualisation parameters, the difference in mean of scores showed the non-inferiority of
gadopiclenol to gadobutrol (lower limit of 95% CI between -0.04 and -0.30 depending on the reader, above the non-inferiority margin
[‑0.35]).
There was no significant difference between the two GBCAs in terms of CNR, while a higher E% was observed with gadopiclenol for
two of three readers (P ≤0.0097), and a higher LBR for the three readers (P ≤0.0036). Readers preferred images with gadopiclenol in
54% to 59% of evaluations, reported no preference for 13% to 24% of evaluations, and preferred images with gadobutrol in 22% to
30% of evaluations.
Conclusion: MRI with gadopiclenol at 0.05 mmol/kg is non-inferior to gadobutrol at 0.1 mmol/kg for brain metastases visualisation.
Limitations: This is a post-hoc analysis with a limited number of patients.
Funding for this study: This study was sponsored by Guerbet.
Ethics committee - additional information: As a post-hoc analysis, this study did not require ethics committee approval.
Quantification and assessment of the chemical form of residual gadolinium in the skin after repeated administration of
gadolinium-based contrast agents in rats (7 min)
Axel Treu; Wuppertal / Germany
718
Author Block: A. Treu1, J. Boyken2, J. Lohrke2, G. Jost2, U. Thuss1, H. Pietsch2; 1Wuppertal/DE, 2Berlin/DE
Purpose: Gadolinium presence in the body has triggered intense research in recent years. Most preclinical studies determined the
overall Gd concentrations, this study investigated the distribution and chemical form of the applied GBCAs in the skin using matrix-
assisted laser desorption/ionisation mass spectrometry imaging (MALDI MSI).
Methods or Background: Rats received either a single (1x0.6 mmol/kg) or multiple intravenous injections within 2 weeks (8x0.6
mmol/kg) of gadobutrol or saline. The Gd distribution and concentration in skin after multiple injections were measured after 5 days
and 5 weeks by inductively coupled plasma-mass spectrometry (ICP-MS) and laser ablation-ICP-MS (LA-ICP-MS). The intact gadobutrol
in the skin (MW 604.72 g/mol) was determined with MALDI MSI 5 days (multiple dose) and 1 week post injection (single dose).
Neodymium- or holmium-butrol was used as internal standard for quantification in MALDI MSI.
Results or Findings: Total gadolinium skin concentration measured by ICP-MS at 5 days after 8x0.6 mmol/kg for gadobutrol was
2.7±1.1 nmol/g. Importantly, gadobutrol was detected within the dermis as intact Gd-chelate using MALDI MSI. There were no
indications for dechelation, as the Gd quantification and distribution by LA-ICP-MS was very similar with the intact gadobutrol detected
by MALDI MSI. Intact gadobutrol was predominantly detected in the sweat glands with an average of 4.1±1.1 compared to 5.0±1.2
nmol/g total gadolinium measured by LA-ICP-MS.
Conclusion: MALDI MSI showed the chemical form of residual gadolinium being intact Gd-chelate. No indications for dechelation of
free gadolinium were observed. Gadobutrol levels in the skin measured as total gadolinium via LA-ICP-MS were low and almost
completely eliminated within 5 weeks. After 5 days, intact Gd-chelate was localised in sweat glands, suggesting a potential
elimination route via sweat.
Limitations: There are species differences in terms of skin physiology between rats and humans.
Funding for this study: All authors were Bayer AG employees.
Ethics committee - additional information: LaGeSo Berlin approved this study (approval for in vivo studies).
Potential of heavily T2-weighted fluid-attenuated inversion recovery sequence for detection of contrast agent (7 min)
Lars-Patrick Schmill; Kiel / Germany
Author Block: L-P. Schmill, S. Seehafer, S. Aludin, N. Larsen, O. Jansen; Kiel/DE

Purpose: After angiographic thrombectomy in acute stroke with partial cerebral infarction, gadolinium-based contrast agents (GBCA)
are rarely detected in the cerebrospinal fluid (CSF) on T1-weighted sequences. Such disruptions of the blood-CSF and brain-CSF
barriers would in principle also be expected in inflammatory or tumour diseases, although they have not yet been observed and
further studied. In order to make the detection of gadolinium as sensitive as possible and to be able to detect a deviation from the
appropriate inversion time of the CSF with only small amounts of gadolinium, we have modified the fluid-attenuated inversion
recovery (FLAIR) technique.
Methods or Background: A MRI phantom was prepared using a series of dilutions of GBCA in isotonic saline (1:250; 1:500; 1:1,000;
1:2,000; 1:4,000; 1:8,000; 1:16,000; 1:32,000; 1:64,000; 1:128,000; 1:256,000) and native controls of pure isotonic saline. Around the
test tubes, the phantom was filled with agarose gel with the addition of MnCl2 (0.08 mmol/l) to match MRI tissue properties of the
brain parenchyma. A 3T MRI (Magnetom, Vida, Germany) was subsequently used to test the FLAIR sequence with different parameters
of TR, TE and TI in order to obtain the most complete suppression of isotonic saline and agarose gel while allowing high sensitivity for
the contrast agent contained.
Results or Findings: Optimal detection of gadolinium at a dilution of 1:256,000 was possible using a FLAIR with a TR of 9,000 ms, a
TI of 1,800 ms and a TE of 650 ms. The signal intensity ratio of sample (44 SI) to isotonic saline (17 SI) was 2.59.
Conclusion: The highly T2-weighted FLAIR sequence is excellent for the detection of subtle amounts of contrast agent in the CSF.
Limitations: This study lacked translation to clinical examinations.
Ethics committee - additional information: The study was approved by the institutional review board and written informed
consent was obtained from all participants.
Assessment of pharmacokinetics and safety of gadoquatrane in renally impaired patients and its dialysability using
clinical trial data, modeling and simulation approaches and in-vitro data (7 min)
Birte Maria Hofmann; Berlin / Germany
719
Author Block: B. M. Hofmann1, G. Sutter1, T. Fadini1, S. Klein1, E. Vendel2, P. Vis2, S. Heitmeier3, T. Frenzel1, W. Ebert1; 1Berlin/DE,
2
Leiden/NL, 3Wuppertal/DE
Purpose: The aim of this study was to evaluate the pharmacokinetics (PK) and safety of the novel, tetrameric macrocyclic
gadolinium-based contrast agent (GBCA) gadoquatrane, using clinical trial data with modeling & simulation (M&S) approaches and in
vitro testing for dialysability assessment, leveraging the broad knowledge of the class of GBCAs avoiding exposure of vulnerable
patient populations, i.e. with severe renal impairment or ESRD, in clinical trials.
Methods or Background: An open-label, single-dose phase 1 study investigated PK (plasma PK, excretion until 7 days p.i. and up to
6 months p.i.) and safety in adult participants (8 per group, dose: 0.025mmol/kg, i.e. 0.1mmol Gd/kg): 1) Mildly impaired renal
function (RF) (eGFR: 60-89mL/min/1.73m2), 2) Moderately impaired RF (eGFR: 30-59mL/min/1.73m2) 3) matched controls (normal RF:
eGFR: ≥90mL/min/1.73m2). PK in patients with severely impaired RF (eGFR<30mL/min/1.73m2) was simulated using an established
popPK model. Dialysability of gadoquatrane was investigated in human blood using a clinical dialysis system versus gadobutrol.
Results or Findings: Similar Cmax and volume of distribution (Vss/BW) independent of renal function were observed. Systemic
exposure (AUC) increased and CL/BW decreased with increasing renal impairment. Within the 7 days collection interval, gadoquatrane
excretion was essentially complete in all groups. Late measurements with a highly sensitive LLOQ confirmed continuous excretion of
trace amounts. Gadoquatrane was safe and well tolerated in all groups. Simulated plasma PK in severely impaired patients showed
expected further increase of AUC and decrease of CL/BW according to renal function. In vitro investigations confirmed dialysability of
gadoquatrane as known for gadobutrol.
Conclusion: Gadoquatrane displays expected PK and safety in mild and moderate renal impairment versus matched controls.
Simulation of PK in severe renal impairment and in vitro testing of dialysability allowed avoiding exposure of vulnerable patient
populations in clinical trials.
Limitations: The number of participants with RF in the clinical study was limited.
Funding for this study: This study received funding from the Bayer AG.
Ethics committee - additional information: The study protocol was reviewed and approved by each of the two study site's IEC/IRB
before the start of the study.
720
RC 1711 - Where neuro meets head and neck: from border zones to common ground
Categories: Head and Neck, Neuro

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Natalia Shor; Paris / France

Natalia Shor; Paris / France
Skull base: anatomy and routes of disease spread (15 min)

1. To recognise the boundaries between the anterior, central and posterior skull base.
2. To describe the relationships between the suprahyoid neck spaces and the skull base.
3. To name and identify the main skull base neurovascular foramen.
4. To predict the potential routes of spread of suprahyoid neck lesions to the intracranial compartment.
Orbit: a window to (systemic) disease (15 min)

Augustin Lecler; Paris / France
1. To develop critical thinking skills to recognise and differentiate between various orbital manifestations related to systemic diseases.
2. To identify common orbital signs and symptoms that may indicate an underlying systemic disease.
3. To evaluate the diagnostic modalities and techniques used to investigate orbital involvement in systemic diseases.
Temporal bone: anatomy and variants the ENTG surgeon wants to know (15 min)
1. To name and identify bony, neural and vascular structures at risk for inadvertent injury during temporal bone surgery.
2. To list anatomical variants of the mastoid, facial nerve, sigmoid sinus and jugular bulb.
3. To describe surgical complications related to anatomical variants.
Panel discussion: How not to miss border zone lesions? (10 min)
721
RPS 1713 - Optimisation and its tools
Categories: EuroSafe Imaging/Radiation Protection, Paediatric, Physics in Medical Imaging, Research

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Agnieszka Kuchcinska; Warsaw / Poland
CT acquisition parameter selection in the real world: impacts on paediatric radiation dose and variation amongst 155
institutions (7 min)
Yifei Wang; Sunnyvale / United States
Author Block: Y. Wang, P. Chu, C. Stewart, R. Smith-Bindman; San Francisco, CA/US

Purpose: Computed tomography (CT) exams have expanded rapidly since introduction, such that theoretical understandings of best
dose optimisation practices and real-world applications are incongruent. This remains the case even in children, where the limitation
of dose is even more important. We seek to quantify how often parameters are adjusted in real-world practice and their degree of
contribution to real-world dose distribution. We identify discrepancies between parameters that are impactful in theory and impactful
in practice.
Methods or Background: This study analyses 25,000 consecutive children routine abdomen exams performed between November
2015 and Jan 2021 in the UCSF International CT Dose Registry of 155 institutions. We calculated geometric standard deviation (gSD)
for five parameters (kV, mAs, spiral pitch, number of phases, scan length) to assess variation in practice. A Gaussian mixed regression
model was performed to predict the radiation dose-length product (DLP) using the parameters. To reflect the theoretical impact, we
predict the increase in DLP per 10% increase in the parameter. To reflect the real-world practical impact, we predict the increase in
DLP per gSD increase in the parameter.
Results or Findings: Among studied examinations, mAs, number of phases, and scan length were frequently manipulated (gSD
1.47–1.79); kV was rarely manipulated (gSD 1.12). Theoretically, kV is the most impactful parameter (31% increase in DLP per 10%
increase in kV, 8–11% for other parameters). In real-world practice, kV is less impactful; for each gSD increase in kV, the DLP
increases by 40%. This value is highly surpassed by the analogous number in mAs (68%), while being comparable to other technical
parameters.
Conclusion: Despite the potential impact of kV on radiation dose, this parameter is rarely manipulated in common practice and its
potential remains untapped.
Limitations: This study uses mainly US Data.
Funding for this study: The registry was funded by NIH and PCORI.
Ethics committee - additional information: The institutional review board of the primary responsible university approved this
research, and collaborating institutions obtained local approval or relied on the primary site's approval.
Establishing indication-based diagnostic reference levels for paediatric computed tomography and international
comparison: data from an international registry (7 min)
Denise Bos; Essen / Germany
722
Author Block: D. Bos1, Y. Wang2, C. Stewart2, S. Zensen1, J. Luong2, P. Chu2, R. Smith-Bindman2; 1Essen/DE, 2San Francisco, CA/US
Purpose: Computed tomography (CT) scans are essential for diagnosing paediatric patients, especially in emergencies. However,
concerns arise about the potential carcinogenic risk associated with radiation exposure. CT radiation doses can vary widely depending
on medical indications, protocols, and local practice. Our aim is to establish diagnostic reference levels (DRLs) based on clinical
indications in paediatric patients.
Methods or Background: A comprehensive analysis was performed using CT data from an international dose registry that includes
CT scans performed in children under 18 years at more than 150 sites between January 2016 and January 2021. DRLs, defined as the
75th percentile, were calculated for volume-weighted CT dose index (CTDIvol) and dose-length product (DLP) across 15 broad CT
categories, which reflect both the anatomic areas and radiation doses (low-dose [LD], routine-dose [RD], high-dose [HD]) required by
the underlying imaging indications. We compared RD categories for head, chest, and abdomen/pelvis scans between facilities in the
United States (US) and Europe.
Results or Findings: A total of 95,047 CT scans (54% male, 45% female) from 41 different indications were included in the analysis.
The DRLs of DLP and CTDIvol mainly increased significantly with increasing age group (p<0.05). For head scans in 10- to 14-year-old
children, the DRLs for DLP varied from 362 mGy·cm (LD) to 734 mGy·cm (RD) to 2,058 mGy·cm (HD). US DRLs were significantly
higher than European DRLs for RD chest and abdomen/pelvis. Differences were inconsistent for RD head.
Conclusion: Optimising radiation dose for paediatric patients presents significant challenges due to variations based on age, size,
clinical indication and protocol. Therefore, we have established indication-based DRLs for different CT categories and age groups to
minimise excessive radiation doses and standardise practice.
Limitations: An identified limitation is that the registry contains mainly US data.
Funding for this study: The registry was funded by NIH and PCORI.
Ethics committee - additional information: The institutional review board of the primary responsible university approved this
research, and collaborating institutions obtained local approval or relied on the primary site’s approval.
Automated dose and noise assessment as a guide for harmonisation of abdominal CT protocols across multiple
scanners (7 min)
Janne Vignero; Herent / Belgium
Author Block: J. Vignero1, N. Fitousi1, B. Miseur1, J. Binst1, K. Lemmens1, V. Nuttens2, H. Bosmans1; 1Leuven/BE, 2Aalst/BE
Purpose: In this study a systematic approach is proposed for performance evaluation and, if necessary, harmonisation of abdominal
computed tomography (CT) protocols across vendors using dose and noise metrics.
Methods or Background: CTDIvol, water equivalent diameter (WED), global noise level (GNL) and metadata of 4,933 abdominal CT
exams from six scanners were calculated by DOSE (Qaelum). Paediatric patients (<15 years), hard kernel reconstructions and scans
with substantial image truncation were excluded. Patients were categorised into five size groups based on WED. Subgroups were
defined based on protocol, scanner, kVp and kernel. Local working points (LWP) for CTDIvol and GNL were calculated from the median
values for every subgroup. Reference working points (RWPs) were determined for each size group based on the median CTDIvol and
GNL across all subgroups. The performance of each subgroup was assessed by comparing LWP to RWP.
Results or Findings: Patient size groups ranged between 200 mm and 422 mm, with corresponding increases in dose and noise
RWP from 5 to 16 mGy and 10.9 to 12.5 HU, respectively. Variations in LWP among subgroups were large. The GNL with iterative
reconstruction was significantly (p=0.0) lower than using filtered back projection. Some subgroups appeared to be overdosing or
requiring higher image quality, while others might benefit from higher radiation doses. One scanner with substantially higher dose
and noise LWP than RWP was identified as underperforming. Suggested actions were confirmed in a team comprising of radiologists, a
radiographer and a physicist.
Conclusion: Combined evaluation of automated image noise evaluation, radiation dose and patient size points to clinically relevant
adjustments of the scan protocols and was successfully applied in a multi-vendor CT scanner scenario.
Clinical CT protocol comparison: an Italian multicentre study (7 min)

Raffaele Villa; Monza / Italy
723
Author Block: R. Villa1, M. Daniotti1, C. Ingraito1, A. Italian Digital Radiology Working Group2, N. Paruccini1; 1Monza/IT, 2Milan/IT
Purpose: In CT imaging, differences in manufacturers, models, scanning parameters, reconstruction algorithms, clinical tasks, and
user’s experience, introduce variability in the trade-off between diagnostic information and patient’s radiation exposure. The extent to
which each of these sources of variabilities impacts the overall procedure performance is still unknown. This study aimed to compare
image quality and radiation dose in five acquisition protocols to ascertain how different technical features can affect CT performance,
to inform protocol optimisation and design.
Methods or Background: Five different acquisition protocols to investigate five different clinical tasks were investigated:
intracranial haemorrhage, pulmonary embolism, acute abdomen, paediatric chest and paediatric acute abdomen.
The Catphan 600 (The Phantom Laboratory, USA), was imaged in 34 CT scanners from 20 hospitals across Italy, for a total of 251
datasets, using the locally optimised clinical protocols. The Catphan body external-annulus was used to assess the effects of tube
current modulation for the different manufacturers.
Image quality was quantified in terms of the detectability index using an object contrast ranging 25-700 HU and lesion size ranging
5-25 mm, representative of clinical scenarios related to the examined protocols.
Results or Findings: Each investigated protocol showed differences between manufacturers in terms of dose levels and image
quality values. The reconstruction algorithm’s generation proved another driving element for image quality. Nevertheless, variability
between CTs was found, even from the same vendor and technology.
Conclusion: This study suggested that manufacturer and reconstruction algorithms play a main role in the optimal trade-off between
diagnostic information and radiation dose levels. Furthermore, the other examined CT procedure technical features, have a non-
negligible impact. All the potential sources of variability should be included when designing optimization actions in clinical CT.
Limitations: Data numerosity should be increased to improve the study strength, especially for paediatric exams.
Implementation of the German lung cancer screening protocol and image quality evaluation over three generations of
CT systems (7 min)
Mishal Ursani; Heidelberg / Germany
Author Block: M. Ursani1, T. D. Do1, H-U. Kauczor1, H-P. Schlemmer1, T. Lasser2, S. O. Schönberg3, S. Sawall1, M. Kachelrieß1;
1 2 3
Heidelberg/DE, Garching/DE, Mannheim/DE
Purpose: To develop methods for the implementation of lung cancer screening using low-dose CT according to the German
“Bundesamt für Strahlenschutz” (BfS) and to evaluate the resulting image quality in three generations of CT systems.
Methods or Background: Experiments were performed using a thorax phantom (QRM, Germany) in three generations of CT
systems (Somatom Flash/Force, Naeotom Alpha, Siemens Healthineers, Germany). The phantom and imaging protocols were tailored
to meet BfS requirements for lung cancer screening. The phantom featured 1 cm sized spherical lung nodules with a contrast of 150
HU and fat extension rings mimicking different patient sizes, namely S (20×30 cm), M (25×35 cm), and L (30×40 cm). Acquisitions
were performed at 1.3 mGy as defined as the maximum dose for a standard patient (BMI 26 kg/m²), employing patient-specific
prefilters, if available. Images were reconstructed iteratively using the smoothest kernel ensuring a spatial resolution with a FWHM≤1
mm in both axial and longitudinal direction. Hence, resolution is only approximately matched among systems. Contrast-to-noise ratio
(CNR) was used to assess the visibility of lung nodules in all images.
Results or Findings: A dose of 1.3 mGy was achieved using tube settings of 20 mAs/120 kV, 306 mAs/100 kV+Sn, and 150 mAs/100
kV+Sn for Flash, Force and Alpha. Image noise varied between 115 HU to 330 HU over all systems and phantom sizes. CNR (S to L)
ranged from 1.30 to 0.36, 1.03 to 0.65, and 1.76 to 0.71 for Flash, Force, and Alpha.
Conclusion: A wide range of CT systems are capable of providing an image quality sufficient for lung cancer screening examinations.
However, larger patients might require an increase in dose or slice thickness to account for the high image noise.
Limitations: The study is limited to phantom acquisitions.
The out-of-plane contact shield and mA-modulation: the effect of absorbed dose on the foetus (7 min)
Heli Riitta Sinikka Larjava; Turku / Finland
724
Author Block: H. R. S. Larjava, C. T. M. Eneh, A. Saikkonen, R. Parkkola; Turku/FI
Purpose: The aim of the patient out-of-plane shield is to reduce the patient radiation dose. Its effect on foetus radiation dose was
evaluated in computed tomography pulmonary angiography with the out-of-plane shield visible in the localiser but absent in the scan
range in chest CT with different CT scanners.
Methods or Background: An anthropomorphic phantom with additional prosthetic pregnancy belly and breasts was scanned with
four different CT scanners from three different vendors. The chest was first scanned without any shielding, and then with the out-of-
plane shield within the localiser but outside the imaged volume. The clinical protocols were used. Radiation dose was measured with
radiophotoluminescence dosimeters from abdominal and naval points, representing the foetal radiation dose nearest the radiation
field and average foetal dose.
Results or Findings: There were notable differences between absorbed foetal doses using the different scanners. The out-of-field
shield decreased the foetus’ absorbed radiation dose with Siemens and Canon scanners. With the GE scanner the shield increased the
dose.
Conclusion: Applying an out-of-plane shield outside the scanned volume, but visible in the localiser images, may increase foetus’
radiation dose, especially with older scanners using only localiser based mA-modulation.
Limitations: The foetus' radiation dose has been evaluated as a point dose, as the measurements have been done in one measuring
point. It represents the foetus' average dose, but it doesn't evaluate the maximum radiation dose.
Funding for this study: Funding for writing has been applied for, but has not yet been granted.
Ethics committee - additional information: The study is a phantom study, so ethical approval was not required.
Physics evaluation of a super resolution deep learning reconstruction algorithm in an anatomical background (7 min)
Kirsten Boedeker; Los Angeles / United States
Author Block: K. Boedeker, D. Shin, N. Akino, D. Mather; Otawara-shi, Tochigi/JP

Purpose: Deep learning reconstruction algorithms for computed tomography (CT) are generally trained on anatomical data.
Therefore, the purpose of this work is to characterise a super resolution deep learning reconstruction algorithm (DLR) compared to
traditional hybrid iterative reconstruction (HIR) and filtered back projection (FBP) in terms of standard quantitative noise and spatial
resolution measures, with the test objects inserted in both an anatomical background as well as in uniform phantom background.
Methods or Background: A commercial super resolution DLR was assessed for cardiac mode. The modulation transfer function
(MTF) was first measured using a standard edge-based approach on images acquired from the Catphan 600 with uniform background
and reconstructed with DLR, HIR, and FBP. The cylindrical test objects used for assessment were next simulated via forward projection
using a system model and inserted into raw patient data. The raw patient data was then reconstructed with DLR, HIR, and FBP. The
MTF and standard deviation of noise were then measured. A 100 mm FOV was used for measurement. The Pearson correlation
coefficient was determined for each reconstruction algorithm’s MTF in patient versus phantom.
Results or Findings: The FBP MTF values in the anatomical background versus uniform phantom were within 10%. In both phantom
and the anatomical background, noise reduction for DLR versus HIR was approximately 45-50%. The 10% of MTF was 11.6p/cm with
DLR versus 8.4lp/cm with HIR in anatomical background. The Pearson correlation in anatomical background versus phantom was
greater than 0.99 for all three reconstruction algorithms.
Conclusion: Traditional in-plane MTF applied in an anatomical background demonstrates significant improvement over HIR and
correlates well with phantom measurements.
Limitations: An identified limitation is that a limited dose range was analysed.
Funding for this study: Funding was received from Canon Medical System Corporation.
Influence of foreign bodies on image quality in dark-field chest radiography (7 min)

Lennard Kaster; Munich / Germany
725
Author Block: L. Kaster1, S. Karl1, T. Urban1, F. Gassert1, R. C. Schick1, T. Koehler2, A. Sauter1, F. Pfeiffer1, D. Pfeiffer1; 1Munich/DE,
2
Hamburg/DE
Purpose: Dark-field radiography is a novel x-ray method that visualises microstructural properties of the examined object. Our
pioneering studies, the first conducted on humans worldwide, have demonstrated that this method outperforms conventional
radiographs in diagnosing and staging pulmonary diseases, like COPD and Covid-19. It is known from conventional x-ray radiographs
and CT that foreign bodies, such as metal implants, can lead to artefacts and thus impair image quality. Therefore, this research
evaluates the influence of foreign bodies on x-ray dark-field chest radiographs.
Methods or Background: Experiments were conducted using a clinical dark-field chest radiography prototype, including
investigating patient images. The "LUNGMAN" thorax phantom was used and measurements were performed with and without foreign
bodies like breast implants, medical compresses and a pacemaker unit. The impact of these foreign bodies on image quality was
analysed.
Results or Findings: The results demonstrated that the impact of a foreign body on image quality is contingent upon its material
composition. Homogeneous materials, such as breast implants, do not contribute additional dark-field signals. Conversely, materials
possessing a microstructure, exemplified by medical compresses, generate their own dark-field signals. Notably, these signals do not
influence the surrounding areas.
In the case of a pacemaker unit, despite the absence of a microstructure, a dark-field signal can be observed. Furthermore, areas
adjacent to the foreign body are also affected.
For both medical compresses and the pacemaker unit, the dark-field signal emanating from tissue within the same radiation path
cannot be reliably evaluated.
Conclusion: Foreign bodies have the potential to generate a substantial true or artificial dark-field signal, which can significantly
impact the image quality and interpretability of dark-field radiographs. Therefore, it is crucial to consider these potential artefacts to
ensure an accurate diagnosis
Funding for this study: Funding was received from the Federal Ministry of Education and Research (BMBF) and the Free State of
Bavaria under the Excellence Strategy of the Federal Government and the Länder, as well as by the Technical University of Munich –
Institute for Advanced Study.
Ethics committee - additional information: The study was approved by the Ethics Commission of the Medical Faculty, Technical
University of Munich, Germany (reference number: 166/20S).
726
US 17 - Comprehensive ultrasound pathways for patient management: how to pick up the pieces from
CT and MR imaging
Categories: Education, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderators:
Adrian K. P. Lim; London / United Kingdom
Mustafa Seçil; Izmir / Turkey

Indeterminate thyroid lesion: one-stop US clinic (14 min)

1. To learn the pathways following incidental detection of a thyroid lesion on CT, MRI or PET-CT.
2. To learn the aspects of ultrasound findings for accurate diagnosis and selecting the ones for fine needle aspiration of incidental
thyroid lesions.
3. To understand the appropriate use of diagnostic methods in incidental thyroid lesions in terms of good clinical practice and cost
effectivity.
Shoulder pain assessment and management following x-ray and MR imaging (14 min)
Bhavna Batohi; London / United Kingdom
1. To understand the place of US in the diagnostic algorithm of patients with a painful shoulder.
2. To appreciate both the advantages and limitations of US in imaging the common causes of shoulder pain as compared to plain films
and MRI.
3. To understand the role of US-navigated therapeutic procedures in the management of shoulder pain.
Indeterminate CT renal lesion: cyst, complex or simple, solid or a normal variant? (14 min)
1. To understand the advantages of ultrasound in characterising focal renal lesions.

2. To appreciate how CEUS can help problem solve for CT and MRI.
3. To be able to come up with a rational diagnostic pathway utilising available tools in the imaging armamentarium to achieve a
diagnosis.
Incidental liver lesions following CT: benign or malignant (14 min)

Tommaso Vincenzo Bartolotta; Palermo / Italy
1. To understand the capabilities of CEUS in the detailed assessment of indeterminate focal liver lesions after CT.
2. To appreciate the limitations of CEUS in assessing focal liver lesions.
3. To acknowledge the correct pathways for further management of indeterminate focal liver lesion safety CEUS.
727
RPS 1704 - Novel technological developments in chest imaging: from research to practice
Categories: Chest, Contrast Media, Imaging Methods, Physics in Medical Imaging

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Andreas Christe; Bern / Switzerland
Dual-layer dual-energy CT characterisation of thrombus decomposition in acute pulmonary hypertension and chronic
thromboembolic pulmonary hypertension (7 min)
Roman Johannes Gertz; Cologne / Germany
Author Block: R. J. Gertz1, M. Pienn2, S. Lennartz1, R. P. Reimer1, K. Kaya1, J. Kottlors1, F. J. Fintelmann3, S. Rosenkranz1, A. Bunck1;
1
Cologne/DE, 2Graz/AT, 3Boston, MA/US
Purpose: The purpose of this study was to evaluate dual-layer dual-energy computed tomography (dlDECT)-based characterisation of
thrombus decomposition in acute pulmonary embolism (PE) and chronic thromboembolic pulmonary hypertension (CTEPH).
Methods or Background: In this single-centre, retrospective study, 33 patients with CTEPH and 49 patients with acute PE, all of
whom presented with central or segmental pulmonary thrombi, were included. All patients underwent CT pulmonary angiography on a
dlDECT. Conventional images (CI), as well as material (iodine density overlay (IDO)) and energy-specific (virtual monoenergetic (VMI))
images, were reconstructed. Region-of-interest (ROI)-based measurements were manually performed within the thrombus, and
morphological features indicative of acute and chronic thrombus morphology were documented. Diagnostic accuracy was determined
through AUC analysis. Intra- and interreader variability was assessed from a randomly selected subset of 20 patients (10 acute PE and
10 CTEPH).
Results or Findings: Thrombi in patients with acute PE revealed a higher attenuation both in CI (67.7 ± 17.7 vs 41.0 ± 8.4 HU) and
VMI50 reconstructions (106.4 ± 35.8 vs 58.2 ± 18.8 HU) and took up less iodine (1.34 ± 0.95 vs 0.58 ± 0.45 mg/ml) (p for all <
0.001). Thrombus attenuation on CI and VMI50 best enabled for identification of CTEPH (CI: AUC 0.92, 95%-CI: 0.86 – 0.98; VMI50:
0.91, 95%-CI: 0.85 – 0.97), yielding higher diagnostic performance than the diameter of the truncus pulmonalis (AUC 0.86, 95%-CI:
0.78 – 0.94). The Bland-Altman test demonstrated agreement among readers.
Conclusion: Thrombus attenuation in large pulmonary thrombi is associated with thrombus age. CI and VMI50keV. derived from
dlDECT yield considerable diagnostic accuracy and could potentially aid in differentiating between acute PE and CTEPH.
Limitations: This study lacked a diagnostic standard comparison and exhibited a selection bias as patients with a history or imaging
features of subacute PE were excluded.
Funding for this study: Supported by the Cologne Clinician Scientist Program (CCSP)/ Faculty of Medicine/ University of Cologne.
Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) (Project No. 413543196).
Ethics committee - additional information: This study was approved by the ethics committee of the Faculty of Medicine of the
University of Cologne, Cologne, Germany.
Assessment of lung perfusion using dynamic digital radiography and comparison with nuclear medicine lung
scintigraphy (7 min)
Satinder Singh; Birmingham / United States
728
Author Block: S. Singh, R. Sai, P. Manapragada, D. Benson; Birmingham, AL/US
Purpose: Assessment of lung perfusion is required in the preoperative evaluation of patients being considered for lung transplant or
resection. Dynamic digital radiography (DDR) is a newer radiographic technique that acquires rapid sequential diagnostic radiographs
of the chest throughout multiple respiratory cycles. The regional lung perfusion can be estimated using DDR by assessing pixel signal
changes throughout the cardiac cycle. This study will compare the results of the differential lung perfusion estimated using DDR with
the same results obtained from lung scintigraphy, a conventional imaging modality.
Methods or Background: A retrospective review of patients evaluated with both nuclear medicine lung scintigraphy and DDR was
performed. The DDR examinations were performed between January 14, 2022, and July 27, 2023. Each patient had a lung
scintigraphy perfusion study within six months of the DDR examination. The percent differential of perfusion between the lungs was
calculated using both modalities and the results were analyzed for statistical correlation.
Results or Findings: Results for 75 patients were reviewed (mean age - 56 years, 32 Females). The mean absolute percent
differential in perfusion between the right and left lungs was (14.7 ± 23.1%) using DDR and (15.08 ± 24.76%) using lung
scintigraphy. There were five patients in which there was a discrepancy between the two modalities in determining which lung was
better perfused. The Pearson coefficient of correlation was (0.915) indicating a strong correlation between the perfusion results
obtained by the two modalities.
Conclusion: Differential lung perfusion estimated by DDR is strongly correlated with the same result obtained using lung
scintigraphy. The speed and cost-effectiveness of DDR make it an attractive option for clinicians, potentially reducing wait times and
healthcare costs for patients.
Limitations: Small numbers. Need more refined computer analysis.
Ethics committee - additional information: The study is IRB approved.
Feasibility of artificial intelligence in detecting pulmonary nodules using virtual non-contrast images from dual-layer
spectral-detector CT (7 min)
Ze Lin; Nanchang / China
Author Block: Z. Lin1, Y. Liu1, Y. H. Yang1, Y. Wang2; 1Nanchang/CN, 2Shanghai/CN

Purpose: The aim of this study was to investigate the feasibility of artificial intelligence (AI) in detecting pulmonary nodules within
chest virtual non-contrast images derived from arterial (VNC-A) and venous (VNC-V) phases of dual-layer spectral-detector CT (DLCT)
comparing them with true non⁃contrast images (TNC).
Methods or Background: 55 patients who underwent DLCT chest scans were retrospectively included in this study. We assessed
116 pulmonary nodules, establishing a gold standard for nodules ≥6 mm using radiologist visual assessment and AI labelling
software. The reconstructed VNC-A, VNC-V and TNC images were imported into the AI pulmonary nodule detection system. We
analysed the AI system's detection performance by sensitivity, precision, and false positive rate (defined as the number of false
positive nodules/the number of enrolled patients), and compared nodule long diameters among three image groups. The statistical
analysis was performed using the chi-square test or Fisher's exact test. The radiation dose for each scan period was recorded.
Results or Findings: The sensitivity of AI in detecting pulmonary nodules showed no significant difference among TNC, VNC-A, and
VNC-V (P=0.345). There were no significant differences in precision and false-positive rate between TNC and VNC-V (P >0.05).
However, using VNC-A led to a significant decrease in precision (P =0.007) and an increase in the false-positive rate (P =0.002).
Bland-Altman analysis yielded a mean difference in nodule long diameter between TNC and VNC-A was 0.151mm (95% CI: -1.264,
1.567mm), and between TNC and VNC-V were 0.057mm (95% CI: -1.391, 1.506mm). Using VNC substitute for TNC can reduce total
effective radiation dose by 31.65%.
Conclusion: Based on AI combined with VNC, the venous phase VNC is recommended to provide excellent nodule detection
performance while significantly reducing the radiation dose.
Limitations: AI couldn't accurately identify nodule boundaries, so CT value evaluation was not performed.
Silver x-ray spectrum modulation filter for lung cancer screening: comparison of capability for radiation dose reduction
and detection performance with copper filter (7 min)
Yoshiharu Ohno; Toyoake / Japan
729
Author Block: Y. Ohno1, H. KImata2, Y. Ito2, K. Fujii2, N. Akino2, H. Nagata1, M. Nomura1, T. Ueda1, Y. Ozawa1; 1Toyoake/JP, 2Otawara/JP
Purpose: The purpose of this study was to compare the capability of low-dose lung cancer CT screening between silver (Ag) and
copper (Cu) x-ray spectral modulation filters at in vitro and in vivo studies.
Methods or Background: A chest CT phantom including simulated ground-glass and part-solid nodules was scanned with a 320-
detector row CT with Ag and Cu filters at 0.6, 1.6 and 2.5 mGy at in vitro study. Then, 95 patients underwent low-dose CT lung cancer
screening with Ag and Cu filters at the same radiation dose level (i.e. 1.6 mGy) as in the in vivo study. At in vitro study, SNR at each
nodule were determined by region of interest (ROI) measurements at all protocols. At in vivo study, SNR of normal lung parenchyma
was also determined by ROI measurement at each CT. Then, the probability of a nodule was assessed with 5-point visual score by two
board-certified chest radiologists at both studies. Student's t-test was performed to compare SNR between Ag and Cu filters at each
study. ROC and JAFROC analyses were performed to compare nodule detection capability between Ag and Cu filters.
Results or Findings: At in vitro study, SNR and AUC of Ag filter were significantly better than those of Cu filter at each dose level (p
<0.05). At in vivo study, SNR of Ag filter was significantly higher than that of Cu filter (p <0.05), although no significant differences of
averaged figure of merit were determined.
Conclusion: Low-dose CT screening with Ag filter can significantly improve image quality compared to screening using a Cu filter.
Limitations: This was a single institution study.
Funding for this study: This study received funding via a research grant from Canon Medical Systems Corporation.
Ethics committee - additional information: IRB approval was received (Fujita Health University Hospital).
Ultra-low-dose photon-counting detector CT for imaging follow-up after lung transplantation (7 min)
Ruxandra-Iulia Milos; Vienna / Austria
Author Block: L. Lechner, R-I. Milos, A. Korajac, D. Kifjak, D. Tamandl, M. Stuempflen, M. Watzenböck, A. E. Strassl, H. Prosch;
Vienna/AT
Purpose: CT surveillance is performed in patients after lung transplantation for the detection of subclinical abnormalities. We
assessed the potential of ultra-low-dose (ULD) photon-counting detector computed tomography (PCD-CT) when compared to routinely
performed low-dose (LD) PCD-CT for the follow-up in lung transplanted patients.
Methods or Background: For this prospective study adult patients underwent same-day ULD and LD PCD-CT (with inspiration and
expiration scans for each radiation dose protocol) between March and May 2023. The 1 mm reconstructions ULD and LD scans were
reviewed by three readers who independently assessed the subjective image quality of anatomical structures (using visual grading
analysis, VGA, scores) and the presence of lung abnormalities (evaluated as present or absent). Interrater agreement was calculated
using Fleiss' kappa.
Results or Findings: 83 patients (median age 62 years, 36 women) were included. The radiation dose of the ULD protocol
comprised on average 15% of the one for LD protocol. For anatomical structures all readers rated ULD scans adequate to excellent in
the majority of patients (92.8 - 97.6% for bronchial wall, 91.6 - 94% for fissures, and 80.7 - 100% for peripheral vessels). Interrater
agreement was better on the LD protocol for the detection of ground glass opacities, consolidations, and atherosclerosis (0.61 vs
0.50; 0.68 vs 0.6; 0.94 vs 0.72), and better for the ULD protocol for the detection of air trapping, tree-in-bud, bronchial wall thickening
and nodules greater than 5 mm (0.66 vs 0.57; 0.69 vs 0.56; 0.44 vs 0.31; 0.45 vs 0.19). Both LD and ULD protocols showed
comparable agreement for the detection of reticulations, linear atelectasis and distal bronchiectasis (between 0.18 and 0.37).
Conclusion: ULD PCD-CT offers sufficient image quality and is a feasible option for post-lung transplantation patient surveillance.
Ethics committee - additional information: The ethics committee notification can be found under the number EK Nr: 1891/2022.
Low-dose high-resolution photon-counting CT in patients with cycstic fibrosis: radiation dose and image quality in the
clinical routine (7 min)
Marko Frings; Essen / Germany
730
Author Block: M. Frings, M. Welsner, S. Zensen, J. Haubold, D. Bos, L. Umutlu, B. M. Schaarschmidt, M. Forsting, M. Opitz; Essen/DE
Purpose: Patients with cystic fibrosis routinely undergo low-dose high-resolution (LD-HR) CT scans as part of their clinical
management, primarily for monitoring chronic exacerbations or assessing treatment response with the triple-combination CFTR
modulator therapy elexacaftor/tezacaftor/ivacaftor. The aim of this study is to compare the radiation dose and image quality of LD-HR
protocols between photon-counting CT (PCCT) and CT scanner with an energy-integrating detector system (EID-CT) in this patient
cohort.
Methods or Background: Thirty-five patients who underwent a LD-HR chest CT scan with PCCT and had previously undergone a LD-
HR chest CT with a standard EID-CT scanner were retrospectively included in this study. The dose-length product (DLP), volumetric
CT-dose index (CTDIvol) and singal-to-noise ratio (SNR) were determined and explorative data analysis was performed. The overall
image quality, image sharpness, and image noise of both full lung scans and enlarged sections highlighting specific findings were
assessed by three radiologists using a 5-point Likert scale.
Results or Findings: The PCCT used significantly less radiation dose compared to the EID-CT (P <0.001), approximately 36%. PCCT
consistently received significantly better ratings for overall image quality and image sharpness compared to EID CT (P <0.003).
Additionally, PCCT exhibited significantly lower image noise (P <0.004) and a lower average signal-to-noise ratio (SNR) compared to
EID CT (P <0.005).
Conclusion: Lung PCCT scans in patients with CF offer superior image quality while utilising substantially lower radiation doses in
comparison to EID-CT scans.
Limitations: The limitations of this study are its retrospective and single-centre study design.
Optimal slice thickness and quantum iterative reconstruction level in low-dose ultra-high-resolution photon-counting
detector CT of the lungs (7 min)
Dirk Graafen; Mainz / Germany
Author Block: D. Graafen1, M. C. Halfmann1, T. S. Emrich1, Y. Yang1, C. Düber1, R. Klöckner2, L. Müller1, T. Jorg1; 1Mainz/DE, 2Lübeck/DE
Purpose: Using photon-counting detector computed tomography (PCD-CT) alongside an innovative iterative reconstruction technique
called quantum iterative reconstruction (QIR) holds the promise of significantly improving the quality of lung images. This research
aimed to investigate how various QIR levels and different slice thicknesses affect the image quality of low-dose ultra-high-resolution
(UHR) PCD-CT in lung examinations.
Methods or Background: The study involved 51 patients who underwent unenhanced UHR-PCD-CT scans. The images were
reconstructed using a 1024 matrix. Three slice thicknesses (1.0 mm, 0.4 mm, and 0.2 mm) were employed, along with three different
QIR levels (2 to 4). Noise levels were calculated for all reconstructions. To evaluate the image quality in comparison to the clinical
reference reconstruction (1.0 mm QIR-3), three radiologists rated the sharpness of different lung structures and the conspicuity of
various lung abnormalities in the images using a 5-point Likert scale.
Results or Findings: The highest QIR level (QIR-4) reduced image noise effectively and received the best image quality ratings.
Decreasing the slice thickness to 0.4 mm enhanced the sharpness of pulmonary structures and the conspicuity of different
pathologies. However, when the thickness was reduced to 0.2 mm, the images were significantly affected by noise, resulting in a
notable decline in quality scores.
Conclusion: The ideal reconstruction approach for low-dose UHR-PCD-CT lung imaging is to use a slice thickness of 0.4 mm together
with the highest QIR level. The clinical implementation of this optimised UHR-PCD-CT protocol could potentially enhance the accuracy
of diagnoses and boost confidence in lung imaging.
Limitations: Limitations of the study are the single-centre design and the small sample size of some lung pathologies.
Ethics committee - additional information: The study was approved by the Institutional Ethics Committee of Rhineland-Palatinate
(Reg. Nr. 2022-16359).
The feasibility and accuracy of using low- and ultra-low-dose photon counting detector CT to detect metastatic lung
nodules in paediatric and young adults with malignant bone tumours (7 min)
Shanshui Zhou; Shanghai / China
731
Author Block: S. Zhou, L. Qin, Z. Xu, F. Yan; Shanghai/CN
Purpose: We aimed to investigate the feasibility of low- and ultra-low-dose photon counting detector CT (PCD-CT) for chest CT follow-
ups in paediatric and young adults with pulmonary metastasis of bone malignant tumours, and compare the accuracy of PCD-CT with
intra-patient previous energy-integrating detector CT (EID-CT).
Methods or Background: 83 subjects (male/female, 52/31) aged 19 [15; 21] years with prior chest EID-CT scans participated in low-
dose (IQ level=16, n=43) and ultra-low-dose (IQ level=10, n=40) chest PCD-CT scanning with a tube voltage of Sn100. The contrast-
noise-ratio (CNR), figure of merit for CNR (FOMCNR, normalising the CNR to effective dose), and edge sharpness of maximum nodule
were calculated. Moreover, a 5-point Likert scale was used to subjectively evaluate the image quality.
Results or Findings: The low-dose and ultra-low-dose groups with effective doses of 0.40 [0.35; 0.44] and 0.22 [0.16; 0.27] mSv,
which were 7.67 and 14.42 times lower than EID-CT, achieved a mean detection rate of 93.56% and 84.48% for preexisting lung
nodules, respectively. PCD-CT with higher FOMCNR and sharpness but lower CNR than EID-CT (all P <0.05). In the low-dose group, in
detail, the CNR, FOMCNR, and sharpness were 18.7 [14.8; 24.0] vs 19.7 [14.1; 26.2], 915 [632; 1578] vs 128 [63.2; 268], and 543
[446; 704] vs 422 [344; 545] Hounsfield Unit (HU)/pixel for PCD-CT and EID-CT, respectively. In the ultra-low-dose group, these values
were 17.1 [15.4; 20.7] vs 22.4 [16.6; 26.6], 1490 [955; 2441] vs 150 [96; 207], and 594 [464; 752] vs 445 [362; 617] HU/pixel,
respectively. PCD-CT showed statistically significant superiority in the image quality, motion artifacts, and display of lung nodules and
skeleton, but inferior in the presentation of mediastinal lymph nodes and lung markings compared to EID-CT (P <0.05).
Conclusion: PCD-CT allowed reliable detection of metastatic lung nodules with significant radiation dose reduction.
Limitations: This technique has relatively poor visualisation of soft tissues, but it does not compromise the detection of crucial
structures like lymph nodes.
was waived.
732
E³ 25D - Venous imaging
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Katarzyna Gruszczyńska; Katowice / Poland

Katarzyna Gruszczyńska; Katowice / Poland
Chest and upper extremities (15 min)

Alexandros Kallifatidis; Thessaloniki / Greece
1. To discuss the role of CT and MR in imaging the chest and upper extremity venous vasculature.
2. To show examples of common conditions involving the chest and upper extremity veins.
3. To discuss the essential elements of the radiology report in patients with suspected or known chest and/or upper extremity venous
disease.
Abdomen (15 min)

Ali Devrim Karaosmanoğlu; Ankara / Turkey
1. To discuss the role of CT and MR in imaging the abdominal venous vasculature.

2. To show examples of common conditions involving the abdominal veins.
3. To discuss the essential elements of the radiology report in patients with suspected or known abdominal venous disease.
Peripheral venous system (15 min)

Carsten Willem Koen Paul Arnoldussen; Venlo / Netherlands
1. To discuss the role of CT and MR in imaging the lower extremity venous vasculature.
2. To show examples of common conditions involving the lower extremity veins.
3. To discuss the essential elements of the radiology report in patients with suspected or known lower extremity venous disease.
733
EIBIR 17 - From seed to success: exciting projects funded by the ESR Research Seed Grant Global
Categories: EuroSafe Imaging/Radiation Protection, Research

ETC Level: LEVEL I
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
This session showcases interim outcomes of three ESR Research Seed Grant Global projects, promoting advancements in the fields of
quality and safety and value-based radiology. It introduces the ESR Research Seed Grant Global funding scheme, a programme
aiming to stimulate innovative projects and pilot studies from Eastern Europe, Central Asia and Africa to propel them towards larger
research efforts and further funding applications.
Moderators:

A Radiology-Driven Approach to Improve Clinical Outcomes for Patients with Hepatocellular Carcinoma in Tanzania: A
Quality Improvement Project (15 min)
Erick Michael Mbuguje; Dar es salaam / Tanzania
Assessing the Safety and Effectiveness of the First Interventional Radiology Training Program in Uganda: A Quality
Assurance Project (15 min)
Eva Nabawanuka; Kampala / Uganda
Improving utilisation of mechanical thrombectomy for acute ischemic stroke by implementation of standardised
protocols for patient selection, imaging and referral (15 min)
Tajana Turk; Osijek / Croatia
Discussion (5 min)
734
OF 17R - Towards equity, diversity and inclusion in radiography and research
Categories: Education, Medico-legal, Professional Issues, Radiographers, Research

Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
This session highlights the essential commitment of the radiography field to fostering an environment that embraces diversity and
champions equity and inclusion. Comprising three thought-provoking talks, this session explores the imperative of EDI in radiography
curriculum, management, and research. The session should be of interest to radiographers, educators and researchers as it seeks to
explore EDI in curriculum design, leadership, and the dissemination of knowledge, with the aim of empowering attendees to actively
contribute to a more inclusive and equitable future for radiography professionals and the broader healthcare community. as it aims to
empower attendees to actively contribute to a more inclusive and equitable future for radiography professionals and the broader
healthcare community.
Moderator:

Working together to enhance curriculum design that meets future needs (16 min)
Janice Mary St John-Matthews; Newport / United Kingdom
Fostering inclusive leadership: driving EDI in radiography management (16 min)

Promoting EDI in journal publishing (16 min)

735
MR 17 - MR safety: towards a practical approach for the European reality
Categories: Education, Evidence-Based Imaging, Management/Leadership

ETC Level: LEVEL II
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Anna Pichiecchio; Pavia / Italy

The MR safety landscape in Europe: results of the ESR survey (15 min)
Francesco Santini; Basel / Switzerland
1. To learn about the current state of MR safety practices and guidelines in Europe based on the comprehensive survey conducted by
the ESR MR Safety and Quality Working Group.
2. To appreciate the challenges and variations in MR safety protocols across different European countries and gain insights into the
common trends and areas for improvement.
3. To understand the importance of harmonising MR safety standards in Europe and the potential impact on patient safety and
outcomes.
To scan or not to scan: maximising patient benefit with a practical MR safety approach (20 min)
Isabella Maria Björkman-Burtscher; Gothenburg / Sweden
1. To learn a practical approach for assessing the risks and benefits of MR scans on an individual patient basis, considering various
factors such as patient history, indications, and specific safety considerations.
2. To analyse decision-making in cases with incomplete safety information.
3. To formulate strategies and recommendations for implementing a practical MR safety approach in clinical practice, ensuring
maximum patient benefit while minimising risks.
Standardising MR safety through nationally certified education: the Austrian experience (20 min)
Siegfried Trattnig; Vienna / Austria
1. To learn about the Austrian experience in developing and implementing a nationally certified education programme for MR safety,
including the curriculum, training requirements, and outcomes.
2. To appreciate the benefits of standardised MR safety education in improving patient safety, reducing errors, and enhancing the
competence and confidence of healthcare professionals involved in MR imaging.
3. To understand the challenges, strategies, and lessons learned from the Austrian model and its potential applicability in other
healthcare systems aiming to standardise MR safety education.
736
RC 1702 - Update of BI-RADS lexicon: case-based radiologic-pathologic correlation
Categories: Breast, Imaging Methods

ETC Level: LEVEL I
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Federica Pediconi; Roma / Italy

Federica Pediconi; Roma / Italy
Mass and non-mass lesions (15 min)

1. To recognise the different presentations of common benign and malignant breast masses on mammography, DBT, ultrasound, and
MRI.
2. To introduce the new "non-mass" category with ultrasound.
3. To describe the main imaging features of common benign and malignant breast masses according to the BI-RADS lexicon.
4. To understand the assessment and management of different breast masses.
Microcalcifications and non-mass enhancement (NME) (15 min)

Füsun Taşkin; Istanbul / Turkey
1. To present different types of breast microcalcifications on mammography according to the updated BI-RADS lexicon.
2. To identify and name different NMEs on MRI according to the BI-RADS lexicon.
3. To list the possible differential diagnoses of these entities and determine patient management.
Architectural distortions and asymmetric densities (15 min)

Theodora Kanavou; Larissa / Greece
1. To correctly identify an architectural distortion according to the BI-RADS lexicon.

2. To list the most common correlations between architectural distortion and other mammographic findings.
3. To describe and explain differences between asymmetric density with other asymmetric mammographic and DBT findings.
Panel discussion: The importance of radiologic-pathologic correlation (10 min)
737
E³ 22D - On-call in the emergency department: the power of radiography
Categories: Education, Emergency Imaging, General Radiology, Imaging Methods, Musculoskeletal

ETC Level: LEVEL I
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Upper extremity (25 min)

Milko Charles De Jonge; Amsterdam / Netherlands
1. To learn helpful signs for fracture detection.

2. To avoid typical pitfalls for misdiagnosis.
3. To learn about the best views and their limitations.
Lower extremity (25 min)

1. To discuss the use of radiographs in identifying fractures of the lower limb.

2. To recognise fracture patterns and their importance.
3. To discuss the strengths and weaknesses of radiographs in the lower limb.
738

ETC Level: LEVEL II
Date: March 2, 2024 | 08:00 - 08:30 CET
CME Credits: 0.5
Moderators:


3. To gain deep knowledge of chest radiology and have the opportunity to grow.


739
E³ 1726 - Imaging response criteria in oncology
Categories: Education, Hybrid Imaging, Oncologic Imaging

ETC Level: LEVEL II
Date: March 2, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

RECIST 1.1 and iRECIST (15 min)

1. To understand the basics of RECIST 1.1 criteria

2. To learn about immunotherapy and iRECIST
3. To investigate the limitations of RECIST and iRECIST.
mRECIST and CHOI (15 min)

1. To understand when we use mRECIST.

2. To learn about which tumours are eligible for CHOI criteria.
3. To investigate desirable future steps ad limitations.
PERCIST and LUGANO (15 min)

1. To understand when PERCIST criteria are applied.

2. To learn about LUGANO criteria.
3. To investigate the pros and cons.
Panel discussion: Are morphological criteria enough? (10 min)
740
ESR/ESSO 17 - Interdisciplinary seminar series: from imaging to cancer surgery - Soft tissue sarcoma
Categories: Abdominal Viscera, Imaging Methods, Multidisciplinary, Oncologic Imaging, Paediatric

Date: March 2, 2024 | 08:30 - 09:30 CET
CME Credits: 1
Retroperitoneal sarcoma is a challenging, rare disease that requires to be managed by expert multidisciplinary teams to achieve the
best chance of cure. Interaction between radiologists and surgeons is key in the initial diagnostic work and in planning the extent of
surgical resection that may often require multivisceral procedures. Adequate diagnostic work is also crucial to rule out any role for
multimodal treatments that are preferably administered before surgery.
Moderator:
Marco Fiore; Milan / Italy

Soft tissue sarcoma (50 min)

Andrea Vanzulli; Tradate (VA) / Italy
Wrap-up (5 min)
741
CUBE 19 - Tool of the trade in Interventional Neuroradiology

Date: March 2, 2024 | 09:00 - 09:30 CET
Neuro IR Day - Topic Coordinator: Prof. Jean-Pierre Pruvo
in the audience.
Moderator:
Jean-Pierre Pruvo; Lille / France

Tool of the trade in Interventional Neuroradiology (28 min)

François Zhu; Nancy / France
1. To present the tools used for mechanical thrombectomy (guide catheter, aspiration catheter, stent retriever).
2. To present the tools used for intracranial aneurysms (stent-assisted coiling, flow diverter, flow disrupter).
742
IF 18 - Large language models (LLM), chatbots, and the next generation of texts in radiology
Categories: Artificial Intelligence & Machine Learning, Education, Management/Leadership, Research

Date: March 2, 2024 | 09:00 - 10:00 CET
CME Credits: 1
Clearly, ChatGPT and other large language models (LLM) are one of the potentially most transformative developments in AI to date.
Those generative models offer potentials not yet fully perceivable – especially if and how those could be applied in medicine and
radiology. This session will discuss potential applications of LLMs in radiology, from documentation and report generation, to teaching
and even academic writing.
Moderator:

The future of reporting and documentation (15 min)

Philippe Puech; LILLE / France
The future of teaching and testing radiology residents (15 min)

AI apps for academic writing: if, when, and how to use it (15 min)
Nicholas Landini; Rome / Italy
Panel discussion: Will large language models and chatbots transform radiology reporting and research? (10 min)
743
ESR eHealth 18 - The EU Health Data Space (EHDS): what does it mean for radiology?
Categories: Imaging Informatics, Management/Leadership, Professional Issues, Research

ETC Level: LEVEL II
Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
The European Health Data Space (EHDS) is a health-specific ecosystem comprising rules, common standards and practices,
infrastructures and a governance framework aiming at empowering individuals through health data access and researchers through a
trustworthy set-up for health data research. This session will provide an overview of what the EHDS is from the EU perspective, how it
is perceived by patients and what challenges to data exchange lie ahead.
Moderator:

Potentials and challenges to data exchange in the EHDS (15 min)

Robin Germain Lucien Decoster; Brussels / Belgium
The EU vision for health data exchange (15 min)

Aleksandra Wesolowska; Luxembourg / Luxembourg
The patient perspective on data exchange (15 min)

Panel discussion: How will the EHDS impact us? How can we use it? (10 min)
744
RC 1811 - Neurovascular intervention for the non-interventionalist
Categories: Interventional Radiology, Neuro, Vascular

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Zsolt Mihaly Kulcsar; Zurich / Switzerland

Stroke imaging: what does the interventionalist need to know from your report? (15 min)
1. To name and identify the occluded vessel and stroke aetiology based on thrombus appearance.
2. To list good outcome predictors based on ASL and SWAN sequences.
3. To describe and reflect on estimating the volume and location of the infarcted tissue and tissue at risk for infarction.
Aneurysm evaluation: what diagnostic information helps better treatment planning? (15 min)
Pedro Vilela; Lisbon / Portugal
Dural fistula and arteriovenous malformations (AVMs): how to predict rupture risk? (15 min)
1. To understand how the hemodynamic of brain arteriovenous malformations (BAVMs) and dural arteriovenous fistulas (DAVFs) of the
venous outflow are related to haemorrhagic risk in natural history.
2. To appropriate the use of clinical classification to predict the natural history.
3. To focalise the importance of focal venous ectasia in the draining system in DAVFs.
4. To focalise the importance of nidal and pre-nidal evaluation of the venous outflow in BAVMs.
Panel discussion: Collaboration between diagnostic and interventional neuroradiologists: how I do it and how I would
like to do it (10 min)
745
Categories: Artificial Intelligence & Machine Learning, Cardiac

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderators:

Segment level coronary artery calcium scoring: an agreement analysis of non-contrast CT data from a multicentre trial
(8 min)
Sotirios Tsogias; Berlin / Germany
Discussant (4 min)
Bernd J. Wintersperger; Toronto / Canada
Quality of life and chest pain outcomes in women and men following initial CT angiography or invasive coronary
angiography strategies for stable chest pain: results from the randomised DISCHARGE trial (8 min)
Jonathan Dermot Dodd; Dublin / Ireland
Discussant (4 min)
Stephen Harden; Southampton / United Kingdom
Artificial intelligence-enabled opportunistic screening for coronary artery calcium and thoracic aortic disease on low-
dose, non-gated CT scans: an updated sub-analysis of the National Lung Screening Trial (8 min)
Aleena Yasin; Stanford / United States
Discussant (4 min)
Comparative analysis of cardiac CT and invasive coronary angiography in guiding functional testing and
revascularisation for suspected stable coronary artery disease (8 min)
Jonathan Dermot Dodd; Dublin / Ireland
Discussant (4 min)
746
ESGAR Quiz - The Young ESGAR Quiz at ECR 2024
Categories: Abdominal Viscera, Education, General Radiology, GI Tract

ETC Level: LEVEL II
Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
This is an interactive session in which the attendees will answer short questions on gastrointestinal radiology. Each round will explore
a key theme. Speakers will navigate the challenging abdominal imaging cases using a fun Quiz Show format, presenting insights on
differential diagnoses and essential diagnostic clues.
Moderators:
Carmelo Sofia; San Filippo Del Mela / Italy
Francesca Castagnoli; London / United Kingdom

Quiz round 1 (15 min)

Nino Bogveradze; Amsterdam / Netherlands

Cesare Maino; Monza / Italy

Ana Maria Villanueva Campos; MADRID / Spain
Conclusion and final remarks (5 min)

747
RPS 1801 - From guidelines to new approaches in liver tumour imaging

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Alternative surveillance using CT/MR improves clinical outcomes in high-risk patients with chronic hepatitis B (7 min)
Dong Ho Lee; Seoul / Korea, Republic of
Author Block: D. H. Lee; Seoul/KR

Purpose: This study aimed to evaluate the outcome of alternative hepatocellular carcinoma (HCC) surveillance using CT/MR
compared to US only in chronic hepatitis B (CHB) patients.
Methods or Background: We enrolled consecutive CHB patients undergoing regular HCC surveillance, classifying into two groups:
US only group and alternative surveillance group. The risk estimation for HCC in CHB (REACH-B) score was calculated to categorise
high and low risk. Outcomes included 10-year overall survival (OS), size and Barcelona Clinic Liver Cancer (BCLC) stage of HCC, and
OS after HCC diagnosis.
Results or Findings: A total of 2,024 patients were enrolled with 1,012 patients in each group. There was no significant difference in
OS (96.0% in US only versus 96.8% in alternative surveillance; P=0.379). In both groups, HCC occurred in 66 patients. Medium size of
HCC in the alternative surveillance was significantly smaller than US only (1.6 cm versus 2.1 cm; P<0.001). The rate of BCLC 0 stage
HCC was also significantly higher in alternative surveillance than US only (71.2% [47/66] versus 42.4% [28/66]; P=0.003). OS after
HCC diagnosis in the alternative surveillance group was significantly better than that in the US only group (83.0% versus 67.0%;
P=0.025). In the high-risk group including 970 patients, alternative surveillance provided significantly better OS (97.3% versus 93.6%;
P=0.029) and OS after HCC diagnosis (83.0% versus 60.6%; P=0.010) than US only. However, there was no significant difference in
both OS (P=0.202) and OS after HCC diagnosis (P=0.937) in 1,054 patients with low risk.
Conclusion: Alternative surveillance using CT/MR enabled the detection of HCC at an earlier stage, with smaller size, than US only
and had potential to improve OS after HCC diagnosis, especially for patients with high risk.
Limitations: The retrospective study design, prone to selection bias, was an identified limitation.
Ethics committee - additional information: This study was designed as a single-centre retrospective cohort study and approved
by the institutional review board of Seoul National University Hospital (spproval number: H-2208-023-1346) with the waiver of
requirement of informed consent.
A comprehensive study on the feasibility and diagnostic potential of fluctuation imaging (FLI) in liver tumour
assessment (7 min)
Dong Ho Lee; Seoul / Korea, Republic of
748
Author Block: D. H. Lee, J. Y. Lee; Seoul/KR
Purpose: B-mode ultrasound (US) is commonly used for liver disease imaging, although it has limitations in characterising liver
masses. This study aims to assess the clinical potential of a newly developed US technique called fluctuation imaging (FLI).
Methods or Background: In this prospective exploratory study, we enrolled 65 patients diagnosed with liver tumours based on
histopathology or typical imaging findings. All participants underwent US examination, which included FLI. FLI involved capturing
approximately 100 frames of images during a 10-second breath-holding period to generate a color-coded FLI map. The presence of
fluctuation signals within liver tumours on the FLI map was noted: the presence of increased fluctuation signal was represented as a
yellow to red colour within the tumour. We compared the incidence of increased fluctuation signals using the chi-square test.
Results or Findings: FLI map creation was successful in 95.4% of cases (62/65), with three cases failing due to respiratory motion.
Among the 62 analysed patients, 38 had 42 haemangiomas (median size: 1.9 cm), 13 had hepatocellular carcinoma (median size: 2.6
cm), two had cholangiocarcinoma (2.6 and 4.5 cm), and one each had cortical adenoma from hepato-adrenal fusion (3.0 cm),
malignant mesothelioma (4.5 cm), and angiomyolipoma (6.1 cm). The remaining six participants had seven metastases from colon
cancer (median size: 2.3 cm). Among the 42 haemangiomas, 23 (54.8%) exhibited increased fluctuation signals on the FLI map. In
contrast, only three out of 25 non-haemangioma liver tumours (12.0%) showed increased fluctuation signals (54.8% for
haemangioma's versus 12.0% for non-haemangiomas, P=0.001).
Conclusion: The acquisition of FLI maps during liver US examinations proved to be feasible, and the identification of increased
fluctuation signals on the FLI map may assist in detecting haemangiomas.
Limitations: The small number of participants was an identified limitation.
Funding for this study: Canon Medical Systems supported this study.
Ethics committee - additional information: The institutional review board of Seoul National University Hospital approved this
prospective study.
Comparison of diagnostic guidelines for hepatocellular carcinoma on gadoxetic acid-enhanced liver magnetic
resonance imaging (7 min)
Jeong-Hee Yoon; Seoul / Korea, Republic of
Author Block: J-H. Yoon, Y. K. Kim, W. Chang, J-I. Choi, B. J. Park, J-Y. Choi, H. S. Park, C. H. Lee, J. M. Lee; Seoul/KR
Purpose: Non-invasive diagnostic guidelines for HCC vary, especially regarding hepatobiliary agent-enhanced magnetic resonance
imaging (HBA-MRI). We evaluated the diagnostic performance of four guidelines and readers’ judgment in diagnosing HCC using HBA-
MRI in high-risk patients.
Methods or Background: This retrospective study included treatment-naive patients at risk of HCC who underwent HBA-MRI from
January 2015 to June 2018. Four radiologists, using a pre-programmed algorithm on a web-based platform, independently reviewed
focal liver lesions (FLLs) according to four guidelines: American Association for the Study of Liver Diseases/Liver Imaging Reporting
and Data System (AASLD/LI-RADS), Korean Liver Cancer Association-National Cancer Center (KLCA-NCC), European Association for the
Study of the Liver (EASL), Asian Pacific Association for the Study of the Liver (APASL). Readers’ judgment was also recorded. The
guidelines’ diagnostic performance was compared in all patients and subgroups. Additionally, the diagnostic odds ratio (DOR) was
assessed.
Results or Findings: We analysed 2,445 FLLs in 2,237 patients; 69.3% (1,694/2,445) were HCC. KLCA-NCC demonstrated the
highest accuracy (80.0%), followed by APASL, AASLD/LI-RADS (77.8%, 76.4%), and EASL (75.1%). APASL exhibited the highest
sensitivity (89.1%), followed by KLCA-NCC (78.2%), while AASLD/LI-RADS displayed the highest specificity (89.6%), followed by EASL
(88.1%). The DORs were 20.7 for AASLD/LI-RADS, 18.9 for KLCA-NCC, 16.8 for EASL, and 8.9 for APASL. The readers’ judgment
demonstrated higher accuracy than the guidelines (86.0%, P<0.001). In small (<20 mm) FLLs, Eastern guidelines showed higher
accuracy than Western guidelines (P<0.001). Diagnostic accuracy of guidelines was 71.8%–79.5% in cirrhosis and 75.2%–82.3% in
chronic hepatitis B without cirrhosis.
Conclusion: Eastern guidelines demonstrated high sensitivity, while Western guidelines displayed high specificity. KLCA-NCC
achieved the highest accuracy, and AASLD/LI-RADS exhibited the highest DOR.
Limitations: The retrospective design inevitably introduced bias which we attempted to minimise by including a large population.
Funding for this study: This work is financially supported by Bayer.
Ethics committee - additional information: This study was approved by the Institutional Review Board of Seoul National
University Hospital and participating institutions.
Diagnostic performance of CT/MRI LI-RADS v2018 in non-cirrhotic steatotic liver disease (7 min)
Justin Ruey Tse; Stanford / United States
749
Author Block: J. Cao, A. Shon, L. S. Yoon, J. R. Tse; Stanford, CA/US
Purpose: A substantial proportion of patients with steatotic liver disease (SLD) develop hepatocellular carcinoma (HCC) without
cirrhosis. Our purpose was to assess the diagnostic performance of CT/MRI LI-RADS v2018 (LI-RADS) among patients with non-
cirrhotic SLD.
Methods or Background: This IRB-approved, retrospective, single-centre study included 103 observations with histopathologic
correlation from 77 adult patients who underwent liver CT/MRI from 2010-2023. All patients had histopathologic evidence of SLD
without cirrhosis. Three board-certified abdominal radiologists (R1-R3) blinded to the tissue diagnoses assessed each observation by
LI-RADS criteria and assigned a final category. Inter-reader agreements were assessed with Cohen’s kappa. The positive predictive
value (PPV), sensitivity, specificity, and accuracy of identifying HCC and overall malignancy were calculated.
Results or Findings: Of 103 observations, 59 (57%) were benign and 44 (43%) were malignant. PPV for HCC was 0-0% for LR-1,
0-0% for LR-2, 0-11% for LR-3, 13-22% for LR-4, 75-88% for LR-5, 0-8% for LR-M, and 50-75% for LR-TIV. For malignancy (including
HCC), PPV was 0-0% for LR-1, 0-12% for LR-2, 5-12% for LR-3, 19-33% for LR-4, 78-88% for LR-5, 65-91% for LR-M, and 100-100% for
LR-TIV. For LR-5 in detecting HCC, sensitivity was 83%, 79%, 83% (R1-3, respectively), specificity was 89%, 96%, 92%, and accuracy
was 87%, 91%, and 89%. For composite categories of LR-5, LR-M, or LR-TIV in detecting overall malignancy, sensitivity was 89%, 89%,
and 86%, specificity was 86%, 95%, and 81%, and accuracy was 87%, 92%, and 84% respectively. Most common false positives for
LR-5 were due to hepatocellular adenomas. Inter-reader agreements for major features ranged from 0.660-0.833 and was 0.751 for
the final category.
Conclusion: CT/MRI LI-RADS v2018 can be applied to non-cirrhotic SLD patients with high but imperfect specificity.
Limitations: This was a single-centre, retrospective study with selection bias of histologically-proven observations.
Ethics committee - additional information: The study was approved by the university Institutional Review Board
Value of preoperative magnetic resonance imaging for diagnosis of LR‑3 and LR-4 lesions coexisting with main
Yuyao Xiao; Shanghai / China
Author Block: Y. Xiao, C. Yang, M. Zeng; Shanghai/CN

Purpose: The diagnostic performance of magnetic resonance imaging (MRI) features favouring hepatocellular carcinoma (HCC) in
LR-3 and LR-4 observations in high-risk patients have been evaluated in previous studies. Our objective was to explore which
preoperative clinical data and conventional MRI findings may indicate the presence of HCC in LR-3 and LR-4 lesions of HCC patients.
Methods or Background: In this study, we enrolled 102 patients with histopathologically confirmed HCC coexisting with 110 LR-3
and LR-4 lesions (HCCs group [n = 66], non-HCCs group [n = 44]). Two radiologists retrospectively assessed the preoperative MRI
features, and each lesion was assigned according to LI-RADS. Preoperative clinical data were also evaluated. Logistic regression
analyses were used to assess the relative value of these parameters as potential predictors of HCC.
Results or Findings: On multivariate analysis, the presence of restricted diffusion (OR: 18.590, p<0.001), delayed enhancement
(OR: 0.113, p<0.001), and T2WI mild-moderate hyperintensity (OR: 3.084, p=0.048) were found to be independent predictors for HCC
diagnosis. The sensitivity and specificity of the above independent variables and their combination for the diagnosis of HCC ranged
from 47.0–80.3% and 56.8–97.7%, respectively. Combining these three findings for the prediction of HCC resulted in a specificity
greater than 97%, and the AUC further increased to 0.874.
Conclusion: The presence of restricted diffusion, delayed enhancement, and T2WI mild-moderate hyperintensity can be useful
features in risk stratifying coexisting LR-3 and LR-4 lesions in HCC patients.
Limitations: Firstly, this was a single-centre retrospective study lacking longitudinal data. Secondly, the number of LR-4 lesions was
also relatively small in our study, so our results might not represent the true spectrum of LR-3 and LR-4 observations in HCC patients.
Funding for this study: This study received funding from the National Natural Science Foundation of China (grant number:
82171897, 82272078).
Ethics committee - additional information: This retrospective study was approved by the local Ethics Committee, and written
informed consent was waived due to its retrospective design.
A new criterion for diagnosis of subcentimetre hepatocellular carcinoma in treatment-naive patients using tumour
biomarker and gadoxetic acid-enhanced MRI features (7 min)
Peng Huang; Shanghai / China
750
Author Block: P. Huang, F. Wu, C. Wang, Y. Xiao, G. Miao, C. Yang, M. Zeng; Shanghai/CN
Purpose: Based on imaging findings, current diagnostic criteria for subcentimetre hepatocellular carcinoma (HCC) have suboptimal
diagnostic performances. The aim of the present study is to develop a diagnostic nomogram including gadoxetic acid-enhanced MRI
features and tumour biomarkers.
Methods or Background: This study enrolled treatment-naive patients with chronic hepatitis B who had a solitary subcentimetre
observation from April 2016 to March 2023. The final diagnosis was confirmed by pathology for HCC and pathology or follow-up for
non-HCC controls. Logistic regression analysis for imaging features and tumour biomarkers was used to identify independent
predictors associated with HCC that were then incorporated into the nomogram. Diagnostic performances of the new criterion were
compared to wash-in and -out.
Results or Findings: A total of 224 patients (116 with HCC and 108 with non-HCC) were divided into a training cohort (including 155
patients) and temporal validation cohort (including 69 patients). The factors associated with HCC diagnosis were alpha-fetoprotein
>20 ng/mL or des-gamma-carboxy prothrombin >40 IU/mL, T2WI mild-moderate hyperintensity, arterial-phase hyperenhancement,
portal-phase washout, and transitional-phase hypointensity. Incorporating these 5 factors, the nomogram achieved good concordance
indexes of 0.92 and 0.95 in diagnosing subcentimetre HCC in the training and temporal validation cohorts, respectively, and had well-
fitted calibration curves. Using a nomogram score of 296 as a cut-off, the new criterion yielded higher sensitivity than wash-in and -
out (training cohort: 72.0% versus 51.2%, P<0.001; validation cohort, 79.4% versus 47.1%, P<0.001) without a significant decrease in
specificity (training cohort: 90.4% versus 91.8%, P>0.999; validation cohort, 91.4% versus 91.4%, P>0.999).
Conclusion: Including tumour biomarkers in the diagnostic algorithm may improve the diagnostic performance for subcentimetre
HCC.
Limitations: Our results may be limited in their generalisability, especially in regions other than HBV-endemic areas.
Funding for this study: Funding was received from The Clinical Research Plan of SHDC (number: SHDC2020CR1029B), National
Natural Science Foundation of China (number: 82171897), and National Natural Science Foundation of China (number: 82371923).
Ethics committee - additional information: The institutional review board of Zhongshan Hospital, Fudan University (number:
B2020-372R) approved this study.
Quantitative CT analysis of hepatocellular carcinoma nodules in liver transplant candidates: impact of volume of
interest and contrast phase on recurrence prediction (7 min)
Francesco Rizzetto; Rho / Italy
Author Block: F. Rizzetto, R. Manzini, L. Centonze, C. B. Monti, S. Garziano, J. Di Napoli, L. A. Carbonaro, A. Vanzulli; Milan/IT
Purpose: To evaluate how different lesion volumes-of-interest (VOIs) and contrast phases affect quantitative CT analysis of
hepatocellular carcinoma (HCC) in liver transplant candidates.
Methods or Background: Liver transplant candidates from 2010 to 2019, with waitlist placement CT showing suspicious HCC
nodules (LI-RADS=4-5) and no treatment, were retrospectively included. HCC nodules, with or without a peritumoural region of 5, 10,
and 15 mm, and "background" liver parenchyma were contoured across arterial, venous, and delayed phases. Radiomic features
(RFs), both directly extracted from segmentations and normalised for "background" parenchyma, were computed and compared for
inter-VOIs and inter-phase relative change (RC), and autocorrelation using Spearman’s rho. RFs with RC<10% and rho>0.9 were
considered redundant. Non-redundant RFs with different distribution in patients with and without post-transplant recurrence were
selected to build multiple logistic regression models. Predictive capability for post-transplant recurrence was assessed comparing
areas under the curve (AUC) from receiver operating characteristic analysis.
Results or Findings: The CT scans of 53 patients were selected, for a total of 1,260 segmentations from 105 nodules, with eight
(15%) recurrence cases. Changing VOIs and phases, the number of redundant RFs was 14-26% and 16-34%, respectively, while 4% of
RFs were autocorrelated. Selected non-redundant RFs allowed the building of eight predictive models (pseudo-r2=0.13-0.46), each
comprising 5-12 RFs, with AUC ranging from 0.77 to 90 (p<0.02). Best performance was achieved when segmenting HCC nodules with
or without 5 mm of peritumoural region in arterial phase. Venous and delayed phases showed predictive capability only for
segmentations including 10-15 mm of peritumoural region after normalisation for background parenchyma.
Conclusion: Selection of optimal VOI and contrast phase combination is critical to maximize post liver transplant HCC recurrence
prediction using quantitative CT imaging.
Limitations: The retrospective single-centre design and the limited sample size are the main limitations of the study.
Ethics committee - additional information: The local ethics committee approved the retrospective data collection in an
anonymous, aggregated form.
Development of the MIEM score to predict survival after recurrence in patients with recurrent HCC following curative
resection (7 min)
Hong Wei; Chengdu / China
751
Author Block: H. Wei, B. Song; Chengdu/CN
Purpose: This study aimed to create a predictive score for survival after recurrence (SAR) based on MRI features and clinical-
pathological characteristics in patients with recurrent HCC following initial curative resection.
Methods or Background: This single-centre, retrospective study included consecutive patients who underwent contrast-enhanced
MRI within one month before curative resection for primary HCC and developed recurrence during follow-up from December 2011 to
June 2021. Clinical (i.e. age, sex, aetiology, BCLC stage, and AFP level), pathological (i.e. tumour differentiation, microvascular
invasion, liver capsule invasion, and cirrhosis), and recurrence-related data (i.e. adjuvant therapy, early versus late recurrence, and
intrahepatic versus extrahepatic recurrence) were collected. MRI images were independently reviewed by two radiologists in terms of
all LI-RADS version 2018 imaging features and several non-LI-RADS imaging features. Uni- and multivariable Cox regression analyses
were used to identify variables associated with SAR. A regression-based predictive model was constructed with five-fold cross-
validation.
Results or Findings: A total of 151 patients with recurrent HCC were included (median age, 53 years; 135 men). During a median
follow-up period of 55 months, 31% (47/151) of patients experienced death. Predictors for SAR included non-smooth tumour margin
(hazard ratio [HR] =4.6), rim arterial phase hyperenhancement (HR=1.9), extrahepatic recurrence (HR=1.9), and male sex (HR=0.5).
By incorporating the above predictors, the MIEM score demonstrated a C-index of 0.71 (95% confidence interval: 0.64, 0.79). Using 11
as the threshold of the MIEM score, all patients were categorised into prognostically distinct low- and high-risk groups (P<0.001).
Conclusion: The MIEM score may serve as a useful tool to help predict SAR for patients with recurrent HCC following hepatectomy.
Limitations: The single-centre nature of the study, the retrospective study design, and lack of external validation were identified as
limitations.
Funding for this study: This work was supported by the National Natural Science Foundation of China (grant number: U22A20343).
Ethics committee - additional information: This study was approved by the Biomedical Ethics Review Committee of West China
Hospital, Sichuan University (approval number: 2022-651). The requirement for informed consent was waived (due to the
retrospective design) by the Biomedical Ethics Review Committee of West China Hospital, Sichuan University.
Nomogram development and validation based on Gd-EOB-DTPA-enhanced MRI to predict IDH1 mutation for
intrahepatic cholangiocarcinoma (7 min)
Xiaoqi Zhou; Guangzhou / China
Author Block: X. Zhou, M. Chen, D. Xu, S-T. Feng; Guangzhou/CN

Purpose: Isocitrate dehydrogenase 1 (IDH1) mutation inhibitors show promise for targeted therapy in intrahepatic
cholangiocarcinoma (ICC), but only for IDH1 mutated tumours. The purpose of this study is to establish and validate a model to
predict IDH1 gene mutation in ICC based on preoperative Gd-EOB-DTPA-enhanced magnetic resonance imaging (MRI) images.
Methods or Background: A total of 85 ICCs were randomly assigned to the training set (n=59) and the test set (n=26). Next
generation sequencing and immunohistochemical analysis were performed for IDH1 mutation status. Preoperative imaging features of
Gd-EOB-DTPA-enhanced MRI were qualitatively and quantitatively reviewed by two radiologists. Nomograms were developed based
on the significant variables for differentiating the IDH1-mutated and IDH1-unmutated ICCs in the training sets.
Results or Findings: The IDH1 mutation rate was 20% (17/85). After filtering the factors (p<0.1) based on univariate analysis, the
serum AFP value (p=0.17), intratumoural vessel (p=0.01), location of liver lobe (p=0.14), T1 change ratio (p=0.004), T2 central
brightness (p=0.01) and ADC value (p=0.008) were incorporated into the nomogram for predicting IDH1-mutated ICCs. Combing all
six significant variables provided a diagnostic accuracy of 88%, sensitivity of 75%, and a specificity of 90% for identifying IDH1
mutation status in the test set. The area under the receiver operating characteristic curve value of the logistical regression
coefficient-based nomogram was 0.927 (95% confidence intervals, 0.838-1.000) and 0.881 (95% confidence intervals, 0.690-1.000)
for the training and test sets. The nomogram exhibited good calibration and clinical usefulness.
Conclusion: Gd-EOB-DTPA-enhanced MRI-based nomogram effectively predicts IDH1 mutation status in ICCs and may assist
clinicians in pretreatment decisions.
Limitations: Firstly, this is a retrospective study with potential patient selection bias. Secondly, this was a single-centre study with a
small sample size.
Funding for this study: Funding was received from the National Natural Science Foundation of China (82271958, 81971684).
Ethics committee - additional information: The study was approved by the institutional review board and ethical committee of
Sun Yat-sen University.
Prognostic role of CT and MRI of resectable intrahepatic mass forming cholangiocarcinoma (7 min)
Silvia De Vizio; Rome / Italy
752
Author Block: S. De Vizio, E. Genco, F. M. Sessa, M. G. Brizi, E. Panettieri, F. Giuliante, E. Sala; Rome/IT
Purpose: Imaging is crucial for diagnosis as well as for prognosis of intrahepatic mass forming cholangiocarcinoma. In the last
decade some studies focused on diffusion weighted imaging (MRI) and arterial enhancement's (CT and MRI) role in prognosis
stratification of resectable intrahepatic cholangiocarcinoma. The aim of this retrospective study was to evaluate the best prognostic
imaging biomarkers for resectable intrahepatic mass forming cholangiocarcinomas.
Methods or Background: We retrospectively evaluated 123 patients with histologically proven intrahepatic cholangiocarcinoma,
surgically treated in our university hospital from 2010 to 2022.
Patients were divided in two groups: a multiphasic contrast-enhanced CT group and a multiphasic hepato-specific contrast-enhanced
MR-DWI group. Each group was further divided into resectable lesion and unresectable lesion sub-groups.
The percentage of arterial enhancement at CT and MR was categorised as hypervascular (>50% of the lesion), peripherally enhancing
(10-50%), or hypovascular (<10%); DWI was qualitatively evaluated on ADC maps and lesions graded as diffusion restriction (<1/3 of
the lesion) and diffusion restriction (>1/3).
Analysis was performed by two expert readers.
Overall survival at 1-year and 3-years was assessed by Kaplan-Meyer curves whose differences were analysed by Cox analysis.
Results or Findings: A total of 26 patients were included in the MRI group (6 hypervascular, 14 peripheral, 6 hypovascular), and 35
in the CT group (7 hypervascular, 11 peripheral, 17 hypovascular); 62 patients were excluded because of poor quality of arterial
phase and/or DWI images.
In both groups overall survival at 1-year was higher for the hypervascular group compared to the hypovascular group (87.3% versus
26.2%).
In the MRI group, 1-year survival was higher in the subgroup with diffusion restriction of <1/3 of the lesion.
At Cox proportional hazards model analysis, hypervascularity was associated with a better overall survival.
Conclusion: CT/MR hypervascularity and lower extent of diffusion restriction are both positive prognostic biomarkers.
Limitations: The retrospective study design was an identified limitation.
Photon-counting detector CT for liver lesion detection: optimising virtual monoenergetic images across different sizes
and doses (7 min)
Andre Euler; Baden / Switzerland
Author Block: D. Racine1, V. Mergen2, A. Viry1, T. Frauenfelder2, H. Alkadhi2, V. Vitzthum1, A. Euler3; 1Lausanne/CH, 2Zurich/CH,
3
Baden/CH
Purpose: To evaluate the optimal energy level of virtual monoenergetic images (VMI) from photon-counting detector CT (PCD-CT) for
the detection of liver lesions as a function of phantom size and radiation dose.
Methods or Background: An anthropomorphic abdominal phantom with liver parenchyma and lesions was imaged on a dual-source
PCD-CT at 120 kVp. The lesions were hypo- or hyperattenuating and had diameters of 5-10 mm. Rings of fat-equivalent material were
added to emulate medium- or large-sized patients. The medium size was imaged at 5, 2.5, and 1.25 mGy and the large size at 5 and
2.5 mGy, respectively. Each setup was imaged ten times. For each setup, VMIs from 40 to 80 keV at 5 keV increments were
reconstructed with quantum iterative reconstruction at a strength level of 4 (QIR-4). Lesion detectability was measured as area under
the receiver operating curve (AUC) using a channelised hotelling model observer with ten dense differences of Gaussian channels.
Results or Findings: Overall, highest detectability was found at 65 and 70 keV for both hypo- and hyperattenuating lesions in the
medium and large phantom, independent of radiation dose (AUC range: 0.910-0.999 for the medium and 0.935-0.987 for the large
phantom, respectively). Lowest detectability was found at 40 keV irrespective of the radiation dose and phantom size (AUC range:
0.784-0.993). Reducing radiation dose decreased lesion detectability more strongly at 40-50 keV as compared with 65-75 keV. At
equal radiation dose, detection as a function of VMI energy differed more strongly for the large size as compared with the medium
size phantom (12 % versus 6 %).
Conclusion: Detectability of hypo- and hyperattenuating liver lesions differed between VMI energies for different phantom sizes and
radiation doses. VMI at 65 and 70 keV yielded highest detectability, independent of phantom size and radiation dose.
Limitations: The phantom study design was an identified limitation.
Ethics committee - additional information: Ethics committee approval was not needed because of the design as a phantom
study.
Detection of spoke-wheel pattern in focal nodular hyperplasia with microvascular flow imaging in a preliminary study
(7 min)
Aladár David Rónaszéki; Gyömrő / Hungary
753
Author Block: B. Zsély, A. D. Rónaszéki, B. K. Budai, R. Surendranath, R. Stollmayer, Z. Zsombor, G. Györi, P. Maurovich-Horvat, P. N.
Kaposi; Budapest/HU
Purpose: Microvascular flow imaging (MVFI) is a novel Doppler ultrasound (US) technique to detect slow flow in capillary vessels. The
spoke-wheel vascularity pattern is characteristic of focal nodular hyperplasia (FNH), and it is consistently detectable with MVFI. The
identification of the typical patterns of vascularity, including the spoke-wheel pattern with MVFI can expedite diagnosis, spare patients
from unnecessary procedures, and reduce costs.
Methods or Background: This study retrospectively collected MVFI US images of 50 patients followed for known focal liver lesions
(FLL). The lesions were diagnosed by contrast-enhanced ultrasound (CEUS), biopsy, or magnetic resonance imaging (MRI) with a liver-
specific contrast agent. Patients were examined using a Samsung RS85 Prestige scanner. The vascularity of the lesions was evaluated
with either conventional colour Doppler US or directional power Doppler imaging using the S-FlowTM application. We used the MV-
FlowTM application for microvascular flow imaging by recording a 5-10 second-long video in a breath-hold.
The Wilk test was used to check the normality of the data. Continuous variables were analysed with the Mann-Whitney U-test,
categorical variables with Fisher’s exact test, and chi-square test. A logistic regression analysis was performed with the individual
morphological characteristics, the variables were characterised by odds ratio (OR) value, sensitivity, and specificity.
Results or Findings: The study patients included 21 FNHs, 7 haemangiomas, 9 hepatocellular carcinomas (HCC), and 13 liver
metastases. The spoke-wheel pattern detectable with MVFI was detected in a higher number in cases of FNHs (OR>100 [95% CI:
45.365 - Inf]; p<0.001), which proved to be specific (1.0) and showed high sensitivity (0.955).
Conclusion: Our preliminary results show the spoke-wheel pattern can be easily detected with MVFI, even in the case of small FNHs,
whose vascularisation pattern can help distinguish them from other FLLs.
Limitations: The retrospective study design was an identified limitation.
Funding for this study: The authors received no funding for this study.
Ethics committee - additional information: The study has been approved by the institutional ethics committee of our university
(SE-RKEB 140/2020).
754
HW 18Sd - Mastering stroke perfusion imaging and MR, CT and AI applications

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1

1. To become familiar with the use of MRI for the purpose of using perfusion MRI in assessing stroke patients.
2. To learn about stroke and treatment options by examining the role of CT scans in guiding treatment decisions for strokes.
3. To develop practical skills in perfusion techniques and post-processing for patient diagnosis, triage, and treatment orientations.
4. To understand available public and commercial platforms in stroke imaging and artificial intelligence-driven applications for acute
stroke.

Dennis M Hedderich; Munich / Germany
755
E³ 1823 - Interventional
Categories: Interventional Radiology, Vascular

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

Basic principles of angiography and image-guided interventions (28 min)

Rene Mueller Müller-Wille; Wels / Austria
1. To describe the normal anatomy and normal variants of the arterial and venous vascular system.
2. To understand the importance of pre-procedure planning and selection of image-guidance techniques.
3. To explain basic percutaneous image-guided techniques, including arterial access as well as biopsy and drainage.
Image-guided interventions in oncology (28 min)

Adrian Kobe; Zurich / Switzerland
1. To describe the basic technical methodological principles and indications of imaging-guided interventions in oncological disorders,
including thermal ablation techniques.
2. To understand the principles and indications for vascular interventions in cancer, such as the transarterial treatment of liver
tumours.
3. To become familiar with post-treatment follow-up, highlighting normal and abnormal pathological imaging findings.
Vascular interventions (28 min)

Dimitrij Kuhelj; Ljubljana / Slovenia
1. To become familiar with the pretreatment imaging flowchart in atherosclerotic diseases.

2. To describe the indications and techniques for arterial angioplasty and stenting.
3. To explain the techniques of arterial embolisation and coiling, as well as thromboaspiration.
756
E³ 1821 - Genitourinary incidentalomas

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Adrenal (45 min)

Olivera Nikolic; Novi Sad / Serbia
1. To illustrate the spectrum of imaging findings in adrenal incidentalomas.

2. To explain the key points in the differential diagnosis.
3. To learn how to avoid interpretation pitfalls.
Renal (45 min)

Dominik Nörenberg; Mannheim / Germany
1. To illustrate the spectrum of imaging findings in renal incidentalomas.

2. To explain the key points in the differential diagnosis.
3. To learn how to avoid interpretation pitfalls.
757
OF 18T - Photon-counting CT: what you wanted to know but were afraid to ask
Categories: Education, General Radiology, Imaging Methods, Professional Issues, Students

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderators:
Mirjam Gerwing; Münster / Germany

Mirjam Gerwing; Münster / Germany
The basics of photon-counting CT (15 min)

Martin J. Willemink; San Diego / United States
1. To describe the basic principles of photon-counting CT.

2. To discuss the advantages of current CT systems.
Clinical application: cardiothoracic imaging (15 min)

1. To describe the clinical application in cardiothoracic imaging.

2. To discuss case examples to show the advantages and limitations of clinical practice.
Clinical application: emerging areas (15 min)

Dietmar Tamandl; Vienna / Austria
1. To describe the emerging areas of clinical application.

2. To discuss case examples to show the advantages and limitations of clinical practic.
Open forum discussion: How will photon-counting CT change my practice? (10 min)
758
RPS 1802 - Advances in breast MRI techniques
Categories: Breast, Contrast Media, Evidence-Based Imaging, Imaging Methods

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Thomas H. Helbich; Vienna / Austria
Abbreviated breast magnetic resonance imaging (MRI) protocols in a very high-risk population: friend or foe? (7 min)
Tamara Suaris; London / United Kingdom
Author Block: L. Metaxa, V. Papalouka, N. Vidyaprakash, S. Lee, S. Dani, T. Suaris; London/UK

Purpose: AbbMRI aims to provide a cost-effective study without compromising diagnostic accuracy. We aim to assess the diagnostic
performance of AbbMRI in high-risk breast cancer screening.
Methods or Background: Four experienced breast radiologists retrospectively reviewed the AbbMRI protocol in 236 MRI breast
studies performed for very high-risk between January 2018 – 2021. Readers were blinded to prior imaging, history, and patient
outcomes. The protocol consisted of six sequences: pre-contrast T1W, T2W axial, two dynamic postcontrast, subtracted axial
sequences and MIP reconstruction. Results were compared with existing reports of the full protocol (19 sequences).
Results or Findings: From the 236 MRI studies, four invasive malignancies and one B3 lesion (radial scar) were identified. The
average time to read the AbbMRI was 69 seconds (range: 28 to 240 seconds; 95% CI: 63, 74 seconds).
The inter-rater rate (IRR) between the abbreviated and full protocol was: 91.5%. The recall rate for AbbMRI was 9.5%, and 6.1% for
full protocol. Sensitivity achieved by AbbMRI (100%) was equal to the full protocol (no cancers were missed). NPV were high for both
reading sessions (100%). Specificity and PPV of the abbreviated MRI was lower compared to the full protocol (91.2 versus 94.7, and
8.9 versus 13.8%, respectively) (p value .01, a=0.5).
Conclusion: Abbreviated MRI is feasible in screening MRI of high risk patients, with sensitivity/CDR comparable to the full MRI
protocol, but with a small increase in the recall rates. Abbreviated breast MRI can be considered as a safe alternative to conventional
MR studies in UK very high-risk screening, reducing MR waiting times/backlog.
Limitations: Identified limitations were: (1) this was a retrospective study, with data for the full protocol collected from existing
reports, where the radiologists were not blinded to the patients' personal history; (2) the abbreviated MRI was not double read.
Ethics committee - additional information: This study was approved by the Director of Research.
Evaluation of pure DCIS with abbreviated breast MRI (7 min)

Nurbanu Basdogan; Istanbul / Turkey
759
Author Block: A. Arslan, N. Basdogan, C. S. Topal, F. Ezberci; Istanbul/TR
Purpose: The most reliable imaging method for detecting breast cancer is magnetic resonance imaging (MRI) of the breast.
'Abbreviated MRI' imaging protocol was developed to shorten scanning and evaluation time to screen large populations. The aim of
our study is to determine the lesion detection rate with the abbreviated MRI protocol in patients with a histopathological diagnosis of
pure DCIS (ductal carcinoma in situ without any invasive components) in the breast
Methods or Background: Patients who underwent breast MRI between May 2021 and October 2023 in our institution were reviewed
retrospectively by two radiologists (a radiologist with nine years of breast MRI experience, and a radiology resident with two years of
MR experience). All the MRIs were carried out using a 1.5T MRI machine (GE Optima MR450w). Patients diagnosed with DCIS by
biopsy were included in the study. Cases who had a history of breast cancer, had invasive tumour foci in addition to DCIS, and cases
who received neoadjuvant chemotherapy, were excluded from the study. Abbreviated MRI protocol includes T1WI without contrast,
T1WI with first minute contrast, and one minute subtraction image.
Results or Findings: A total of 36 DCIS diagnosed by biopsy were evaluated in our centre. Eleven cases were excluded from the
study because they had invasive tumour foci in addition to DCIS. Except for five of the 25 pure DCIS, the others showed contrast
enhancement in the first minute of contrast and subtraction images (sensitivity 80.0%). Of the five cases, one case had low-grade
DCIS, two cases had intermediate-type DCIS, and two cases had high-grade DCIS. The mean age of the patients included in the study
was 51.72 (39-80) years.
Conclusion: Pure DCIS could be detected with the high sensitivity abbreviated MRI protocol.
Limitations: The limited number of patients and the retrospective nature of the research are limitations of this study.
Ethics committee - additional information: The study was approved by Umraniye Reserch and Training Hospital.
Abbreviated breast MRI protocols in the screening of high-risk females (7 min)

Nada Mohamed Kotb; Cairo / Egypt
Author Block: M. A. Adel, N. M. Kotb; Cairo/EG

Purpose: This study aims to decrease scanning time with the abbreviated MRI protocol and to achieve the same high level of cancer
detection while providing greater efficiency, improved patient tolerance of the examination, and substantial resource savings.
Methods or Background: This study was carried out on high-risk women with a lifetime risk of over 20%. The majority of the
examinations were performed on women with previous history of breast cancer (63.3%), while 36.7% had moderate to high family
history. An abbreviated protocol was created from the sequences obtained from the full diagnostic protocol, which consisted of a pre-
contrast T1, postcontrast T1, subtraction and MIP images.
Results or Findings: On MRI examination using the conventional protocol, 36% of the women in our study were diagnosed with
breast mass lesions, of which 20% were malignant masses. Benign findings, whether mass or non-mass lesions, constituted a total of
50.1%, and about 26.7% were normal. All the malignant lesions were identified on the abbreviated protocol by the designated reader.
Conclusion: The use of abbreviated breast MRI is feasible without compromising the cancer detection rate (compared to a full
protocol). It also achieved a significant reduction in scanning time, interpretation time, and total cost of the study, proving its efficacy
and making it a more accessible first-line in the screening of high-risk females.
Limitations: There were many exclusions to participation, including: patients who had a cardiac pacemaker, aneurysmal clippings,
cochlear implants, intracranial aneurysmal clips, ferromagnetic surgical clips/staples, hearing aids, metallic shrapnel and/or bullets,
metal foreign bodies in any region, patients with raised kidney functions or suffering from contrast allergy, pregnant females in the
first trimester or lactating females, male patients and patients disinclined to perform the study.
Funding for this study: The cancer detection rate in our study using the abbreviated MRI protocols revealed a Kappa agreement
test of 1.000, i.e., perfect agreement with the conventional protocols for female participants.
Abbreviated vs full protocol in breast MRI: a meta-analysis comparing their diagnostic performances (7 min)
Ottavia Battaglia; Milan / Italy
760
Author Block: O. Battaglia, F. Pesapane, A. Rotili, O. D'Ecclesiis, S. Penco, G. Signorelli, L. Nicosia, S. Gandini, E. Cassano; Milan/IT
Purpose: In the last decade, the interest in magnetic resonance imaging abbreviated protocols (AB-MRI) has grown as shorter scan
times translate into cost savings, enhanced patient tolerance and improved accessibility.
The aim of this meta-analysis is to compare the diagnostic performances of AB-MRI and full protocol MRI (FP-MRI) in breast cancer
assessment.
Methods or Background: We conducted a systematic review and meta-analysis from 01/11/2019 up to 12/12/2022. A biavariate
model was used to calculate the summary estimates of sensitivity and specificity. Random effect models were used to calculate
summary AUC and 95% CIs. Probability distributions for negative and positive predictive values were obtained.
Results or Findings: Out of 367 studies, a total of 11 studies were included. The summary sensitivity estimated from the AB-MRI
compared to the FP-MRI is slightly lower (86.2%, 95% CI: 82.1-89.6 and 93.7%, 95% CI: 88.7-96.6, respectively). Summary estimates
(SEs) of specificity are similar (79.7%, 95% CI: 73.2-84.8 for AB-MRI and 78.0%, 95% CI: 70.6-84.0 for FP-MRI). A better accuracy has
emerged for FP-MRI (84.4% for AB-MRI and 87.6% for FP-MRI). The SEs of positive predictive value is similar (22.3% for AB-MRI and
20.4% for FP-MRI).
Conclusion: AB-MRI protocols have emerged as a potential alternative to overcame FP-MRI limits, offering increased efficiency and
cost-effectiveness by reducing acquisition and reading times, improving accessibility, and enhancing the patient experience with
similar diagnostic performances.
Further research is needed to optimise the AB-MRI and to better define its indications. This includes investigating the role of machine
learning algorithms to enhance lesion detection and characterisation within the AB-MRI. Long-term outcome studies are needed to
evaluate AB-MRI’s impact on patient outcomes, including detection rates and interval cancers.
Limitations: An identified limitation was the heterogeneity of the proposed AB-MRI protocols in the included studies.
Enhancing the Kaiser score for unenhanced breast MRI (7 min)

Nina Pötsch; Vienna / Austria
Author Block: N. Pötsch, P. Clauser, P. Kapetas, T. H. Helbich, P. A. Baltzer; Vienna/AT

Purpose: Unenhanced abbreviated breast MRI protocols aim to reduce scanning time and costs while enhancing access to breast
MRI. However, in order to further exploit the advantages of the Kaiser score, a well-established clinical decision rule for breast MRI, it
must be adapted to unenhanced protocols.
Methods or Background: We studied 120 patients who underwent breast MRI between 2021 and 2023 for further evaluation of
suspicious or equivocal findings in conventional imaging (BI-RADS 0, 3-5 in mammography and ultrasound) or for staging purposes
(BI-RADS 6). All patients underwent biopsy for histologic verification or were followed for a minimum of 12 months to confirm lesion
stability. MRI scans were conducted using 1.5T or 3T scanners using dedicated breast coils and a protocol in line with international
recommendations additionally including DWI and ADC. Lesion characterisation relied solely on T2w and DWI/ADC-derived features
(such as lesion type, margins, shape, internal signal, surrounding tissue findings, ADC value). Statistical analysis was done using
decision tree analysis with benign (B2, B3) and malignant (B5a/b) as dependent variables.
Results or Findings: We analysed a total of 161 lesions, including 81 non-mass lesions, with a malignancy rate of 40%. The lesion
margins (spiculated, irregular, or circumscribed) were identified as the most critical criterion within the decision tree. Further
subclassification considered the ADC value as the second most important criterion. The resulting score demonstrated a strong
diagnostic performance with an AUC of 0.840, providing clear rule-in and rule-out criteria.
Conclusion: Lesion characterisation in unenhanced MRI is feasible, with lesion margins and ADC value as the most important criteria.
Limitations: An identified limitation was that our analysis did not address lesion detection.
Ethics committee - additional information: Informed consent was waived.
Role of ultra-fast breast MRI in characterisation of breast masses: a step toward a faster MRI technique (7 min)
Rasha Karam Mahmoud Mohammed; Mansoura / Egypt
761
Author Block: R. Karam Mahmoud Mohammed, D. M. Bayoumi, F. A. Shokeir; Mansoura/EG
Purpose: To evaluate the efficacy of parameters derived from ultra-fast breast MRI, maximum slope (MS), initial enhancement phase
(IE phase) and time to enhancement (TTE), in differentiation of benign and malignant breast lesions.
Methods or Background: This is a prospective study that included 264 patients with 273 breast lesions, from which 120 were
pathologically proven to be benign and 153 were pathologically proven to be malignant. An ultra-fast study was performed between
the pre-contrast study and the first postcontrast sequence in the dynamic protocol using the TWIST sequence without fat suppression.
The TWIST sequence was composed of a pre-contrast phase followed by 15 postcontrast TWIST series with a temporal resolution of
4.32 seconds each. We put a ROI at the hot spot of the lesion on subtracted images and a wash-in curve was automatically generated.
From the ultra-fast wash in curve parameters were manually calculated (TTE and MS).
Results or Findings: TTE was significantly lower in malignant lesions compared to benign ones (median=8 seconds [0-20] versus 20
seconds [8-40], respectively. P value<0.001). However, MS was higher in malignant lesions compared to benign ones (median=15
%/s [2-140] versus 5 %/s [1-17] respectively. P value=<0.001).
TTE cut-off value of 15.5 seconds showed a sensitivity of 78.4%, specificity of 80%, AUC of 0.876, and accuracy of 79%. However, MS,
a cut off value of 7.25 %/s showed sensitivity of 82.4%, specificity of 70%, AUC of 0.86 and accuracy of 76.9%.
Conclusion: Ultra-fast breast MRI can be used as a faster and accurate MRI technique for differentiation between benign and
malignant breast lesions.
Limitations: Identified limitations were: (1) the relatively small number of cases and (2) the fact that the ultra-fast technique was not
compared to the conventional dynamic MRI technique.
Ethics committee - additional information: The study was approved by our institution’s ethics committee (Mansoura Faculty of
Medicine Institutional Research Board, reference number: R.23.05.2174). Written informed consent was obtained from all patients
included in the study.
Application of the Kaiser score in the assessment of suspicious contrast-enhancing lesions on breast MRI in patients
with equivocal conventional imaging findings (7 min)
Katarina Obradovic; Belgrade / Serbia
Author Block: K. Obradovic, I. Petkovic, D. Pavlovic Stankovic, V. Urban, Z. C. Milosevic; Belgrade/RS

Purpose: To assess whether applying the Kaiser score might downgrade breast MRI BI-RADS categories in patients with equivocal
conventional imaging findings (mammography and/or US) and avoid unnecessary biopsies.
Methods or Background: Equivocal conventional imaging findings are one of the indications for problem-solving breast MRI.
Generally, a combination of high sensitivity (98–100%) and suboptimal specificity (up to 88%) is inherent to breast MRI, which
frequently results in false-positive diagnoses.
Our single-centre study enrolled 61 consecutive patients from January 2017 to December 2018, with equivocal conventional imaging
findings and 69 breast lesions on MRI (62 lesions in BI-RADS category 4, seven in BI-RADS category 5), undergoing US-CNB and/or
surgery. All MRI examinations were re-evaluated after 4 years of the patients' follow-up, using the Kaiser score by two breast imaging
radiologists and one radiology resident, individually and blinded to the pathological diagnosis and previous reading results, with the
consensus about different opinions through consultation.
Results or Findings: Histopathology revealed 18 malignant and 51 benign lesions, resulting in a predominance of benignancy of
73.9%. The mean size of lesions on MRI was 15±10 mm; 50 lesions presented as masses (72.5%), 19 as a non-mass enhancement
(27.5%), and there was no difference between malignant and benign lesions. The Kaiser score was as follows: 1–4 (BI-RADS 2/3) in 48
lesions (69.6%), 5–7 (BI-RADS 4) in 11 (15.9%), and 8–11 (BI-RADS 5) in ten (14.5%) (Figs. sensitivity and specificity were 100% and
94.1%, respectively, with three false-positive findings and no false-negative findings. The AUC was 0.971 (95% CI: 0.932–1.000).
Conclusion: Applying the Kaiser score to breast MRI, BI-RADS categories are reclassified (particularly BI-RADS 4 to BI-RADS 2/3),
which can obviate unnecessary biopsies in breast lesions with equivocal conventional imaging findings.
Ethics committee - additional information: The study was approved by the Institute of Oncology and Radiology of Serbia:
1140-01
Diffusion-weighted breast MRI: is there a role in the surveillance of BRCA mutated carriers? (7 min)
Anna Rotili; Milan / Italy
762
Author Block: A. Rotili, G. Signorelli, F. Pesapane, S. Penco, A. C. Bozzini, V. Dominelli, E. Cassano; Milan/IT
Purpose: We determined the diagnostic performances of diffusion-weighted imaging (DWI) in asymptomatic women with high-risk of
breast cancer.
Methods or Background: Asymptomatic women with breast cancer susceptibility gene (BRCA)1/2 mutations who underwent breast
MRI in a single centre from January 2019 to December 2021 were retrospectively evaluated. A radiologist with experience of breast
imaging (R1) and a radiology resident (R2) independently evaluated DWI/ADC map and, in case of doubts, T2-WI. The standard of
reference was the pathological diagnosis through biopsy or surgery, or ≥1 year of clinical and radiological follow-up. Diagnostic
performances were calculated for both readers with a 95% confidence interval (CI). The agreement was assessed using Cohen’s
kappa (κ) statistics.
Results or Findings: Of 313 women, 145 women were included (49.5±12 years), totalling 344 breast MRIs with DWI/ADC map. Per-
exam cancer prevalence was 11/344 (3.2%). Sensitivity was 8/11 (73%; 95% CI: 46-99%) for R1 and 7/11 (64%; 95% CI: 35-92%) for
R2. Specificity was 301/333 (90%; 95% CI: 87-94%) for both readers. Diagnostic accuracy was 90% for both readers. R1 recalled
40/344 exams (11.6%) and R2 recalled 39/344 exams (11.3%). Inter-reader reproducibility between readers was moderate agreement
(κ=0.43).
Conclusion: In women with the BRCA1/2 mutation, breast DWI supplemented with T2-WI allowed breast cancer detection with high
sensitivity and specificity by a radiologist with a high level of experience of breast imaging, comparable to other screening tests.
Limitations: Our analysis was performed at a single tertiary-level cancer care institution, with dedicated breast equipment and
radiologists, so the results may not be generalisable to general practice.
Another identified limitation was the small size of the lesions.
Ethics committee - additional information: Institutional Review Board approval was obtained for this retrospective study and all
patients signed written informed consent before breast MRI was performed.
Classification of MRI-only suspicious breast lesions: development and validation of a nomogram combining MRI
features and apparent diffusion coefficient histogram (7 min)
Xue Li; Beijing / China
Author Block: X. Li, M. Chen; Beijing/CN

Purpose: Up to half of suspicious breast lesions (BI-RADS category 4 or 5) only visible on MRI (MRI-only lesions) are false-positive
results and undergo unnecessary biopsies. To this end, we aimed to develop and validate a nomogram combining clinicoradiologic
features and ADC-based histogram parameters to improve the diagnostic accuracy of MRI-only suspicious lesions.
Methods or Background: Ninety patients who underwent breast MRI and were found to have MRI-only suspicious lesions were
retrospectively included and randomly divided into training (n=62) and validation (n=28) cohorts. The clinical information and MRI
features of each patient were reviewed and analysed. The ADC maps obtained from DWI of each patient were subjected to histogram
analysis, and then 17 histogram parameters were extracted. Univariate and multivariate logistic regression analyses were performed
to identify significant variables associated with predicting MRI-only malignant lesions, which were then included in the nomogram. The
diagnostic performances of significant variables and the nomogram were evaluated and compared in terms of the ROC curve and
DeLong’s test.
Results or Findings: The kinetic pattern in clinicoradiologic features (P=0.004, odds ratio [OR]=2.268) and ADC entropy in
histogram parameters (P=0.003, OR=6.49) were significant variables associated with the classification of MRI-only suspicious lesions.
Combined ADC entropy and kinetic pattern yielded an AUC of 0.820, sensitivity of 80.65%, and specificity of 77.42%, which were
higher than any single variable alone. The C-index values for the nomogram were 0.820 for the training cohort and 0.728 for the
validation cohort.
Conclusion: ADC-based histogram parameters facilitated the classification of MRI-only suspicious lesions. The nomogram combining
kinetic pattern and ADC entropy can be used as a simple and non-invasive tool for classifying MRI-only suspicious lesions.
Ethics committee - additional information: The Institutional Review Board of our hospital approved this retrospective single-
centre study (number: 2022BJYYEC-388-01) and waived the requirement for informed consent.
Breast multiparametric MRI (MP-MRI) to discriminate between pure ductal carcinoma in situ (DCIS) and microinvasive
carcinoma (MIC): the importance of DWI (7 min)
Federica Cicciarelli; Latina / Italy
763
Author Block: F. Cicciarelli1, V. Rizzo2, F. Galati3, M. Pasculli3, R. Maroncelli4, F. Pediconi3, C. Catalano3; 1Latina/IT, 2Maglie/IT, 3Rome/IT,
4
Guidonia/IT
Purpose: DCIS is a common subtype of breast cancer that is often upgraded to invasive cancer on the final pathology report of the
surgical specimen. DCIS has an upgrade rate ranging in MIC from less than 10% to more than 40%. In these cases, histologic upgrade
should be considered before surgical planning, to include sentinel lymph node biopsy in view of possible axillary lymph node
metastasis.
The aim of our study is to investigate whether breast multiparametric magnetic resonance imaging (MP-MRI) can distinguish between
pure DCIS and MIC or invasive ductal cancer (IDC).
Methods or Background: Between January 2018 and November 2022, patients with biopsy-proven DCIS who performed
preoperative breast MP-MRI on 3T magnet were enrolled in this retrospective study. MP-MRI features (size, morphology, margins,
extent of disease, presence of peritumoural oedema and curve I/T value, type and distribution of enhancement, restriction of diffusion
and ADC value) were evaluated. For categorical variables chi-square test was assessed to investigate the association with the
pathological outcome. Logistic regression analysis was performed to evaluate their prognostic value.
Results or Findings: Out of 129 patients with biopsy-proven DCIS, 36 presented foci of micro-infiltration on surgical specimens and
eight was IDC. The presence of micro-infiltration foci was significantly associated with several MP-MRI features, such as tumour size
≥2 cm (p=0.02), clustered ring enhancement (p<0.001), and segmental distribution (p<0.01). Interestingly, a mean ADC value <1.3
x 10-3 mm2/s revealed to be a prognostic factor for the presence of micro-infiltration foci (p< 0.05).
Conclusion: Breast MP-MRI can predict the presence of micro-infiltration foci in biopsy-proven DCIS, and could be considered a valid
tool for therapeutic planning in patients with DCIS.
Limitations: The relativity small number of patients was an identified limitation.
Screening without mammography using abbreviated breast MRI alone (7 min)

Stephanie Morscheid; Aachen / Germany
Author Block: S. Morscheid, M. Bode, L. C. Huck, K. Strobel, C. K. Kuhl; Aachen/DE

Purpose: Breast MRI is recommended for supplemental screening for women with dense breasts. However, in women undergoing
screening MRI, the additional cancer detection rate attributable to mammography has been shown to be limited. Based on recently
published cost-effectiveness analyses, using MRI alone for screening would represent the most cost-effective screening method. The
purpose of this study was to report first results on the outcome of women undergoing breast MRI as a stand-alone screening test
(without mammography).
Methods or Background: This is an ongoing study which has so far enrolled 1,085 women (aged 40-85 years, median 57 years)
without personal history of breast cancer who underwent screening using abbreviated MRI alone. Breast MRI was performed on a 1.5T
system with multichannel-coil, according to a standardised protocol. Validation of MRI diagnoses was achieved by follow-up
(977/1,085) or biopsy (108/1,085). A further 54 biopsies were performed within the screening period.
Results or Findings: Women underwent a total 2,771 screening MRIs (mean 2.6 studies/participant), covering a total of 3,220
women-years and a subsequent follow-up period of 2,425 women-years. MRI was positive in 392/2,771 examinations (14.1%). Breast
cancer was confirmed in 67, yielding a PPV3 (of biopsy) of 41.4%. The overall cancer detection rate was 20.8 per 1000 women-years.
The median age of women diagnosed with cancer was 59 years. Cancers were DCIS in 18/67 and invasive cancers in 49/67. Of the
invasive cancers, cancers were staged pT1a in 7/49, pT1b in 21/49, pT1c in 16/49, pT2 in 5/49, and as N0 in 42/49 (85.7%); N1 and
N1mi in the remaining cases; all were staged M0. 60/67 patients with MRI-detected cancer underwent bilateral preoperative
mammography. The MRI-known cancer was occult on mammography in 40/60 (67%).
Conclusion: Abbreviated breast MRI alone appears sufficient for breast cancer screening, ensuring early cancer detection.
Limitations: The pseudo-prospective study design was an identified limitation.
Funding for this study: No external funding was received for this study.
Ethics committee - additional information: This study was conducted in accordance with the Declaration of Helsinki and was
approved by the Institutional Review Board of RWTH University Aachen.
Looking beyond standard dosage in dynamic contrast enhanced breast MRI using a pre-trained deep-learning model (7
min)
Srivathsa Pasumarthi Venkata; Santa Clara / United States
764
Author Block: S. Pasumarthi Venkata1, P. Gulaka1, G. Zaharchuk2, R. Chamberlain1; 1Menlo Park, CA/US, 2Stanford, CA/US
Purpose: Gadolinium-based contrast agents (GBCAs) are widely used in breast MRI. Increasing GBCA dosage improves lesion
visualisation but safety concerns have been raised regarding long-term GBCA retention. In this work, we repurposed a dose reduction
deep-learning (DL) model to synthesise contrast boosted (CB) breast MRI images that have better lesion visualisation without
increasing dosage.
Methods or Background: A DL model for dose reduction in brain MRI was previously trained to enhance 10% to 100% dose. This
model was repurposed for boosting contrast in breast MRI by inputting pre-contrast and standard postcontrast (SC) sequences to
synthesise contrast boosted (CB) sequences. Dynamic contrast enhanced (DCE) breast MRI data from 60 patients (30 internal, 30
Duke Breast MRI public dataset) were used to evaluate CB algorithm. CE images were normalised and co-registered to pre-contrast
sequences, and the first time point post-injection was used as the postcontrast. Axial images were manually cropped to remove
signals posterior to the pectoral muscles. Left and right sides were processed individually by the model. Lesion-to-background ratio
(LBR) and contrast enhancement percentage (CEP) metrics were calculated on SC and CB images. ROIs were manually drawn for
calculating quantitative metrics. Enhancement kinetic curves (time post-injection versus signal-intensity [SI]) were drawn with pre-
contrast, SC and other CE time points along with the average SI of CB images.
Results or Findings: The mean LBR and CEP of CB (9.63±0.53 and 182.3%) were significantly better (p<0.001) than those of SC
images (6.73 ± 0.34 and 93.4%). It was also observed that the DL model improves the lesion enhancement and visibility.
Conclusion: Quantitative and qualitative assessment has shown the feasibility of DL based contrast boosting in breast MRI to
improve lesion visualisation.
Limitations: Detailed clinical evaluation required.
765
RC 1814 - The role of radiographers in translating research into clinical practice

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderators:
Bo Redder Mussmann; Odense C / Denmark
Gertraud Heinz; St. Pölten / Austria

Appraising the evidence: key skills and tips for radiographers (12 min)
Nejc Mekis; Ljubljana / Slovenia
1. To describe radiographic research and its relevance to clinical practice.

2. To explore strategies for effectively incorporating evidence-based findings into radiographic practice.
3. To discuss and disseminate results.
Using the "complex interventions framework" to implement research (12 min)

Karen Knapp; NEWTON ABBOT / United Kingdom
1. To outline the framework for developing and evaluating complex interventions.

2. To consider how radiography research can fit within the complex interventions framework.
3. To describe how using structured research frameworks can enhance radiography research.
Strategies to disseminate research results to influence change (12 min)

Louise A. Rainford; Dublin / Ireland
1. To discuss the importance of effective research dissemination.

2. To explore diverse strategies for disseminating research results.
3. To list top tips for impactful research dissemination.
Tips for adopting research evidence into radiography practice (12 min)
1. To define evidence-based research.

2. To learn how to apply effective methods in disseminating evidence-based research findings.
3. To address the challenges encountered in implementing evidence-based research findings into practice.
Panel discussion: How can radiographers bring evidence-based practice into clinical departments (10 min)
766
RC 1806 - Non-FDG PET tracers in oncologic imaging
Categories: Hybrid Imaging, Molecular Imaging, Oncologic Imaging, Translational Imaging

ETC Level: LEVEL II
Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Michel Eisenblaetter; Detmold / Germany

Michel Eisenblaetter; Detmold / Germany
PSMA-PET: state-of-the-art and current developments (15 min)

Daniela Elena Oprea-Lager; Amsterdam / Netherlands
1. To become familiar with the development of the PSMA-PET tracers.

2. To get insight into the different radiolabelled PSMA-PET tracers.
3. To assess the rationale of PSMA theranostics in prostate cancer
FAPI-PET: a new workhorse in oncologic imaging? (15 min)

Frederik L Giesel; Heidelberg / Germany
1. To learn about the radiopharmaceutical basics of FAPI-PET imaging.

2. To learn about current clinical and scientific applications of FAPI-PET imaging.
3. To discuss ongoing developments and future directions.
SSR-PET: imaging of neuroendocrine tumours and beyond (15 min)

Luigi Aloj; Cambridge / United Kingdom
1. To learn about the radiopharmaceutical basics of SSR-PET imaging.

2. To learn about current clinical and scientific applications of SSR-PET imaging.
3. To discuss ongoing developments and future directions.
Panel discussion: When and how? Choosing the right tracer for the right patient (10 min)
767
RC 1804 - Back to basics: diffuse lung diseases
Categories: Chest
ETC Level: LEVEL II
Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Anna Kerpel-Fronius; Budapest / Hungary

Anna Kerpel-Fronius; Budapest / Hungary
Navigating the secondary pulmonary lobule (12 min)

Eva Kocova; Hradec Kralove / Czechia
1. To learn about anatomy to CT imaging correlation.

2. To understand the basic imaging patterns in CT lungs.
3. To understand the current limitations of CT imaging in ILD.
How to approach nodular lung diseases (12 min)

Lukas Ebner; Bern / Switzerland
1. To recognise various nodular patterns on lung imaging.

2. To understand the limitations of imaging in nodular lung diseases.
3. To learn the differential diagnosis of nodular patterns.
How to approach reticulation on HRCT (12 min)

Anand Devaraj; London / United Kingdom
1. To recognise various linear patterns on lung imaging.

2. To understand the limitations of imaging in linear lung diseases.
3. To learn the differential diagnosis of linear patterns.
Diagnosis of cystic lung diseases made easy (12 min)

1. To understand various presentations of cysts on imaging.

2. To learn when cysts are incidental and when they are pathological.
3. To learn the differential diagnosis of cystic lung diseases.
Panel discussion: The importance of a systematic approach in diffuse lung diseases (8 min)
768
PC 18 - Embracing clinical care
Categories: Interventional Radiology, Management/Leadership, Multidisciplinary, Oncologic Imaging, Professional Issues

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

Strategy for DR and IR to be visible for the patients (12 min)

1. To understand the current situation in Europe.

2. To understand the importance of the radiologist's visibility.
3. To define possible approaches to improve visibility.
Organisation of an outpatient clinic (12 min)

1. To understand the requirement of a suitable infrastructure for an outpatient clinic.

2. To understand the importance of dedicated time to consulting patients.
3. To identify the necessary personnel to run an outpatient clinic sustainably.
In-patient clinical services (12 min)

1. To discuss the rationale in supporting admitting rights for IRs, highlighting advantages for patients, IRs, and hospitals.
2. To highlight service requirements and infrastructures for elective IR in-patient admissions.
3. To discuss clinical knowledge needed for patient management and principal team-based training options.
The role of DR and IR at tumour boards (12 min)

Otto M Van Delden; Amsterdam / Netherlands
1. To learn about the role of DR at multidisciplinary tumour boards.

2. To learn about the role of IR at multidisciplinary tumour boards.
3. To learn about ways to increase the visibility and impact of DR and IR at multidisciplinary tumour boards.
Value proposition of clinical care in IR (12 min)

Charles E. Ray Jr.; Chicago, IL / United States
1. To understand the importance of non-procedural clinical care for patients undergoing interventional radiology procedures.
2. To recognise the financial implications of providing non-procedural clinical care.
3. To identify strategies for discussing the importance of non-procedural clinical care to hospital administrators and third-party
payers.
769
Panel discussion: How to change the image of IR from a technical service to a clinical discipline? (25 min)
770
E³ 1819 - MR: the future is now

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Jürgen Hennig; Freiburg / Germany

Jürgen Hennig; Freiburg / Germany
Low-field MR: a renaissance (25 min)

Elmar M. Merkle; Basel / Switzerland
1. To describe current technology in low-field MRI.

2. To examine the advantages of using low-field MRI and in which diagnostic setting.
3. To analyse possible limitations of low-field MRI.
MR beyond 3T: technical challenges and current applications beyond neuroimaging (25 min)
Tom Scheenen; Nijmegen / Netherlands
1. To describe the technical evolution of ultra-high field MRI.

2. To examine the advantages of ultra-high field MRI in body imaging.
3. To analyse possible limitations and how to overcome them.
AI in MR examinations: no more compromise between time and quality (25 min)

Marta Zerunian; Rome / Italy
1. To describe current AI technology implemented in MR imaging.

2. To evaluate the impact of AI in image quality and scanning time.
3. To describe AI advantages in contrast media reduction.
Discussion (10 min)
771
SF 18b - Incidental lesions in children
Categories: Abdominal Viscera, Chest, Musculoskeletal, Neuro, Paediatric

ETC Level: LEVEL II
Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Karen Rosendahl; Tromsø / Norway
Neuroimaging (17 min)

1. To acknowledge developmental changes during maturation of the brain, as assessed on ultrasound, CT or MRI.
2. To recognise the most common incidental findings in immature brains.
3. To understand if and when to follow up.
Thoracic imaging (17 min)

Maria Raissaki; HERAKLION / Greece
1. To recognise the most common incidental findings in chest imaging, as assessed radiographically, by CT or MRI.
2. To understand when to follow up and how to guide the clinicians.
Abdominal imaging (17 min)

Simon G. F. Robben; Maastricht / Netherlands
1. To identify the more common incidental findings in abdominal imaging, as assessed on radiographs, ultrasound CT or MRI.
2. To acknowledge if and when to follow up.
Musculoskeletal imaging (17 min)

Ignasi Barber; EspluguesdeLlobregat / Spain
1. To recognise developmental changes of the skeleton during maturation.

2. To identify the more common incidental findings.
3. To acknowledge the need for, or lack of the need for, a follow-up.
Panel discussion: How do we report incidental findings? (17 min)
772
OF 18R - Latest advances and considerations in radiation therapy and oncology
Categories: Hybrid Imaging, Imaging Methods, Multidisciplinary, Oncologic Imaging, Radiographers

Date: March 2, 2024 | 09:30 - 10:30 CET
CME Credits: 1
This session aims to provide a comprehensive overview of the evolving landscape in radiation therapy and oncology. It encompasses
three pivotal talks, each shedding light on cutting-edge technologies, challenges, and quality control considerations that
radiographers must navigate in their pursuit of excellence in cancer care. Indeed, the session will include talks relating to some of the
latest advances and considerations in radiation therapy and oncology. Whether you are a seasoned practitioner or a newcomer to the
field, this session offers invaluable knowledge and practical insights that will empower radiographers to provide the highest quality
care in the dynamic landscape of cancer treatment.
Moderator:

Proton therapy: what radiographers need to know (16 min)

Brigitte Florentine barones van Asbeck; Utrecht / Netherlands
Opportunities and challenges of MR-guided adaptive radiotherapy (16 min)

Cynthia Eccles; Buxton / United Kingdom
Keeping up to date with the latest QC for RT equipment (16 min)

Ainars Bajinskis; Riga / Latvia
773
SF 18a - Cardiac CT in the emergency department
Categories: Cardiac, Chest

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

Which patients benefit from cardiac CT in the emergency department and how do we do it? (20 min)
1. To understand spectrum of current clinical indications to CCTA in emergency departments.

2. To review recommended scanning protocols according to the on-site available technology.
3. To comprehend drawbacks of using CT as first line tool in acute chest pain.
Do we need CT-FFR and CT-perfusion in the emergency department? (20 min)

1. To review scientific evidences supporting the advantages of a functional vs an anatomic approach for in acute chest pain patients.
2. To learn principles and technical optimisation of a CT perfusion protocol in emergency setting.
3. To understand impact of CTP/FFR results on patient's workflow and clinical decision-making.
What does plaque analysis add in acute chest pain patients? (20 min)
1. To understand the importance of plaque composition as a predictor of events in acute chest pain patients.
2. To review CCTA features of stable vs. unstable plaques, including the added value of artificial intelligence and radiomics evaluation.
3. To analyse whether atherosclerotic burden quantified with calcium score could be used in acute chest pain patients.
Panel discussion: Are we ready for 24/7 cardiac CT in the emergency department? (25 min)
774
MS 18 - Muscle-tendon tears: from the basis to the frontiers of imaging

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Xavier Tomas Batlle; Barcelona / Spain

Muscle-tendon tears: a sports and exercise medicine specialist's point of view (15 min)
Gil Rodas; Cardedeu / Spain
1. To manage MTT: from genetics to return-to-play (RTP).

2. To analyse how OMICS and RADIOMICS Biomarkers could help us to prevent, diagnose and treat MTT.
3. To explain what radiologists should know about regenerative medicine for MTT: PRPs, cellular therapy, and scaffolds.
MRI approach to acute muscle injuries and the impact of imaging on RTP estimation (15 min)
Sandra Mechó Meca; Barcelona / Spain
1. To characterise acute muscle injuries by MRI.

2. To recognise the radiological signs that may or may not be associated with a prolonged RTP.
MRI approach during the repair process after a muscle injury and estimation of reinjure risk (15 min)
Jaime Isern Kebschull; Barcelona / Spain
1. To characterise the healing process after a muscle injury by MRI.

2. To recognise MRI signs related to the risk of re-injury.
Imaging of muscle injuries in elite and non-elite athletes (15 min)

Fernando Idoate Saralegui; PAMPLONA / Spain
1. To discuss imaging strategies: role of MRI and US.

2. To discuss current trends and possible future directions in assessing muscle injuries.
3. To put it all together: the decision-making process.
Case discussion: Let's go to work (25 min)

775
HW 18Cb - Exploring cardiomyopathies: imaging insights and reporting techniques
Categories: Cardiac, Imaging Methods, Multidisciplinary

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:


1. To become familiar with typical and atypical imaging findings of most common cardiomyopathies.
3. To discuss the limits and technical drawbacks of cardiac MRI.
4. To learn how to report cardiac MRI using specific templates.
776
RPS 1808 - Thyroid and salivary gland imaging
Categories: Artificial Intelligence & Machine Learning, Head and Neck, Imaging Methods, Interventional Radiology
Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Role of superb microvascular imaging for differentiation between benign and malignant neck lymph nodes (7 min)
Hye Ree Cho; Changwon / Korea, Republic of
Author Block: H. R. Cho, M. Sunwoo, Y. Nam; Changwon-si, Gyeongsangnam-do/KR

Purpose: Ultrasound examination reveals different types of disease that cause lumps in the cervical lymph nodes. However, different
types of disease can produce similar ultrasound findings, and despite the criteria that have been proposed to distinguish between
them, greyscale ultrasound is still limited. Therefore, methods are being used to analyse vascular patterns using colour or power
Doppler ultrasound. However, benign and malignant vascular patterns still overlap, and the use of superficial microvascular imaging
techniques that detect microvasculature may help distinguish them.
Methods or Background: From January to August 2023, 163 patients who received FNAB or biopsy for enlarged neck lymph nodes
were included. All patients underwent US examinations of LNs consisting of greyscale US, PDUS and SMI. After sonography, each
lymph node was evaluated histologically by ultrasound-guided FNAB or biopsy.
Results or Findings: Out of 163 patients, 18 patients were excluded due to non-diagnostic results. Of the 145 total lymph nodes, 49
were identified as malignant. SMI evaluation of nodal vascular pattern was of high sensitivity (69.4%), specificity (90.6%), and
accuracy (83.4%) for differentiating metastatic and benign nodes, however SMI alone does not have a significant advantage in
differentiating metastatic nodules compared to greyscale or PDUS. There were 38 indeterminate nodes on the greyscale Doppler
image, of which five were histologically confirmed to be malignant. Of 38 indeterminate nodes, 23 were reclassified as benign by SMI,
all of which were confirmed as benign.
Conclusion: Using a nodal classification based on a combination of sonographic features and SMI is effective in detecting and
diagnosing metastatic lymphadenopathy, leading to improved treatment and prognosis for patients.
Limitations: Identified limitation were: (1) this was a short-term prospective study with a small number of cases; (2) inter-observer
agreement could not be assessed due to on-site pattern analysis.
Blood leukocyte DNA methylation biomarker for distinguishing malignant from benign thyroid nodules (7 min)
Feihang Wang; Shanghai / China
777
Author Block: F. Wang, D. Zhao, L. Liu; Shanghai/CN
Purpose: Discriminating between malignant thyroid nodules (MTNs) and benign thyroid nodules (BTNs) presents challenges in clinical
practice. Our objective was to develop a blood leukocyte-based classifier to distinguish MTNs from BTNs, and to assess its potential
for enhancing diagnostic performance.
Methods or Background: This study included 91 MTNs and 114 BTNs. Differential methylation haplotype blocks (MHBs) in blood
leukocytes between MTNs and BTNs were detected using reduced representation bisulfite sequencing (RRBS). Subsequently, a blood
leukocyte DNA methylation (BLDM) classifier was developed.
Results or Findings: A 60-marker BLDM classifier achieved an area under the curve (AUC) of 0.86 in the validation cohort. The
specificity of it was 90.91%, outperforming the 43.64% specificity of ultrasonography, albeit with a slightly lower sensitivity compared
to ultrasonography (83.33% versus 97.62%). The BLDM classifier correctly identified 93.55% of patients whose nodules were
suspected to be malignant by ultrasonography, but which were finally histologically benign. In micronodules, a higher specificity of
93.33% and accuracy of 88.24% for diagnosis of MTNs was displayed by this classifier, in comparison to the 6.67% specificity and
72.55% accuracy of ultrasonography.
Conclusion: We identified a novel BLDM classifier for distinguishing MTNs from BTNs. The clinical application of this classifier leads to
enhanced diagnostic specificity compared to ultrasonography.
Limitations: The main limitation of this study is that all participants were recruited from a single centre, and this classifier will
require validation in larger external cohorts in the future.
Funding for this study: Funding was provided by the National Health Commission Capacity Building And Continuing Education
Center (GWJJ2022100303) and Fudan University Integrated Medical Engineering Program (yg2022-6).
Ethics committee - additional information: This study was approved by the Institutional Review Committee of Zhongshan
Hospital, Fudan University (B2022-390R).
CEUS as an additional tool in differentiating hyperthyroid diffuse disease: destructive vs hyperproduction conditions (7
min)
Andreea Borlea; Timisoara / Romania
Author Block: A. Borlea, D. I. Stoian; Timisoara/RO

Purpose: This study aims to evaluate the efficacy and added value of contrast-enhanced ultrasound (CEUS) in differentiating
between destructive thyroiditis and hyperproduction states in patients presenting with diffuse thyroid disease.
Methods or Background: Hyperthyroid diffuse diseases present as multifaceted pathological entities, often categorised into
destructive thyroiditis and hyperproduction-induced hyperthyroidism. A prospective cohort of 56 patients diagnosed with hyperthyroid
diffuse disease underwent CEUS, alongside conventional ultrasound and other standard diagnostic investigations. Qualitative CEUS
parameters, including perfusion patterns, enhancement intensity, and washout timing, were analysed and compared between the
destructive thyroiditis group and the hyperproduction group.
Results or Findings: Preliminary results indicated distinctive CEUS perfusion patterns between destructive thyroiditis and
hyperproduction-induced hyperthyroidism. Destructive thyroiditis demonstrated rapid enhancement (12 seconds, 10-14.5) followed
by a swift washout (20 seconds, 18-26), reflective of increased vascular permeability and inflammation. In contrast, hyperproduction
states typically exhibited prolonged enhancement, suggestive of increased vascularity due to important thyroid activity. CEUS
exhibited very good diagnostic accuracy (AUC 0.905, sensitive 91%, specificity 88.3%), with an appreciable sensitivity and specificity
when integrated with existing diagnostic paradigms. Furthermore, CEUS findings correlated with biochemical markers, strengthening
its diagnostic usefulness.
Conclusion: By presenting distinct perfusion patterns and enhancement characteristics, CEUS enables a more precise imaging-based
understanding of the underlying pathophysiological processes, thereby facilitating informed diagnostic and therapeutic decision-
making.
Limitations: One primary limitation of this study resides in the relatively small sample size of 56 patients, which potentially
constrains the generalisability of the findings. While the preliminary results provide insightful correlations and trends within the
examined parameters, a more expansive, multicentric study with a larger patient cohort is crucial to validate and consolidate these
findings
Funding for this study: The study received no external funding.
Ethics committee - additional information: The study was approved by the UMF Victor Babes Timisoara Ethics Committee.
Radiomics analysis of thyroid ultrasound in relation to the radioactive iodine therapy-related sialadenitis (7 min)
Hyein Kim; Suwon / Korea, Republic of
778
Author Block: H. Kim, D. H. Lee; Suwon/KR
Purpose: Radioactive iodine therapy (RAIT) is a widely employed therapy after total thyroidectomy but is often associated with the
development of RAIT-sialadenitis. This study aims to investigate whether preoperative thyroid ultrasound can identify the occurrence
of sialadenitis in patients who have undergone total thyroidectomy followed by RAIT.
Methods or Background: This retrospective single-centre study was conducted at Ajou University Hospital, a tertiary referral centre
in Suwon, Korea. The study included patients who underwent total thyroidectomy and subsequent RAIT between January 2010 and
October 2022.
Regions of interest were drawn as 2D square regions, each with a minimum size of 100×100 point, encompassing only non-cancerous
thyroid parenchyma. Radiomics features were extracted using a commercial program (AVIEW Research, Coreline Soft, Seoul, Korea).
Radiomics features were selected through univariable logistic regression (p-value<.05). The performance of the diagnostic model in
distinguishing the occurrence of RAIT-sialadenitis was evaluated by the area under the curve (AUC) of the receiver operating
characteristic curve.
Results or Findings: The study included 62 normal patients and 20 patients with RAIT-sialadenitis. RAIT-sialadenitis was predicted
by the excess kurtosis (p-value=.003) from the first order and histogram feature, as well as contrast (p-value=.043) from the
neighbouring grey-tone difference matrix (NGTDM). The combination of the excessive kurtosis and contrast of NGTDM demonstrated
good diagnostic ability to differentiate the occurrence of RAIT-sialadenitis (AUC 0.75, 95% confidence interval: 0.64–0.87).
Conclusion: Radiomics features from thyroid ultrasound, excess kurtosis, and contrast of NGTDM, may help the prediction of RAIT-
sialadenitis and enhance patient care before RAIT.
Limitations: A relatively small number of patients were included. In addition, there was data heterogeneity derived from the
ultrasound scanner and practitioners. Furthermore, there was no pathologic confirmation of the salivary gland.
Funding for this study: This study was supported by the Ajou University (M-2023-C0460-00050).
Ethics committee - additional information: This retrospective study was approved by the Institutional Review Board of Ajou
University Hospital (approval number: AJOUIRB-MDB-2022-226). The requirement for written informed consent was waived by the
Clinical Research Review Committee of Ajou University Hospital. All procedures performed in this study complied with both the US
Health Insurance Portability and Accountability Act regulations and the Declaration of Helsinki.
Arterial spin labelling MR technique in assessing blood perfusion of thyroid nodules and differentiating the malignancy
by comparing with CT enhancement (7 min)
Yuyun Xu; Hangzhou / China
Author Block: Y. Xu; Hangzhou/CN

Purpose: To investigate the feasibility of arterial spin labelling (ASL) MR technique to evaluate thyroid and nodule blood perfusion by
comparing with enhanced thyroid CT examination, and to differentiate between benign and malignant nodules. .
Methods or Background: A retrospective analysis was conducted on data of patients with thyroid nodules who underwent routine
MRI and ASL examinations with thyroid CT enhancement from August 2022 to August 2023 in our hospital. Two radiologists
independently rated the ASL image quality, measured the TBF (thyroidal blood flow) of thyroid and nodules, as well as the CT values
of thyroid nodules in plain, arterial, and venous phases. The consistency of image quality between the two radiologists was analysed
using Kappa analysis. The consistency of TBF and CT values measured by the two radiologists for thyroid and nodules was analysed
using ICC analysis. The correlation between CT values and thyroid blood flow (TBF) in the thyroid and nodules was evaluated using
Pearson correlation.
Results or Findings: A total of 70 patients with thyroid nodules underwent thyroid CT enhancement and ASL examination. A total of
11 cases were excluded. Finally, 59 patients with a total of 64 thyroid nodules were included. The consistency of image quality scores
between the two radiologists was good (Kappa=0.745, P<0.001). The consistency of TBF measurements was ICC=0.780, 0.856,
respectively (P<0.001). The correlation between TBF of thyroid and arterial CT enhancement value was mild (r=0.283, p=0.023),
while the correlation between TBF of thyroid nodules and CT enhancement value was significant (r=0.754, P<0.01). TBF of thyroid
nodule between benign and malignant was significant (p=0.034).
Conclusion: ASL can non-invasively quantitatively evaluate the blood perfusion of the thyroid and nodules with good repeatability,
and can help differentiate between benign and malignant nodules.
Limitations: The small sample size was an identified limitation of this study.
Funding for this study: This study was supported by the Zhejiang Medicine and Health Project of Science and Technology
(2022KY581, 2022KY525).
Ethics committee - additional information: This study was conducted in accordance with the requirements of the institutional
medical ethics committee (Hangzhou Medical College, Zhejiang Provincial People's Hospital Research Ethics Committee, document
number: 202101291419000446756).
Risk stratification of incidental 18F-Fluorodeoxyglucose-avid thyroid nodules based on six societies’ ultrasound risk
stratification systems (7 min)
Chae Young Shin; Gangneung-si / Korea, Republic of
779
Author Block: C. Y. Shin1, D. G. Na1, H. S. Lee1, H. S. Ahn2; 1Gangneung-si/KR, 2Seoul/KR
Purpose: This study was aimed to determine the risk of malignancy in thyroid incidentalomas (TIs) detected on 18F-FDG PET/CT
using the six societies’ ultrasound (US)-risk stratification systems (RSSs) and to evaluate the diagnostic value of the SUVmax in PET-
TIs.
Methods or Background: This study included 215 consecutive patients with 236 TIs detected by focal thyroid uptake on 18F-FDG
PET/CT. The final diagnoses were obtained by US-guided biopsy or thyroidectomy. The malignancy risk of all TIs were assessed
according to the six US-RSSs including ACR-, Chinese (C)-, European (EU)-, and Korean-TIRADS.
Results or Findings: The overall malignancy rate of TIs was 47% and the malignancy risk of TIs was stratified by all US-RSSs (all,
P<0.001). The observed malignancy risks of high-suspicion TIs were higher than the estimated malignancy risks only in ACR-, C-, and
EU-TIRADSs (P≤0.035). The malignancy risks of intermediate-suspicion TIs were higher than the estimated malignancy risks in all US-
RSSs (24.8-48.2%, P≤0.002). The malignancy risks of low-suspicion TIs were higher than the estimated malignancy risk only in the C-
and EU-TIRADS (35% and 10%, respectively, P≤0.01). The SUVmax of malignant tumours was higher than that of benign nodules (4.7
versus 2.8, P<0.001) and the AUC for malignancy was 0.708. The SUVmax of high-grade malignancy was higher than that of low-
grade malignancy (7.4 versus 4.0, P=0.002).
Conclusion: The malignancy risk of intermediate-suspicion TIs was increased in all US-RSSs, whereas the malignancy risks of high-
and low-suspicion TIs was increased in only some RSSs. The biopsy size thresholds for intermediate suspicion TIs should be lowered in
all US-RSSs. The SUVmax showed a fair diagnostic performance for malignancy in PET-TIs and the SUVmax needs to be considered in
biopsies.
Limitations: No limitation were identified.
Ethics committee - additional information: This study is under the review of the ethics committee (number: 2023-09-007).
Quantitative evaluation of thyroid gland vascularisation with microvascular ultrasonography and contribution of
vascular index for differential diagnosis of Graves' disease and subacute thyroiditis (7 min)
Zeynep Sezgi Erdal; Ankara / Turkey
Author Block: Z. S. Erdal, M. N. Cerit, H. N. Şendur, S. Özhan Oktar, M. Yalçın, Ş. E. Doğan, M. Coşkun; Ankara/TR
Purpose: Microvascular ultrasonography (MVUS) is a new Doppler technique that provides higher sensitivity in differentiating slow
flow from motion artifacts. The vascular index obtained by MVUS measures the blood flow rate in the tissue by determining the ratio
of coloured pixels (blood flow) to all pixels in the ROI (region of interest) and gives a quantitative real time evaluation. The aim of the
present study is to test the ability of the MVUS technique in differentiating cases with Graves' disease and subacute thyroiditis.
Methods or Background: 86 cases (28 Graves' disease, 28 subacute thyroiditis, and 30 control group) were selected. Age, gender,
BMI, fT3, fT4, TSH, TRAb, ESR, and CRP values were recorded. During the thyroid gland ultrasonography, vascular indexes were
measured in the transverse and longitudinal axes in both lobes after measuring the thyroid gland dimensions. The correlations
between mean vascular index value, age, BMI, and laboratory results were assessed. ROC (Receiver operating characteristic) curve
analysis was performed to investigate the diagnostic performance of the vascular index in differentiating Graves' disease from
subacute thyroiditis.
Results or Findings: In conclusion, mean thyroid gland volume and vascular index values were significantly higher in both disease
groups than in the control group. The mean vascular index value was higher in the group with Graves' disease than in subacute
thyroiditis. Remarkably, 88% sensitivity and 92% specificity were achieved with an optimum cut-off value of 26.31% for vascular
index in differentiating Graves' disease from subacute thyroiditis.
Conclusion: Our results highlight the potential benefit of microvascular ultrasonography in the differential diagnosis of Graves'
disease and subacute thyroiditis.
Limitations: The limitation of the study is the limited number of cases. In addition, we didn't compare the MVUS with other Doppler
techniques.
Ethics committee - additional information: The study was approved by the Gazi University School of Medicine ethics committee
(Reference number 782, 24.10.2022).
Possibility to predict BRAFV600E mutations of papillary thyroid carcinoma via nomogram based on dual-layer detector
spectral CT and clinical characteristics (7 min)
Dan Zhang; Chongqing / China
780
Author Block: D. Zhang; Chongqing/CN
Purpose: The aim of this study was to investigate the effectiveness of a nomogram based on dual-layer detector spectral computed
tomography (DLCT) for predicting BRAFV600E mutation in PTC.
Methods or Background: The BRAFV600E is the most common mutant oncogene in thyroid cancer and is closely related to the
aggressiveness of papillary thyroid carcinoma (PTC). The medical records of 253 patients with PTCs who underwent DLCT and
BRAFV600E mutation detection (mutant group: n=203; wild group: n=50) were reviewed. DLCT quantitative parameters of arterial
phase, typical radiological features and clinical information were compared by univariate and multivariate analysis between the
mutant and wild-type BRAFV600E groups. A nomogram was developed based on the significantly different variables using
multivariate logistic regression analysis. The nomogram performance was evaluated by the area under the receiver operating
characteristic curve (AUC), calibration curve and decision curve analysis (DCA).
Results or Findings: The normalised iodine concentration (NIC), calcification and Hashimoto's thyroiditis (HT) were identified as
independent risk factors of BRAFV600E mutation in PTC. The AUC value of the nomogram based on the three parameters was 0.75.
The calibration curve of the nomogram revealed that the prediction result was in good agreement with actual observation. The
decision curve demonstrated that the nomogram can provide a more satisfactory net benefit than the simple all or none-intervention
standard strategy within a large range of threshold probabilities.
Conclusion: The DLCT-based nomogram with NIC, calcification, and HT has good effectiveness for predicting the BRAFV600E
mutation in PTC.
Keywords: dual-layer detector spectral CT, papillary thyroid carcinoma, BRAFV600E mutation, nomogram
Limitations: The two samples were somewhat imbalanced.
Funding for this study: No funding was received for this work.
Magnetic Resonance (MR) sialographic assessment of masseter muscle and ductal kinking in patients with recurrent
parotitis (7 min)
Maria Cristina Firetto; Milan / Italy
Author Block: M. C. Firetto, S. Triggiani, G. Conte, L. Pignataro, G. M. M. Nicolino, G. Carrafiello; Milan/IT

Purpose: The aim of this study was to assess the anatomical relationship between Stensen's duct and masseter muscle and analyse
their implications in etiopathogenesis of recurrent parotitis secondary to masseter muscle dysfunction.
Methods or Background: We enrolled 50 patients with suspect of masseter dysfunction due to bruxism: 41 unilateral recurrent
parotitis, 9 bilateral parotitis. 33 patients made up the control group. All underwent power doppler ultrasonography, dynamic MR
sialography, sialendoscopy. Basal and dynamic images of salivary glands were acquired at US by a small-parts transducer. MRI were
acquired by 1.5T with head bobina. The before lemon sequences were STIR coronal and trasversal; T1 tse tra; T2 spc tra; T2 haste tra
fs; T2 spc tra rst iso. 3-5 cc of lemon are administered through the mouth. Post lemon sequences were T2 haste tra fs; T2 spc tra; T2
spc rst tra. The scan lasts 40 minutes. At sialendoscopies, Stensen's duct ostium was dilated, then angled sialendoscope was inserted;
exams end when all viable branches have been explored, pervious, without mucous plugs.
Results or Findings: Ultrasonography did not reveal Stensen's duct dilatation or obstruction. MRI sialography showed no statistically
significant difference in imaging metrics. A subgroup analysis by gender, showed both parotids of male subjects with sialadenitis had
a longer duct length compared to control subjects. In unilateral sialadenitis, comparison showed that parotids with sialadenitis had a
wider duct. Sialendoscopy revealed acute duct angle in middle third of main duct; some mucous plugs were detected. Complete
concordance between evidence of acute duct angle during sialendoscopy and wider duct in patients with parotitis was observed
although not statistically significant.
Conclusion: This is one of the first studies in the literature analysing anatomical features relevant for the etiopathogenesis of
recurrent parotitis secondary to the masseter muscle; we aim to increase the cohort population.
Limitations: There was a small number of patients.
The role of magnetic resonance elastography in the evaluation of benign parotid tumours (7 min)
Vitaliy Atamaniuk; Rzeszów / Poland
781
Author Block: V. Atamaniuk, L. Hanczyk, M. Obrzut, I. Palasz-Krasowska, W. Domka, M. Cholewa, B. Obrzut; Rzeszow/PL
Purpose: Magnetic resonance elastography (MRE) is the method of choice for the non-invasive evaluation of the biomechanical
properties of numerous organs. MRE can be regarded as an advanced version of palpation, where the tissue is evaluated both
qualitatively and quantitatively. We hypothesise that MRE can be a valuable tool for evaluating parotid tumours, which are often first
diagnosed as palpable masses inside the parotid gland. This study aims to determine the feasibility of the parotid tumour MRE and
assess the added value of tissue stiffness in tumour diagnosis and treatment. We believe that this new method can potentially enable
us to circumvent the need for aspiration biopsy before surgery.
Methods or Background: Seven patients with benign parotid tumours (two men and five women), aged between 45 and 70,
participated in this study. A 1.5 T whole-body MRI scanner and a motion-sensitised 2D GRE MRE sequence were utilised for data
acquisition. Shear waves in the parotid gland were induced using a custom passive driver fixed against the patient's face. Excitation
frequencies of 40, 60, 70, 80, and 100 Hz were tested. Mean tumour stiffness was calculated using manually drawn masks delineating
tumour areas.
Results or Findings: Among the tested frequencies, 60 Hz vibrations provided optimal shear wave illumination in the tumour area,
allowing for the assessment of tumour stiffness in all patients. The estimated mean ± SD stiffness of parotid tumours was 1.44 ± 0.43
kPa. Tumour stiffness exhibited high heterogeneity both between and within patients.
Conclusion: This study demonstrates the feasibility of MRE for parotid tumours using a specific passive driver and appropriate
excitation frequency. However, reliable reference data from healthy parotid glands are required for results comparison.
Limitations: The primary limitation of this study is its small sample size.
Funding for this study: No external funding was obtained for this study.
Ethics committee - additional information: This study received approval from the ethics committee of the Regional Medical
Chamber (Resolution No 60/2022/B) and each volunteer provided written informed consent before participating in this study.
Ultrasound-assisted identification of the cricothyroid ligament: use in head and neck ultrasound education (7 min)
Andreas Weimer; Heppenheim / Germany
Author Block: J. Künzel1, J. Weimer2, L. Müller2, C. Chrissostomou1, R. Klöckner3, M. Rink1, A. Weimer4; 1Regensburg/DE, 2Mainz/DE,
3
Lübeck/DE, 4Heidelberg/DE
Purpose: When other attempts to secure the airway are unsuccessful, a cricothyrotomy may be indicated. Sonographic guidance can
support the identification of the cricothyroid ligament. The aim of this prospective study was to establish a structured training as part
of head and neck ultrasound courses.
Methods or Background: The training consisted of a 10-min lecture and a worksheet with an additional 10-min video tutorial
followed by hands-on (45 min in groups of 4). To measure previous skills and satisfaction with the training as well as subjective and
objective level of competence, evaluations were completed before (T1) and after (T2) the training. Furthermore, a practical test using
a pocket device at T2 was performed. The answers of the evaluations were recorded using a Likert scale.
Results or Findings: Most participants had neither seen a cricothyrotomy (64.6%) nor had they performed it (79.6%) and they had
no previous experience in using pocket devices (76.1%). Regarding the subjective competence, a significant improvement (T1 to T2)
was measured in all competency areas (P <0.001), with the "Sonographic identification of Lig. conicum" (Δ=3.1 ± 1.6 scale points)
reaching a particularly high increase. In the practical test, 89.2% of the possible points were achieved and an average of 101 seconds
was needed for identification. Participants who already performed a cricothyrotomy, a tracheostomy or at least 30 head and neck
sonographies tended to achieve higher scores and required less time.
Conclusion: Integration of a training concept for sonographic identification of the cricothyroid ligament was well accepted.
Significant improvement in subjective competence as well as a high objective level of competence was achieved. Therefore, such
training concepts should be standardised in head and neck ultrasound training.
Limitations: No evaluation of objective competence level at T1 was conducted; there was no control group without training.
Ethics committee - additional information: This study was approved by the ethics committee of the University of Regensburg.
782
RPS 1816 - Liver, biliary tract and pancreas

Date: March 2, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Philipp Wiggermann; Regensburg / Germany
A risk predictive model for proliferative hepatocellular carcinoma and its prognostic value: using LI-RADS v2018 and
other MRI features (7 min)
Mengtian Lu; Nantong / China
Author Block: M. Lu, X. Zhang, T. Zhang, Q. Qu; Nantong/CN

Purpose: The aim of this study was to develop a predictive model for proliferative hepatocellular carcinoma (HCC) using liver
imaging reporting and data system (LI-RADS) and other MRI features and to investigate its prognostic value for HCC.
Methods or Background: 241 HCC patients who underwent gadoxetic-enhanced MRI were retrospectively included. In the training
set, a diagnostic nomogram was developed using logistic regression analysis to identify proliferative HCC. The model performance
was tested in the validation set and characterised by the area under the receiver operating characteristic curves (AUC). All patients
were divided into high- and low-risk subgroups, and the recurrence-free survival (RFS) in different subgroups was analysed by Kaplan-
Meier curves and the log-rank test. Cox regression analysis was performed to determine risk factors for RFS.
Results or Findings: Corona enhancement (OR, 3.373; P = 0.006), rim arterial phase hyperenhancement (OR, 2.787; P = 0.037),
infiltrative appearance (OR, 7.818; P = 0.018), intratumoural artery (OR, 4.706; P = 0.001), and substantial hypoenhancing
component (OR, 2.684; P = 0.033) are identified as independent predictors of proliferative HCC. The model performed well, with AUCs
of 0.823 and 0.803 for the training and validation sets, respectively. Differences in RFS were significant between the subgroups (high-
risk vs low-risk, P =0.001; proliferative HCC vs non-proliferative HCC, P <0.001). Serum PIVKA level (HR, 1.013; P <0.001) and risk
score (HR, 8.026; P <0.001) were risk factors for RFS.
Conclusion: LI-RADS v2018 and other MRI features may predict proliferative HCC before surgery, which is also predictive for RFS of
HCC.
Limitations: First, this retrospective study only included patients who underwent surgical resection, possibly introducing a selection
bias. Second, the determination of proliferative HCC currently relies solely on histologic subtypes, and genomic studies regarding this
classification deserve more attention.
Ethics committee - additional information: This study was approved by the medical ethics committee of Nantong Third People's
Hospital.
An image-based comparison of histological subtypes in intraductal papillary mucinous neoplasm of the pancreas (7
min)
Jessica Ritter; Munich / Germany
783
Author Block: J. Ritter, I. Pergolini, I. E. Demir, F. Lohöfer, R. Braren; Munich/DE
Purpose: Intraductal papillary mucinous neoplasms (IPMN) are precursor lesions of pancreatic ductal adenocarcinoma (PDAC). IPMN
can be distinguished by histomorphological and immunohistochemical characteristics as intestinal, oncocytic, pancreatobiliary and
gastric subtypes. The Fukuoka consensus guideline is a classification system for risk assessment in IPMN. The aim of the study was to
compare these subtypes according to the Fukuoka criteria.
Methods or Background: MRI scans of 104 patients undergoing pancreatic surgery for IPMN were re-examined for the Fukuoka
criteria. Statistics on the distribution of Fukuoka criteria and the incidence of malignancy were performed.
Results or Findings: In IPMN with pancreatobiliary subtype (n=16), 75% had carcinoma compared to 44% with intestinal subtype
(n=41) and only 25% with gastric subtype (n=47). Further analysis using the Fukuoka criteria compared all malignant versus benign
IPMNs, regardless of the histological subtype. Here significantly more worrisome features (P-value 0.017) and high-risk stigmata (P-
value 0.0077) were detected in the presence of malignancy. For the individual histological subtypes, no significant difference in the
number of worrisome features and high-risk stigmata was found between the benign and malignant cases. A cross-subgroup
comparison within the malignant IPMN showed that patients with PDAC based on pancreatobiliary or intestinal subtype had
significantly more worrisome features than the gastric cohort (2.08/2.22 vs 1.25; P-value 0.040/0.026). Regarding the high-risk
stigmata all three subtypes presented similarly.
Conclusion: According to our findings pancreatobiliary and intestinal subtype are more susceptible to malignant progression than
gastric subtype. At the same time, our results show that the Fukuoka criteria are only of limited use for preoperative subtyping in
IPMN.
Limitations: A limitation of this study is the relatively small number of patients of different subtypes. At least similarly sized patient
groups could be formed in the case of malignancy.
Ethics committee - additional information: The study was approved by the ethics committee of the university hospital rechts der
Isar, technical university of Munich (number 87/18S)
Dual-layer detector spectral CT: a non-invasive preoperative tool for predicting histopathological differentiation in
pancreatic ductal adenocarcinoma (7 min)
Wei Liu; Shanghai / China
Author Block: W. Liu, Z. Zhou, T. Xie, Y. Wang; Shanghai/CN

Purpose: The study aimed to predict histopathological differentiation grades in patients with pancreatic ductal adenocarcinoma
(PDAC) before surgery, with quantitative and qualitative variables obtained from dual-layer spectral detector CT (DLCT).
Methods or Background: A total of 128 patients with histopathologically confirmed PDAC and preoperative DLCT were
retrospectively enrolled and categorised into the low-grade (LG) (well and moderately differentiated, n=82) and high-grade (HG)
(poorly differentiated, n=46) subgroups. The regions of interest were placed on tumours, and both conventional and spectral
variables for PDAC were determined, including CT attenuation, contrast enhancement fraction (CEF), iodine concentration (IC),
effective atomic number (Zeff), iodine enhancement fraction (IEF), and the slope of the spectral curve (K-slope). Necrosis was visually
assessed on both conventional CT images (necrosis_con) and virtual mono-energetic images (VMIs) at 40 keV (necrosis_40keV).
Receiver operating characteristic (ROC) curves were used to evaluate the efficiencies of variables in predicting tumour grade.
Results or Findings: Necrosis_con (odds ratio [OR]=2.84, 95% confidence interval [CI]: 1.13–7.13; p<0.001), necrosis_40keV
(OR=5.82, 95% CI: 1.98–17.11; p=0.001) and IEF in the arterial (AP) and venous (VP) phases (OR=1.12, 95% CI:1.07–1.17; p<0.001)
were significantly different between the LG and HG PDAC groups by univariable and multivariable analyses. IEF_AP/VP (AUC=0.754,
p=0.016) and the combination model (AUC=0.812, p<0.001) had better predictive performances than necrosis_con (AUC=0.580). The
combination model yielded the highest sensitivity (72%) and accuracy (79%), while IEF_AP/VP exhibited the highest specificity (89%).
Conclusion: Variables derived from DLCT have the potential to preoperatively evaluate PDAC tumour grade. Furthermore, spectral
variables and their combination exhibited superior predictive performances than conventional CT variables.
Limitations: The small sample size, and at single institution, was an identified limitation.
Funding for this study: Funding was received from: Shanghai Science and Technology Innovation Action Plan, Hong Kong, Macao
and Taiwan Science and Technology Cooperation Project (22490760800) and Shanghai Minhang District Medical Characteristic
Specialty Development Project (2020MWFC05).
Ethics committee - additional information: The study was approved by the Ethics Committee of Fudan University Shanghai
Cancer Center.
Preoperative scoring model for identifying proliferative hepatocellular carcinoma on multiphase liver MRI and its
implication for surgical resection (7 min)
Junhan Pan; Hangzhou / China
784
Author Block: J. Pan, Y. Zhu, F. Chen; Hangzhou/CN
Purpose: The study aimed to develop and validate a scoring model for preoperative identification of proliferative hepatocellular
carcinomas (HCCs) using multiphase liver MRI, and to compare the early recurrence rates between patients with model-predicted
proliferative and nonproliferative HCCs after surgical resection.
Methods or Background: Between September 2019 and August 2021, 476 consecutive patients with surgically proven HCC who
underwent preoperative multiphase liver MRI were retrospectively enrolled. Logistic regression analyses were conducted to determine
predictors associated with proliferative HCC. We established a scoring model based on the identified predictors (training cohort
n=332) and verified it in a time-independent validation cohort (n=144). Early recurrence rates were evaluated by the Kaplan–Meier
method with log-rank test.
Results or Findings: Serum alpha-fetoprotein level >100 ng/mL, irregular tumour margin, rim arterial phase hyperenhancement,
marked diffusion restriction, and tumour-to-liver ratio at arterial phase ≤1.4 were independent predictors of proliferative HCCs. The
optimal threshold of the scoring model for diagnosing proliferative HCCs was >18 points, with an area under the curve of 0.83 and
0.80 in the training and validation cohort, respectively. In both two cohorts, patients with model-predicted proliferative HCCs
exhibited significantly higher early recurrence rates compared to those with model-predicted nonproliferative HCCs (all P<0.05).
Moreover, patients with BCLC stage B-C and predicted nonproliferative HCCs exhibited similar early recurrence rates to those with
BCLC stage 0-A and predicted proliferative HCCs in both cohorts (all P>0.05).
Conclusion: The developed scoring model, incorporating four MRI features and serum alpha-fetoprotein level, showed promising
potential for predicting proliferative HCCs and identifying suitable surgical candidates for patients with HCC.
Limitations: This was a single-centre retrospective study with inherent selection bias.
Funding for this study: This study received funding from the National Natural Science Foundation of China (12090020 and
12090025) and Zhejiang Provincial Natural Science Foundation Committee-Zhejiang Society for Mathematical Medicine Joint Fund
Major Project (LSD19H180003).
Ethics committee - additional information: This study was approved by the Institutional Review Board of the First Affiliated
Hospital, College of Medicine, Zhejiang University.
Intraindividual comparison of CT and MRI for predicting vessels encapsulating tumour clusters in hepatocellular
carcinoma: which imaging modality is better? (7 min)
Junhan Pan; Hangzhou / China
Author Block: J. Pan, Y. Zhu, F. Chen; Hangzhou/CN

Purpose: The study aimed to establish and validate scoring models for identifying vessels encapsulating tumour clusters (VETC)
pattern in hepatocellular carcinoma (HCC) using CT and MRI, and to compare the predictive performance across the modalities.
Methods or Background: Between June 2019 and August 2020, 324 consecutive patients with surgically confirmed HCC who
underwent preoperative multiphase CT and MRI were retrospectively enrolled. Logistic regression analyses were conducted to
determine predictors associated with the VETC positive HCC. We established two separate scoring models based on the identified
predictors (training cohort n=227) and verified it in an independent validation cohort (n=97). Generalised estimating equations were
used to compare the diagnostic performance of the two models.
Results or Findings: Tumour size >5.0 cm, intra-tumoural artery, three or four types of enhancement pattern, tumour-to-liver ratio
at arterial phase >1.57, and serum AFP > 400 ng/ml were identified as independent predictors for VETC positive HCC on CT.
Conversely, independent predictors on MRI included incomplete "capsule", necrosis or serve ischaemia, marked diffusion restriction,
tumour-to-liver ratio at ADC map ≤1.03, and serum AFP >400 ng/mL. The CT-based and MRI-based scoring models were constructed
using these predictors, with cut-off values of >17 points and >21 points, respectively. Compared to the MRI-based model, CT-based
model exhibited lower sensitivity but higher specificity. However, the CT-based and MRI models demonstrated similar AUC values in
both training (0.83 versus 0.85, P=0.468) and validation cohorts (0.81 versus 0.80, P=0.966).
Conclusion: Preoperative CT and MRI may show comparable predictive performances for identifying the VETC pattern in HCC. The
application of CT or MRI should be determined by the preferred trade-off between sensitivity and specificity in clinical practice.
Limitations: This was a single-centre retrospective study with inherent selection bias.
Funding for this study: This study received funding from the National Natural Science Foundation of China (12090020 and
12090025) and Zhejiang Provincial Natural Science Foundation Committee-Zhejiang Society for Mathematical Medicine Joint Fund
Major Project (LSD19H180003).
Ethics committee - additional information: This study was approved by the Institutional Review Board of the First Affiliated
Hospital, College of Medicine, Zhejiang University.
CT evaluation of hepatic steatosis as a predictive factor for the development of surgical complications in patients with
pancreatic adenocarcinoma resected after neoadjuvant chemotherapy (7 min)
Elisa Boffa; Padova PD / Italy
785
Author Block: E. Boffa, G. Zamboni, M. Antolini, A. Spezia, G. Malleo, G. Mansueto; Verona/IT
Purpose: The study aimed to assess the impact of neoadjuvant chemotherapy (NAT) on the onset of moderate-to-severe liver
steatosis and postoperative complications in patients with pancreatic adenocarcinoma (PDAC).
Methods or Background: This retrospective study included patients with PDAC who received NAT, underwent CT before and after
NAT and underwent major pancreatic resections. All patients were included in a prospectively maintained registry; 71 patients were
also enrolled in an IRB-approved prospective study and underwent intraoperative liver biopsy. Two readers in consensus drew
multiple ROIs on liver and spleen parenchyma on the venous phase: the difference between the two mean densities was calculated
and used to quantify steatosis. Imaging assessment of steatosis was compared with biopsy results in 71 patients. The population was
divided into two groups based on the steatosis degree (group 1: <30%; group 2: >30%). Post-surgical complications and Clavien-
Dindo Index (CDI) were compared between groups applying the chi-squared test.
Results or Findings: A total of 234 patients were included (113 males, 121 females; mean age 62 years). Liver steatosis at CT was
significantly correlated with biopsy results (p=0.0002). After NAT, 82 patients (35%) developed steatosis or worsened their degree of
steatosis. 109 patients (47%) developed post-surgical complications (POPF, liver failure, bile leak, PPAP). Moderate-severe steatosis
before NAT was significantly correlated with a CDI ≥3 (p=0.0453) and with 90-days postoperative mortality (p=0.0262). A positive
trend of correlation was observed between moderate-severe steatosis after NAT and CDI ≥3 and PPAP.
Conclusion: In patients affected by PDAC undergoing neoadjuvant treatment, liver steatosis evaluated on routine CT is associated
with a higher risk of developing complications after major pancreatic resections.
Ethics committee - additional information: Informed consent was provided by all patients (PAD-R registry, n1101CESC). For this
retrospective study, IRB approval was not required. IRB approval was obtained for the prospective LIMBO study.
Evaluation of percutaneous placement of gold fiducial marker using real-time CT/MR guidance in SBRT targeting of
Jeong Min Lee; Seoul / Korea, Republic of
Author Block: J. M. Lee, J. H. Park, S. J. Hwang; Seoul/KR

Purpose: The study aimed to elucidate our preliminary experience utilising a gold fiducial marker via real-time CT/MR-guided
percutaneous placement. Furthermore, we aimed to evaluate its safety, technical efficacy, and clinical utility in stereotactic body
radiotherapy (SBRT) targeting hepatocellular carcinoma (HCC).
Methods or Background: Between May 2021 and August 2023, 19 patients with 24 HCCs underwent percutaneous fiducial marker
placement. This procedure was facilitated using real-time multi-modality-US fusion guidance as a preparatory step before SBRT for
HCCs, which were not visible on pre-contrast CT. Outcomes were gauged regarding technical and clinical success, along with
identifying any procedure-associated complications. We defined technical success as the accurate positioning of the fiducial marker at
the designated tumour site.
Results or Findings: The technical outcomes were exemplary, with a 100% success rate in marker placement. On the clinical front,
one patient with three HCCs experienced marker migration for a subcapsular HCC, which led to a delayed subcapsular haematoma
five days post-procedure. This was managed using trans-arterial embolisation. The other 18 patients underwent SBRT without any
marker migration, resulting in a clinical success rate of 94.7%. During follow-up after SBRT, there was no case of development of HCC
recurrence.
Conclusion: Percutaneous placement of a gold fiducial marker, facilitated by real-time CT/MR guidance, establishes a safe and
efficacious method that enhances the precision and outcomes of SBRT for patients with HCC.
Limitations: Firstly, the sample size was relatively small, which might impact the generalisability of our results to a larger population
of HCC patients. Additionally, our study was conducted over a specific timeframe, and longer follow-up periods might yield more
comprehensive insights into the long-term implications of gold fiducial marker placements and their effects on SBRT outcomes.
Ethics committee - additional information: The study has been submitted to the IRB and we are waiting for IRB approval.
Informed consent is not required as this is a retrospective study.
Apparent diffusion coefficient and tissue stiffness are associated with different tumour microenvironment features in
Jie Chen; Chengdu / China
786
Author Block: J. Chen, T. Chen, Z. Zhang, W. Zhenru, B. Song; Chengdu/CN
Purpose: The study aimed to investigate associations between tissue diffusion, stiffness, and cell density, nuclear volume fraction,
tumour-stroma ratio (TSR), tumour stemness, and CD8+ T cell infiltration in resected hepatocellular carcinoma (HCC).
Methods or Background: We included 72 HCC patients with preoperative magnetic resonance (MR) diffusion-weighted imaging and
MR elastography examination. Free hands region of interests was placed on the central four slices of the apparent diffusion coefficient
(ADC) map and the stiffness map, and the mean value of the tumour and peri-tumour parenchyma was estimated. Cell density,
nuclear volume fraction, TSR, CK19 expression, and CD8+ T cell infiltration was estimated in the whole slide histopathological image
of the resected tumour sample. Spearman’s rank correlation coefficients and intraclass correlation coefficients were calculated.
Results or Findings: Inter-reader agreement was excellent regarding ADC and stiffness measurement. Tumour ADC correlated with
cell density (r=-0.39, p=0.001) and nuclear volume fraction (r=-0.39, p=0.001). Tumour stiffness (TS) correlated with nuclear volume
fraction (r=-0.28, p=0.02) and TSR (r=-0.33, p=0.005). Peri-tumour ADC correlated with CK19 expression (r=0.40, p=0.001) in HCC.
Peri-tumour stiffness correlated with CD8+ T cell infiltration (r=0.27, p=0.02) in HCC.
Conclusion: In HCC, tumour ADC reflects tumour cellularity and nuclear volume fraction, tumour stiffness reflects nuclear volume
fraction and TSR. Peri-tumour ADC is associated with CK19 expression in tumour, and the peri-tumour stiffness is associated with
CD8+ T cell infiltration in tumour.
Limitations: The limitations of the study are its retrospective nature and small sample size.
Funding for this study: Funding was provided by the Science and Technology Support Program of Sichuan Province (grant number:
2022YFS0072) and the 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (grant number: ZYJC21012).
Ethics committee - additional information: The study was approved by the Ethics Committee on Biomedical Research, West
China Hospital of Sichuan University
Pancreatic cancer in photon-counting CT: low keV virtual monoenergetic images improve tumour conspicuity (7 min)
Matthias Michael Wöltjen; Minden / Germany
Author Block: M. M. Wöltjen, J. R. Kroeger; Minden/DE

Purpose: The aim of the study was to identify differences in the tumour conspicuity of pancreatic
adenocarcinomas in different monoenergetic or polyenergetic (T3D) series and
contrast phases in photon-counting CT (PCCT).
Methods or Background: A total of 34 patients were retrospectively enrolled in this study. Quantitative image analysis was
performed with ROI measurements in different monoenergetic levels ranging from 40 up to 70 keV (5-point steps) and polyenergetic
series. Tumour-parenchyma attenuation differences and CNR were calculated. A qualitative image analysis was accomplished by four
radiologists. Differences between groups were evaluated for statistical significance using the Friedman test and in case of significant
differences pair-wise post-hoc testing with Bonferroni correction was applied.
Results or Findings: Tumour-parenchyma attenuation difference was significantly different between the different image
reconstructions for both arterial- and portal-venous-phase-images (p<0.001). Tumour-parenchyma attenuation difference was
significantly higher on arterial-phase images at mono 40 keV compared to T3D images (p<0.001) and mono 55 keV images or higher
(p<0.001). For portal-venous-phase images tumour-parenchyma attenuation difference was significantly higher on mono 40 keV
images compared to T3D images (p<0.001) and mono 50 keV images (p=0.03) or higher (p<0.001). The same trend was seen for
CNR. Tumour conspicuity was rated best on mono 40 keV images with 4.3±0.9 for arterial-phase images and 4.3±1.1 for portal-
venous-phase images. In contrast, overall image quality was rated best on T3D images with 4.8±0.5 for arterial-phase images and
4.7±0.6 for portal-venous-phase images.
Conclusion: Low keV virtual monoenergetic images significantly improve the tumour conspicuity of pancreatic adenocarcinomas in
PCCT based on quantitative and qualitative results. On the other hand, readers prefer polyenergetic images for overall image quality.
Limitations: Limitations were the monocentric study design and the limited number of patients.
and approved by the institutional review board (AZ: 2021-805).
Is computed tomography-based body composition a reliable predictor of chemotherapy-related toxicity in pancreatic

cancer patients? (7 min)
Vito Chianca; Naples / Italy
787
Author Block: V. Chianca1, M. Cefalì2, C. Reiner3, S. Stocker3, M. Kissling4, F. Del Grande1, S. De Dosso2, S. Rizzo1; 1Lugano/CH,
2
Bellinzona/CH, 3Zurich/CH, 4Basle/CH
Purpose: Malnutrition, loss of weight and of skeletal muscle mass are frequent in pancreatic cancer patients, a majority of whom will
undergo chemotherapy over the course of their disease. Available data suggest a negative prognostic role of these changes in body
composition on disease outcomes; however, it is unclear whether tolerance to chemotherapeutic treatment is similarly and/or
negatively affected. We aimed to explore this association by retrospectively assessing changes in body composition and
chemotherapy-related toxicity in a cohort of advanced pancreatic cancer patients.
Methods or Background: 131 patients (mean age 69.7 ± 9.0 years, 45% women and 55% men), most of whom (81.5%) had
metastatic disease at diagnosis, were enrolled in this study.
Body composition was evaluated through clinical parameters and through radiological assessment of muscle mass, skeletal muscle
area, skeletal muscle index and skeletal muscle density; an assessment of fat distribution by subcutaneous adipose tissue and
visceral adipose tissue was performed. We performed descriptive statistics, pre/post chemotherapy comparisons and univariate and
multivariate analyses to assess the relation between changes in body composition and toxicity.
Results or Findings: Toxicity risk increased with an increase of skeletal muscle index (OR: 1.03) and body mass index (OR: 1.07),
whereas it decreased with an increase in skeletal muscle density (OR: 0.96). Multivariate analyses confirmed a reduction in the risk of
toxicity only with an increase in skeletal muscle density (OR: 0.96).
Conclusion: This study suggests that the retrospective analysis of changes in body composition is unlikely to be useful in predicting
toxicity to gemcitabine-nab-paclitaxel.
Limitations: The small sample size was an identified limitation.
Ethics committee - additional information: The Ethics Committee approved this retrospective study with a waiver for informed
consent (2020-01085).
Intrahepatic mass-forming cholangiocarcinoma: is there additional prognostic value in using Gd-EOB enhanced MRI? (7
min)
Sebastian Halskov; Berlin / Germany
Author Block: S. Halskov, F. Krenzien, L. K. Segger, D. Geisel, B. Hamm, J. Ihlow, W. Schöning, T. A. Auer, U. Fehrenbach; Berlin/DE
Purpose: The study aimed to investigate the prognostic value of enhancement patterns of intrahepatic mass-forming
cholangiocarcinomas (IMCCs) during the hepatobiliary phase (HBP) in gadoxetic acid (Gd-EOB) enhanced MRI.
Methods or Background: We retrospectively identified 66 consecutive patients with histopathologically proven IMCCs (reference
standard: resection) and preoperative Gd-EOB-enhanced MRI. Gd-EOB retention area was subjectively rated based on areas of
intermediate signal intensity. Lesions were classified as either hypointense (0-25% retention area) or significantly-retaining (>25%
retention area). Clinical, radiological and prognostic features were compared between these groups. The primary endpoint was
recurrence-free-survival (RFS) after primary surgical resection.
Results or Findings: In total, 73% (48/66) of lesions were rated as hypointense and 29% (19/66) as significantly-retaining. While the
hypointense subgroup more frequently featured local and distant intrahepatic metastases (p=0.039 and p=0.022) and an infiltrative
growth pattern (p=0.005), RFS and clinical features did not differ significantly with Gd-EOB retention area or quantitatively measured
HBP enhancement ratios. Lymph node metastasis was an independent predictor of poor RFS (p=0.001).
Conclusion: Gd-EOB-enhanced MRI reveals two subtypes of IMCC in the HBP: hypointense and signal-retaining. The hypointense
subtype is associated with more frequent intrahepatic metastases and an infiltrative growth pattern, indicating potential tumour
aggressiveness. However, this did not result in a significant difference in survival after primary resection of IMCC.
Limitations: This was a retrospective study of a surgical cohort, which could lead to selection bias: the prognostic value of imaging
features in unresectable HCC remains unclear. Histopathological analysis to identify small-duct and large-duct IMCCs could have been
of added value for identifying prognostic imaging subtypes. Quantitative analysis of HBP enhancement evaluated the whole tumour
without taking intra-lesion heterogeneity in enhancement into account.
Ethics committee - additional information: Our institutional review board approved this retrospective study (internal registration
number: EA2/016/14) and waived informed consent due to the retrospective nature. The study protocol conforms to the ethical
guidelines of the 2002 Declaration of Helsinki.
Efficacy and evaluation of response in hepatocellular carcinoma (HCC) treated with stereotactic body radiotherapy
(SBRT) (7 min)
788
Author Block: F. Castagnoli1, A. Villanacci2, M. Bertuletti2, B. Frittoli2, L. Grazioli2; 1London/UK, 2Brescia/IT
Purpose: The study aimed to evaluate changes in ADC values, size, wash-in, wash-out, T2 signal intensity, necrosis and capsule of
HCCs treated with SBRT.
Methods or Background: A total of 62 patients with HCC (65 lesions), treated with SBRT, who underwent baseline gadoxetate-
enhanced MRI within four weeks before treatment and follow-up MRI within three months after SBRT were prospectively enrolled; 44
patients underwent a further MRI follow-up 12 months after SBRT. Three radiologists evaluated tumour size (∂), ADC, wash-in, wash-
out, capsule, necrosis and T2 signal of all lesions. Local control (LC) was evaluated after a median follow-up (FU) of 2.1 years.
Delta(Δ)∂ and ADC were correlated using Spearman correlation coefficient. Δ∂ and ΔADC were compared (Mann-Whitney test)
between patients with and without LC. A logistic regression model was trained to estimate the probability of response of patients
using categorical imaging variables.
Results or Findings: The median HCC ADC (10⁻³mm²/s) and dimension were 0.94 and 22.5 mm at baseline; 1.326 and 16.4 mm at
3-month follow-up; 1.352 and 10.6 mm at 12-month follow-up. At 12-months, the ADC values increased by 41.1% (p<0.001), whereas
tumour size decreased by 52.7% (p<0.001).
At 3-month follow-up there was no correlation between the increase in ADC values and decrease in lesion dimension (p=0.41).
At 12-month follow-up, there was correlation between the increase in ADC values and decrease in lesion dimension (p=0.04).
At 3- and 12-month follow-up, there was significant difference in Δ(∂) and presence of wash-out between patients with and without LC
(p=0.05, p=0.03; p=0.04, p=0.02). No significant correlation were found for Δ(ADC), wash-in, T2w signal, necrosis and capsule
between patients with and without LC.
Conclusion: SBRT showed excellent LC for HCC, which was associated with the degree of size change after treatment and absence of
wash-out at follow-up.
Limitations: Identified limitations were the small sample size and the limited cases of local recurrence (n=7).
Ethics committee - additional information: The current study was carried out according to the Declaration of Helsinki (1964) and
was approved by the Ethics Committee of the ASST Spedali Civili, Brescia.
789
MD 11 - Brain tumours: new developments in imaging and treatment - recommendations for clinical
Categories: Multidisciplinary, Neuro, Oncologic Imaging

Date: March 2, 2024 | 10:00 - 11:00 CET
CME Credits: 1
Moderator:

1. To highlight new developments in the treatment of brain tumours.

The neuro-oncologist's perspective (8 min)

Martin Jacques van den Bent; Rotterdam / Netherlands
The neurosurgeon's perspective (8 min)

Niels Verburg; Amsterdam / Netherlands
The neuroradiologist's perspective (8 min)

790
EFRS 8 - EFRS Educational Wing in the Future

ETC Level: LEVEL I
Date: March 2, 2024 | 10:30 - 11:30 CET
This session aims to learn the outcomes of the 1st EFRS summit, to meet the EFRS Educational Wing Management Team and to
consider the need for OPTIMAX summer school, as also to share the thoughts and suggestions for EFRS EW webinars.
Moderator:

Outcomes from EFRS Summit (10 min)

EFRS Strategy for Education (5 min)

Summer School experiences (10 min)

Supporting members in education - Discussion (20 min)
Closing (5 min)
791
CUBE 20 - Case Studies: Individualized Decision Making for Rescue Therapy
Categories: Emergency Imaging, Interventional Radiology, Neuro, Research, Vascular

Date: March 2, 2024 | 10:30 - 11:00 CET
being explored.
Moderator:

Case Studies: Individualized Decision Making for Rescue Therapy (28 min)
Johannes Kaesmacher; Bern / Switzerland
1. To discuss new imaging modalities for mechanical thrombectomy patient selection.

2. To understand Flat-Panel Perfusion as tool for individualized decision making for rescue therapy.
792
AI-SC 7 - Economic impact and clinical risk of AI

Date: March 2, 2024 | 11:15 - 12:15 CET
Moderator:

Economic impact and clinical risk of AI (57 min)

Hugues Brat; Sion / Switzerland
1. To appreciate how to assess the economic impact of AI applications.

2. To appreciate how to address risks associated with AI applications.
3. To discuss use cases of economic impact and clinical risks of AI applications.
793
ST 16 - EDiR: a certificate of excellence. Why you should take it?
Date: March 2, 2024 | 11:30 - 12:00 CET

The European Diploma in Radiology is a certificate of excellence with a great prestige not only at European level but also
international.
Moderators:
Interview (30 min)

794
PL 3 - Advancing radiology in times of dwindling resources and increasing demands
Categories: Management/Leadership, Professional Issues

Date: March 2, 2024 | 11:30 - 12:00 CET
CME Credits: 0.5
Moderator:
Advancing radiology in times of dwindling resources and increasing demands (28 min)
795
CUBE 21 - EYMINT E-fellowship programme: current state and future directions

Date: March 2, 2024 | 12:00 - 12:30 CET
Moderator:

EYMINT E-fellowship programme: current state and future directions (28 min)
Uta Hanning; Münster / Germany
1. To present the EYMINT, the trainee association of the European Society of Minimally Invasive Neurological Therapy (ESMINT).
2. To discuss about the E-Fellowship programme: a Remote access of trainees to training centers via video streaming.
796
RPS 1908 - Head and neck: advances in tumour imaging
Categories: Artificial Intelligence & Machine Learning, Head and Neck, Imaging Methods, Interventional Oncologic Radiology,
Oncologic Imaging
Date: March 2, 2024 | 12:15 - 13:45 CET
CME Credits: 1.5
Moderator:
Jussi Hirvonen; Turku / Finland
The prediction of biological features using MR imaging in head and neck squamous cell carcinoma: a systematic review
and meta-analysis (7 min)
Hedda Joanne van der Hulst; Amsterdam / Netherlands
Author Block: H. J. van der Hulst, R. W. Jansen, C. Vens, P. Bos, R. Martens, Z. Bodalal, R. G. H. Beets-Tan, M. van den Brekel, J. A.
Castelijns; Amsterdam/NL
Purpose: Magnetic resonance imaging (MRI) is an indispensable, routine technique, providing morphological and functional imaging
sequences. MRI can potentially capture tumour biology, allowing for longitudinal evaluations of such biology. This systematic review
and meta-analysis evaluates the ability of MRI to predict tumour biology in primary head and neck squamous cell carcinoma (HNSCC).
Methods or Background: Studies were screened, selected, and assessed for quality using appropriate tools according to the
PRISMA criteria. Fifty-eight articles were analysed, examining the relationship between (functional) MRI parameters and biological
features and genetics.
Results or Findings: Most studies focused on associations between individual MRI features and HPV status, revealing a consistent
pattern wherein HPV-positive tumours exhibit significantly lower ADCmean (SMD: 0.82; p<0.001) and ADCminimum (SMD: 0.56;
p<0.001) values. Lower ADCmean values are also associated with high Ki-67 levels, consistent with a diffusion restriction caused by
high cellularity. Several perfusion parameters of the vascular compartment were found to be significantly associated with HIF-1α.
Analysis of other biological factors (vascular endothelial growth factor [VEGF], epidermal growth factor receptor [EGFR], tumour cell
count, p53, and micro vessel density [MVD]) yielded inconclusive results. This review showed that larger datasets with homogenous
acquisition are required to develop and test radiomic-based prediction models capable of capturing different aspects of the underlying
tumour biology.
Conclusion: Overall, this meta-analysis demonstrates that rapid and non-invasive characterisation of tumour biology features by MR-
imaging is feasible and could enhance clinical outcome predictions and personalised patient management for HNSCC.
Limitations: While comprehensive, the review of MRI parameter associations with biological features is limited by the large variety of
methods used in the included articles for calculating biological and MRI parameters.
Ethics committee - additional information: This was a review article.
Comparison of diffusion-weighted MRI using imaging split acquisition of fast spin echo signal (SPLICE) and single-shot
echo-planar imaging (SS-EPI) in tumours of the head and neck (7 min)
Hedda Joanne van der Hulst; Amsterdam / Netherlands
797
Author Block: H. J. van der Hulst, R. Martens, B. Westerink, L. Braun, L. C. ter Beek, R. Tissier, R. G. H. Beets-Tan, M. van den Brekel,
J. A. Castelijns; Amsterdam/NL
Purpose: Traditionally, diffusion-weighted imaging (DWI) uses single-shot echo planar imaging (SS-EPI) for its fast read-out but
struggles with distortion, predominantly near bone-air interfaces. Turbo Spin Echo (TSE) DWI, like split acquisition of fast spin echo
signal for diffusion imaging (SPLICE), may reduce this issue, especially in the head and neck area. This study compares DW-SPLICE
with DW-SS-EPI for head and neck lesions (HN-lesions).
Methods or Background: Between August 2020 and January 2022, 57 potential MRI HN-lesion exams were conducted using both
SS-EPI and SPLICE techniques on a 3.0T MRI system. Lesions >0.5cm³ were delineated on both DWI-MRIs. The study assessed
differences in whole lesion ADC-values through interclass correlation (ICC), Bland-Altman plots, and mixed-effect model regression.
Two blinded radiologists evaluated image quality and distortion levels using a standardised questionnaire.
Results or Findings: Preliminary findings:
The eligible 41 MRI exams of 41 patients included 40 benign and 20 malignant HN-lesions, visible on both DWI techniques. Strong
interclass correlation coefficients were found for mean ADC values (ICC agreement 0.864, 95% CI 0.782-0.917), consistency 0.862
(95% CI 0.780-0.915). The Bland-Altman regression showed discrepancies between SPLICE and EPI for higher average ADC values
(ß=0.16, p=0.02), but not when assessing benign and malignant lesions separately. Radiologist evaluations of visual distortion
showed a trend towards less distortion and better image quality for SPLICE-DWI.
Conclusion: High agreement and consistency in mean ADC values were seen for HN-lesions measured with both techniques.
Qualitative analyses of radiologists resulted in a trend towards better distortion and image quality scores for SPLICE-DWI. While a
potential ADC value discrepancy was observed in the Bland-Altman analyses, its clinical impact may be minimal if DWI-SPLICE can
effectively replace SS-EPI.
Limitations: Identified limitations were: (1) the cohort consists of a mixed group of HN-lesions; (2) data is currently still preliminary.
Ethics committee - additional information: The study was approved by the IRBd22-270, registered at 28.10.2022.
Does radiotherapy influence the diagnostic performance of ultrasound-guided fine-needle aspiration cytology of neck
nodes in head and neck squamous cell carcinoma? (7 min)
Sophie Neveu; Geneva / Switzerland
Author Block: S. Neveu, A. Escriva, M. Becker; Geneva/CH

Purpose: The study aims to evaluate and compare the diagnostic performance of ultrasound-guided fine-needle aspiration cytology
(US-FNAC) in neck nodes of head and neck squamous cell carcinoma (HNSCC) patients with and without previous radiotherapy (RT+
versus RT-).
Methods or Background: Retrospective study comprising a series of 99 consecutive HNSCC patients (66 RT- and 33 RT+, 25
females, 74 males, mean age = 67±12 years). A total of 136 nodal aspirations (92 RT- nodes, 44 RT+ nodes) were analysed. US-FNAC
results were compared with the standard of reference (neck dissection and follow-up).
Results or Findings: The minimum/maximum axial diameter of biopsied RT- versus the RT+ nodes was similar (10.97±7.99
mm/15.9±10.9 mm versus 9.07± 5.9 mm/13.6±6.7 mm), t(134)=1.42/1.266, p<0.05. Likewise, nodal level distribution and US
imaging characteristics of benign and malignant nodes were similar in both the groups (p>0.05). For diagnostic US-FNAC probes, the
sensitivity, specificity, accuracy, positive and negative predictive values (PPV and NPV) of the US-FNAC in the RT- versus the RT+
group were 91.3%, 95.5%, 93.3%, 95.5%, 91.3% versus 94.7%, 88.9%, 91.8%, 89.5%, 94.4%, respectively (p>0.05). However, the
proportion of non-diagnostic cytology results was significantly higher (7/44 = 15.9%) in the RT+ versus the RT- group (2/92 = 2.2%),
p=0.005.
Conclusion: Although, irradiation significantly increases the proportion of non-diagnostic samples, it has no influence on the
diagnostic performance of US-FNAC in neck nodes of HNSCC patients.
Limitations: Limitations of the current study include its retrospective design and a potential selection bias as only patients with US-
FNAC were included in the current study.
Funding for this study: Funding was received from the institution.
Ethics committee - additional information: This retrospective study was approved by the local Ethics Committee and was done in
accordance to the Helsinki declaration (CER 11-092R). Informed consent was waived.
Detection of recurrent nodal disease in head and neck squamous cell carcinoma with multiparametric contrast-
enhanced FDG-PET/DWIMRI (7 min)
Sophie Neveu; Geneva / Switzerland
798
Author Block: S. Neveu1, A. Varoquaux2, O. Rager1, C. De Vito1, N. Dulguerov1, C. Combescure1, H. Zaidi1, M. Becker1; 1Geneva/CH,
2
Marseille/FR
Purpose: The purpose of this study was to determine the diagnostic performance of contrast-enhanced FDG-PET/DWIMRI to detect
recurrent nodal disease in head and neck squamous cell carcinoma (HNSCC) treated with radio(chemo)therapy (RCT).
Methods or Background: This prospective study included 47 patients treated with RCT ± surgery for advanced HNSCC who
underwent 67 contrast-enhanced FDG-PET/DWIMRI examinations as part of their surveillance protocol. Lymph nodes (LNs) were
evaluated prospectively using a 5-point Likert scale by experienced readers. Lymph node size, ADCmean/ADCmin and
SUVmean/SUVmax were measured for all detected LNs. The standard of reference was histology of neck dissection specimens in 34%
and follow-up ≥ 5years in 66% of patients. The diagnostic performance was assessed both in terms of sensitivity, specificity and
accuracy and using Receiver Operating Analysis (ROC) in 67 examinations, 134 neck sides, 938 LN levels and 220 LNs.
Results or Findings: The prevalence of LN metastases depended on the type of analysis performed (50% per neck side, 5% per LN
level and 44% per LN). Mean minimum axial diameter was 6.5±4.0 mm in benign and 7.0±5.7mm in malignant LNs (p>0.05) and only
11% of all nodes had a diameter >10 mm. ADC mean/ADC min values were 1.14±0.31|0.62±0.29x10-3 mm2/s and
1.17±0.29|0.78±0.29x10-3 mm2/s in benign and malignant LNs, respectively (p>0.05) while SUV mean/SUV max values were
1.87±0.84|2.38±1.15 in benign and 3.40±2.38|4.79±3.45 in malignant LNs (p<0.05). FDG-PET/DWI-MRI had a
sensitivity/specificity/accuracy of 90%/64%/72% per examination, 88%/82%/83% per neck side, 74.5%/97%/96% per LN level, and
49.5%/76.4%/64.5% per LN. The areas under the curve (AUC) were 0.926 per examination, 0.933 per neck side, 0.874 per LN group
and 0.661 per LN.
Conclusion: Analysis type heavily influences results. As most recurrent metastatic LNs are <1 cm in size, the sensitivity of contrast-
enhanced FDG-PET/DWIMRI is limited in the per level and per LN analysis.
Limitations: The limitation was that it was a single study center.
Funding for this study: Funding was received from the Swiss National Science Foundation (SNSF) (grant number: N0
320030_173091/1).
Ethics committee - additional information: This prospective clinical study was approved by the Swissethics committee (Project ID
2017-00748) and performed in accordance with the guidelines of the Helsinki II declaration. Written informed consent was obtained
from all subjects.
CT-based radiomic markers are independent prognosticators of survival in advanced laryngeal cancer: a pilot study (7
min)
Amar Rajgor; Newcastle Upon Tyne / United Kingdom
Author Block: A. Rajgor, A. S. Mcqueen, J. Cowley, A. Mill, S. Rushton, B. Obara, T. Bigirumurame, K. Kallas, D. Hamilton; Newcastle
Upon Tyne/UK
Purpose: The objective of this study was to identify CT-based radiomic features that may predict survival and guide prognostication
in advanced laryngeal cancer.
Methods or Background: Laryngeal cancer affects 2400 new patients yearly. Half of these patients present with advanced disease
where at most DSS rates are 50-60%. During the diagnostic pathway, medical imaging is typically acquired. These images hold a
wealth of information about the solid tumour and its microenvironment. With advances in computational techniques we can extract
such information and correlate this with oncological outcomes to produce a ‘radiological biomarker.’
Methods:
Pre-biopsy contrast-enhanced CT scans were assembled from a retrospective cohort of 72 patients with advanced laryngeal cancers
(T3-T4). Two consultant radiologists performed tumour segmentation using the LifeX platform and extracted 68 radiomic features.
Two radiomic features were selected via LASSO-Cox regression model: shape compacity (irregularity of tumour volume) and
GLZLM_ZLNU (tumour heterogeneity). The prognostic potential of both features was explored via Cox-regression analysis. The cohort
was additionally stratified to upper, middle, and lower terciles according to radiomic feature values for Kaplan-Meier analysis.
Results or Findings: Multivariable Cox-regression analyses determined that greater shape compacities (HR 2.89, 95% CI 1.40-5.93,
p=0.004) and GLZLM_GLNU (HR 1.64, 95% CI 1.02-2.63, p=0.041) were significantly associated with worse 5-year disease-specific
survival. Patients with the upper tercile of shape compacity values had poor 5-year disease-specific mortality (51% vs 76% [middle
tercile] vs 83% [lower tercile], p=0.032), as was observed for GLZLM_GLNU (47% [upper tercile] vs 63% [middle tercile] vs 97% [lower
tercile], p=0.0013).
Conclusion: Two radiomic features were identified as independent prognostic biomarkers for oncological outcome. A large, multi-
center prospective study is necessary to expand on these important findings.
Limitations: Two major limitations of this study are that (1) Retrospective, single-centre and (2) Small sample size .
Funding for this study: The funding was obtained from NIH.
Predicting response to exclusive combined radio-chemotherapy in naso-oropharyngeal cancer: the role of texture
analysis (7 min)
Eleonora Bicci; Florence / Italy
799
Author Block: E. Bicci, L. Calamandrei, M. Pietragalla, S. Busoni, S. Paolucci, F. Mungai, C. Nardi, L. Bonasera, V. Miele; Florence/IT
Purpose: The aim of the study was to identifying MRI texture features able to predict response to radiochemotherapy (RT-CHT) in
patients with naso-oropharyngeal carcinoma before treatment to help clinical decision making.
Methods or Background: 41 patients with naso-oropharyngeal carcinoma (NPC-OPC) were enrolled in the study and were divided
into 2 groups (responders/non-responders) according to MRI, 18F-FDG-PET/CT performed 3-4 and 12 months after RT-CHT, and biopsy.
Pre-RT-CHT lesions were segmented and radiomic features were extracted from ADC-maps and post-gadolinium T1 MRI images
acquired on a single MRI scanner.
A radiomic model differentiating responders from non-responders was built using the LASSO algorithm on all the features extracted,
which allows at the same time to make feature reduction and selection.
Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) values were generated a 95% confidence interval
(CI) was reported.
Results or Findings: LASSO model selected five statistically significant features: Maximum 2D Diameter Slice, Dependence Non-
uniformity, Gray Level Non-Uniformity, Small Area high gray level emphasis from post-gadolinium T1 MRI images and 10 Percentile
from ADC maps AUC associated with the ROC was: 0.978 with 95% CI: 0.937-1.
Conclusion: Texture analysis on post-gadolinium T1-images and ADC-maps could potentially predict response to therapy in patients
with NPC-OPC who will undergo exclusive treatment with RT-CHT, being, therefore, a useful tool in clinical decision-making.
Limitations: The limitations of this study were that (1)The use of an MRI-unit only is an advantage to make the sample as
homogeneous as possible, but currently a disadvantage for the lack of generalizability of results. (2) The relative low sample size.
Nevertheless, most studies on texture analysis of OPC and NPC are based on Computed Tomoghrapy (CT) and not MRI
examinations.(3) The small number of patients with residual cancer for the well-known excellent response to RT-CHT of OPC
(especially HPV+) and NPC.
Ethics committee - additional information: The study was approved by the Ethics Committee of our Institution (study protocol n:
21800)
Early intra-treatment T1 rho imaging predicts treatment response to induction chemotherapy in patients with
advanced nasopharyngeal carcinoma (7 min)
Qi Yong AI; Hong Kong / China
Author Block: Q. Y. AI, Y. M. Tsang, Z. Yu, B. B. Ma, E. P. Hui, W. Chen, A. D. King; Hong Kong/HK
Purpose: The aim of the study was early prediction of treatment response to induction chemotherapy (IC) in patients with advanced
nasopharyngeal carcinoma (NPC) that could help clinicians to adjust treatment plans. This study evaluates whether quantitative T1
rho imaging, a pulse sequence that is newly applied to research in cancer, can predict treatment response to IC at early intra-
treatment stage in patients with advanced NPC.
Methods or Background: Twenty-five eligible patients who had advanced NPC planned for treatment with 3-cycle of IC +
concurrent chemoradiation therapy were prospectively recruited. T1rho MRI scans of the primary tumour were performed pre-
treatment and early intra-treatment (2-week after IC started) and the T1rho mean value was obtained. Treatment response to IC was
evaluated at the third cycle of IC based on the RECIST 1.1. T1rho value of pre-treatment (T1rhopre) and percentage change in T1rho
between pre- and intra-treatment scans (%Δ) =[(T1rhointra-T1rhopre)/T1rhopre *100%] were compared between responders
(complete response and partial response) and non-responders (stable disease and progression disease) using Mann-Whitney U test.
Median (interquartile) values were reported.
Results or Findings: Comparing responders (n= 15) and non-responders ( n=10), responders showed higher pre-treatment T1
rhopre [66.6 (66.2, 74.3) x msec vs. 63.8 (61.7, 69.5) x msec, p=0.03] and greater reduction in T1 rho [ %Δ decrease of -9.0%
(-12.1%, 3.9%) vs. 3.2% (-4.9%, 6.6%), (p=0.03)].
Conclusion: Pre-treatment and early intra-treatment change in the T1 rho mean values show potential for predicting treatment
response to IC in patients with advanced NPC.
Limitations: Preliminary results were reported. Further evaluation is needed.
Funding for this study: This study was funded by Research Grant Committee of The Hong Kong S.A.R. Government (Ref. 14100419)
Ethics committee - additional information: This study was approved by the Joint Chinese University of Hong Kong and New
Territories East Cluster Clinical Research Ethics Committee.
Radiologic predictors of response to neoadjuvant chemotherapy in patients with advanced oral cavity tumours (7 min)
Iris Burck; Frankfurt a. Main / Germany
800
Author Block: I. Burck, A. Gleich, E. Herrmann, J-E. Scholtz, J. Gruen, P. Thönissen, M. Fleischmann, R. Winkelmann, D. Pinto Dos
Santos; Frankfurt a. Main/DE
Purpose: The aim of this study was to explore whether radiological imaging features could predict response to neoadjuvant
radiochemotherapy in patients with locally advanced oral-cavity tumour.
Methods or Background: We included 28 Patients (14 women, median age 59+-9.5) with stage IVa oral cavity carcinoma who
underwent neoadjuvant radiochemotherapy at our institution between February 2012 and August 2022. All patients received MRI
scans before treatment, after 15 days of treatment and preoperatively. MRI images were judged qualitatively based on a Lickert scale
with respect to their overall intensity and change in intensity and tumour extent between the scans. Additionally, scans were
analysed quantitatively by measuring the absolute signal intensity (SI) of the tumour in ADC, DWI and T2 and normalising the values
to the signal intensity of the spinal cord. Differences and quotients of normalised ADC-, DWI- and T2-SIs were calculated. Patients with
pathological T1 or T0 stage were classified as responders, while all others were classified as non-responders.
Results or Findings: We found significant differences between responders and non-responders for normalized ADC-SI at both follow-
up timepoints (p = 0.014 and p = 0.024, respectively), as well as for the quotient between normalized DWI-SI at baseline and
preoperatively (p = 0.011) and the difference between normalized ADC-SI at baseline and preoperatively (p = 0.035).
Conclusion: Diffusion weighted imaging parameters (normalized ACD-SI and normalized DWI-SI) could predict response to
neoadjuvant radiochemotherapy in jaw carcinoma and could potentially be used to guide treatment or extent of surgery in these
patients.
Limitations: Limitations of this study include small sample size, single-centre and retrospective study design.
Ethics committee - additional information: This study was approved by the Institutional Review Board (approvals number 208/12
)
Role of intraoral ultrasound in assessment of depth of invasion of oral cavity squamous cell carcinoma and its
correlation with histopathology: a prospective diagnostic validation study (7 min)
Pankaj Sharma; Rishikesh / India
Author Block: P. Sharma; Rishikesh/IN

Purpose: The aim of this study was to investigate the accuracy of IOUS in the assessment of the DOI in early oral SCC and
management in different phases and potential applications in the future.
Methods or Background: Oral cavity carcinomas accounts for 1/3rd of head and neck carcinomas. Eight edition of AJCC introduced
DOI as one of the main criteria in staging the oral cavity carcinomas. Depending on the disease's stage, surgery with or without
adjuvant radiotherapy (RT) or chemotherapy and radiation therapy is the preferred method of treatment. In order to design the
surgical strategy and do preoperative assessments of the DOI, imaging is crucial. IOUS was performed on 86 OCSCC patients using
high frequency hockey stick probe. DOI and cervical lymph node metastasis was assessed and correlated with postoperative
histopathological DOI.
Results or Findings: The usDOI and pDOI had a high degree of agreement, according to a Bland-Altman plot. IOUS and
histopathological staging assessments showed nearly complete agreement (k=0.894, p value 0.001) for the T stage and for N stage
between IOUS and histopathological staging evaluations (k = 0.81, p value 0.001). Mean depth of invasion by histology and IOUS was
9 mm and 9.7 mm, respectively. Strong correlation (interclass correlation coefficient = 0.95) was noted between radiological and
pathological tumour depth.
Conclusion: The IOUS can be recommended as an alternative to MRI in the preoperative staging of tongue SCC since it was reliable
in a preoperative assessment of a pDOI and T stage. Ultrasound (US) is a quick, painless, affordable, and efficient diagnostic tool.
Without the danger of radiation exposure, it can portray soft tissues very clearly. Can perform core needle biopsy (CNB) or US-guided
fine needle aspiration (FNA) simultaneously.
Limitations: It was single centre study with a learning curve for technique.
Funding for this study: No funding was obtained for the study.
Ethics committee - additional information: The study was approved by the Institutional Ethical Committee of AIIMS Rishikesh.
Image-guided intraoperative assessment of surgical margins in oral cavity squamous cell cancer: a diagnostic test
accuracy review (7 min)
Giorgia Carnicelli; Rome / Italy
801
Author Block: G. Carnicelli, S. Lusi, G. M. Frigerio, L. Disconzi, G. Mercante, L. Balzarini, M. Francone, C. Giannitto; Milan/IT
Purpose: The objective of this study was to assess the imaging guidance of oral cavity squamous cell cancer. Incomplete primary
resection of oral cavity squamous cell cancer has a dramatic impact on prognosis and re-treatment. Imaging guidance may improve
the rate of negative margin resection (currently around 45%). A diagnostic test accuracy review was designed to assess the
performance of intraoperative ultrasound and magnetic resonance imaging (MRI) in determining free margin status.
Methods or Background: Two independent reviewers systematically screened articles from EMBASE, Cochrane and MEDLINE
databases. The study protocol was designed according to standards of the Cochrane handbook for diagnostic test accuracy (DTA)
reviews. Among inclusion criteria were MRI and ultrasound-based studies (2016-2023), a free-margin resection cut-off of 5 mm,
frequency of US probes ≥ 15 MHz. Negative predictive value (NPV), sensitivity and specificity of intraoperative imaging were
calculated for selected studies. The software used for the review process was Review Manager 5.4 .
Results or Findings: Ten articles were included in the final analysis, four used for data extraction (Figure 1). Overall NPV of both
techniques for margins < 5 mm ranged from 0.50 to 0.91, with higher diagnostic yield of ultrasound compared to MRI. Sensitivity of
imaging ranged from 0.07 to 0.75; specificity ranged from 0.81 to 1.00 (Figure 2). Overall, the mean rate of free margin resection was
55.8% ± 28 . In all studies with control group, imaging guidance always provided a benefit, with increase in free-margin resection of
35% ± 13; the accuracy decreased for stages ≥ T1-2 and with increasing depth of invasion (DOI) values .
Conclusion: NPV of intraoperative ultrasound in guiding oral cavity cancer resection remains suboptimal with 40-50% margins being
missed, and varies with DOI and T stage. Nevertheless, it provides a mean 35% improvement in free margin resection rate.
Limitations: Heterogeneity of the included studies limits the scope of this study.
Ethics committee - additional information: There was no ethical committee approval for reasons related to the nature of the
study.
Reproducibility and prognostic value of NI-RADS score in early post-operative CT for oral cavity cancer (7 min)
Giovanni Mazzon; Padua / Italy
Author Block: M. Panfili, G. Mazzon, S. Longo, S. Settimi, F. Pastore, D. A. Mele, M. Massaccesi, R. Calandrelli, S. Gaudino; Rome/IT
Purpose: This retrospective study aims to assess the inter-observer concordance of the NI-RADS score applied both for primary site
and regional lymph-node levels in early post-operative CT scans in patients with surgically treated oral cavity cancer and its
correlation with outcomes. The Neck Imaging reporting and Data System (NI-RADS) is a valuable tool for post-treatment imaging in
head and neck cancers with good capability to predict disease recurrence. Yet, it remains underexplored its specific utility and
prognostic value in early post-operative computed tomography (CT) scans in surgically treated patients with oral cavity cancer (OC)
and eligible for adjuvant therapy.
Methods or Background: Two experienced radiologists independently reviewed CT scans performed less than 90 days from surgery
and assigned NI-RADS scores both for T site and N site. NI-RADS scores ranged from 1 to 3. The Cohen's kappa was used to test the
interobserver agreement and the Kaplan-Meier survival analysis was used to evaluate correlation between NI-RADS scores and
patients overall survival.
Results or Findings: Forty-two CT scans were reviewed with a median post-surgery time of 40 days. The Cohen's kappa statistic
indicated substantial inter-observer agreement for the primary tumour (kappa = 0.91, 95% CI 0.79-1.00) and regional lymph nodes
(kappa = 0.92, 95% CI 0.77-1.00). Patients with a NI-RADS score of 3 at the primary tumour level had a shorter median OS of 7
months compared to those with NI-RADS scores of 1, 2a, and 2b, who had a median OS of 68 months. This difference was statistically
significant with a p-value of 0.006.
Conclusion: The study demonstrates that NI-RADS score in early post-operative CT has good reproducibility and could have
prognostic implications.
Limitations: Retrospective nature and small number of patients.
Funding for this study: No funding was obtained for the study.
Diagnostic accuracy of MRI in detecting perineural spread of head and neck tumours: a systematic review and meta-
analysis (7 min)
Umida Bafoevna Abdullaeva; Tashkent / Uzbekistan
802
Author Block: U. B. Abdullaeva1, B. Pape2, J. Hirvonen3; 1Tashkent/UZ, 2Vaasa/FI, 3Tampere/FI
Purpose: The aim of this study was to review the diagnostic accuracy of MRI in detecting perineural spread (PNS) of head and neck
tumours using histopathological or surgical evidence from the afflicted nerve as the reference standard.
Methods or Background: Previous studies in the English language published in the last 30 years were searched from PubMed and
Embase databases. We included studies that used MRI (with and without contrast enhancement) to detect PNS, histological or surgical
confirmation of PNS, and reported exact numbers of patients required for assessing diagnostic accuracy. The outcome measures were
sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Heterogeneity was assessed with the
Higgins inconsistency test (I2). P values smaller than 0.05 were considered statistically significant.
Results or Findings: A total of 11 retrospective studies were found reporting 319 nerve samples from 245 patients. Sensitivity
ranged from 0.46 to 1.00, and specificity from 0.83 to 1.00, with median values of 0.96 and 0.88, respectively. Meta-analytic
estimates and their 95% confidence intervals were: 0.85 (0.70-0.95), specificity 0.85 (0.80-0.89), PPV 0.86 (0.70-0.94), and NPV 0.85
(0.71-0.93). We found statistically significant heterogeneity for sensitivity (I2=72%, p=0.003) and PPV (I2=70%, p=0.038), but not for
NPV (I2=65%, p=0.119) or specificity (I2=12%, p=0.842). The most frequent MRI features of PNS were nerve enlargement and
enhancement. Squamous cell carcinoma and adenoid cystic carcinoma were the most common tumour types, and the facial and
trigeminal nerves were the most commonly affected nerves in PNS. Only a few studies provided examples of false MRI diagnoses.
Conclusion: MRI has high diagnostic accuracy in depicting PNS of cranial nerves, yet this statement is based on scarce and
heterogeneous evidence.
Limitations: The study is based on a limited number of retrospective studies with some studies subject to sampling bias.
Funding for this study: This study was financially supported by the Sigrid Jusélius Foundation.
Ethics committee - additional information: Since this is a systematic review, Institutional Review Board approval was not
necessary.
803
RC 1916 - Post-treatment changes in oncologic patients
Categories: Abdominal Viscera, Chest, Genitourinary, Interventional Oncologic Radiology, Interventional Radiology
Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Prostate (15 min)

1. To understand prostate MRI for post-treatment evaluation.

2. To learn about signs of recurrence.
3. To investigate pitfalls in follow-up examinations.
Lung (15 min)

Fernando Gomez Muñoz; Valencia / Spain
1. To understand advances in percutaneous lung biopsies.

2. To learn about ablative techniques in thoracic malignancies.
3. To investigate imaging changes after procedures.
Liver (15 min)

1. To understand which interventional approaches are available.

2. To learn about liver changes after radioembolisation.
3. To investigate a new approach (MRI based).
Panel discussion: Is radiology enough, or do we still need surgery? (10 min)
804
RPS 1912 - The newest techniques in paediatric radiology: faster imaging, less radiation and contrast
agents
Categories: Artificial Intelligence & Machine Learning, Contrast Media, EuroSafe Imaging/Radiation Protection, Imaging Methods,
Paediatric
Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Élida Vazquez; Barcelona / Spain
Factors associated with CT Scan repetition among paediatrics and its association with cancer risk: a systematic review
and meta-analysis study (7 min)
Tahani Alshangeeti; Kuala Lumpur / Malaysia
Author Block: T. Alshangeeti1, R. Ahmad1, P. D. M. Abdullah Alshawsh1, M. Elzaki2; 1Kuala Lumpur/MY, 2Madinah/SA
Purpose: The study aimed to evaluate and quantify the risk of cancer in paediatric patients due to exposure to CT scans, including
single and repeated exposure. To identify the types of cancer associated with radiation from CT and investigate the repeat rate,
cumulative radiation doses, and reasons behind multiple exposures in paediatric populations.
Methods or Background: A review of pediatric patients' CT scan repetition found factors contributing to it and its link to cancer risk.
The study, which included 33 studies through Web of Science, Scopus, and PubMed, had fivestudies with over 7 million participants in
the meta-analysis. The Newcastle-Ottawa Scale was used to assess potential bias. The study offers a comprehensive evaluation of the
evidence regarding exposure and outcome.
Results or Findings: A meta-analysis found an increased risk of overall cancer and brain tumours for children who underwent CT
scans. The risk of brain tumours increased by 53% and the evidence for an increased risk of leukaemia was less conclusive. A dose-
response effect was observed, with patients receiving two or more repeat CT scans showing a markedly elevated risk compared to
those who had only one scan. Many patients undergo repeat scans for injury reassessment rather than a change in their clinical
condition, particularly in head injuries. A study found a disparity in radiation exposure levels and the lack of implementation of CT
dose optimization strategies in hospitals, resulting in unnecessarily high radiation doses.
Conclusion: CT scan exposure poses a significant risk of cancer, especially brain tumours. Repeat scans may increase the risk of
leukaemia.
Limitations: The heterogeneity in the selection of variables used to assess cancer risk and the lack of clear explanations for why CT
scans were repeated in the articles were identified.
Ethics committee - additional information: We registered the protocol of this systematic review in the International Prospective
Register of Systematic Reviews (PROSPERO) database, which can be accessed at http://www.crd.york.ac.uk/PROSPERO, and it is
identified with the record number CRD42022342579
Myocardial strain assessment in the human foetus by cardiac MRI using Doppler ultrasound gating and feature
tracking (7 min)
Björn Schönnagel; Hamburg / Germany
805
Author Block: B. Schönnagel, M. Tavares de Sousa, P. Bannas, G. Adam, R. Fischer, L. Huber, M. Kaul, J. Herrmann, M. Dargahpour
Barough; Hamburg/DE
Purpose: The aim of this study was the assessment of myocardial strain by feature tracking magnetic resonance imaging (FT-MRI) in
human fetuses with and without congenital heart disease (CHD) using cardiac Doppler ultrasound (DUS) gating.
Methods or Background: Forty-three human foetuses (gestational age: 28-41 weeks) underwent dynamic cardiac MRI at 3T. Cine
balanced steady-state free-precession imaging was performed using foetal cardiac DUS gating. FT-MRI was analysed using dedicated
post-processing software. Endo- and epicardial contours were manually delineated from foetal cardiac 4-chamber views, followed by
automated propagation to calculate global longitudinal strain (GLS) of the left (LV) and right ventricle (RV), LV radial strain, and LV
strain rate.
Results or Findings: Strain assessment was successful in 38/43 fetuses (88%), 25 of them had postnatally confirmed diagnosis of
CHD (e.g. coarctation, transposition of great arteries) and 13 were heart healthy. Five foetuses were excluded due to reduced image
quality. In foetuses with CHD compared to healthy controls median LV-GLS (-13.2% vs. -18.9%; P<0.007), RV-GLS (-7.9% vs. -16.2%;
P<0.006), and LV strain rate (1.4s-1 vs. 1.6s-1; P<0.003) were significantly higher (i.e., less negative). LV radial strain was without a
statistically significant difference (20.7% vs. 22.6%; P=0.1). Bivariate discriminant analysis for LV-GLS and RV-GLS revealed sensitivity
of 67% and specificity of 93% to differentiate between foetuses with CHD and healthy foetuses.
Conclusion: Myocardial strain was successfully assessed in the human foetus performing dynamic fetal cardiac MRI with DUS gating.
Our study indicates that strain parameters may allow for differentiation between foetuses with and without CHD.
Limitations: No direct comparison with echocardiography. Myocardial strain was assessed from four chamber views only.
Funding for this study: This study received a grant by the Deutsche Forschungsgemeinschaft.
Ethics committee - additional information: The ethical approval was obtained from the local ethics committee.
Location comparison of radiation doses in paediatric Hickman line insertion (7 min)

Sara Hosseinzadeh; Glasgow / United Kingdom
Author Block: S. Hosseinzadeh; Oxford/UK

Purpose: Percutaneous Hickman line insertion is a common and necessary procedure for the delivery of chemotherapeutics, long
term medications, and patients with difficult vascular access. In our institution, paediatric Hickman line insertion occurs in the
radiology department fluoroscopy room or in surgical theatres. Ultrasound-guided puncture of internal jugular vein is followed by
guide-wire insertion and subsequently Hickman line tunnelled in the anterior chest wall. Position of the wire and line tip are confirmed
fluoroscopically. We assessed differences in radiation exposure between fluoroscopy room and surgical theatres.
Methods or Background: Radiation dose and screening time were recorded from all paediatric Hickman line insertions over a three
year period. Inter-room and inter-operator data was analysed. Mann-Whitney U Test was used for statistical analysis.
Results or Findings: Three hundred and thirty seven Hickman lines were inserted between January 2020 and May 2023. Radiation
doses were significantly lower in the fluoroscopy room than in theatre (p<0.0001), and this effect remained when data was analysed
in age subgroups and controlling for operators. Screening times were significantly reduced in the fluoroscopy room compared to
theatre (p<0.05). Lower radiation dose (p<0.001) and screening time (p<0.01) was also observed in radiologist-performed insertions
compared with surgeon-performed insertions in theatre.
Conclusion: The location and operator of Hickman line insertions significantly impacts the radiation dose and screening time in the
procedure. Even controlling for operator, patients were exposed to lower radiation doses in our fluoroscopy room than in theatres.
Limitations: Theatre cases may be more complex than patients suitable for fluoroscopy room. Of note, however, this study spans
the COVID-19 pandemic, during which many ‘routine’ procedures ordinarily carried out in fluoroscopy were performed in theatre,
which may somewhat mitigate this effect.
Funding for this study: This was a student project. No specific funding was acquired.
Ethics committee - additional information: This study was performed as internal service evaluation using pooled anonymised
data, and hence no ethics approval was required.
Intracranial gadolinium deposition after using gadolinium-based contrast agents in children: a systematic review and
meta-analysis (7 min)
Min Wu; Cheng Du / China
806
Author Block: L. Bao1, M. Wu2; 1Chengdu/CN, 2Cheng Du/CN
Purpose: The objective of this study was to determine whether the use of GBCAs induces gadolinium deposition in the brains of
children.
Methods or Background: We searched PubMed, MEDLINE, Web of Science, Embase and Cochrane Central Register of Controlled
Trials databases until April 2023. Studies on changes in T1-weighted MR images (T1WI) signal intensity in the globus pallidus (GP) and
dentate nucleus (DN) in the brains of children after administration of GBCAs were included. Quality evaluation, heterogeneity
analyses, meta-regression, subgroup analyses, sensitivity analyses and publication bias analyses were performed.
Results or Findings: Ten studies were finally included in the meta-analysis, including 459 paediatric patients with an age range of
0.1 to 18.1 years. In the linear GBCAs studies, the changes in T1-weighted imaging intensity in the GP and DN exhibited significant
difference (effect size=1.55, 95% CI: 0.75 to 2.34; effect size=1.96, 95% CI: 1.01 to 2.91, respectively) and notable heterogeneity
(I2=83.3%, P<0.001; I2=89.4%, P<0.001, respectively). In the macrocyclic GBCAs studies, after excluding interfering studies, no
significant change existed in T1WI signal intensity in GP or DN (effect size=0.13, 95% CI: -0.08 to 0.35; effect size=-0.02, 95% CI:
-0.21 to 0.16, respectively) and no heterogeneity existed (I2=0%, P=0.884; I2=0%, P=0.959, respectively).
Conclusion: Regrding linear contrast agents, a history of brain radiation therapy and the number of uses of GBCAs affected
gadolinium deposition in the brains of children. Macrocyclic GBCAs had higher safety in children. These data lay the foundation for
future prospective large-scale clinical trials in children.
Limitations: As some included studies failed to adequately explore the potential influence of age-related changes in signals of brain
structures, our study could not analyze the correlations between age and gadolinium deposition. Besides, the number of included
patients was limited and high-quality studies such as randomized controlled trials were lacking.
Funding for this study: This work was supported by the Sichuan Foundation for Distinguished Young Scholars (2022JDJQ0049); the
Scientific and technological Achievements Transformation Fund of West China Hospital, Sichuan University (Grant CGZH21002).
Ethics committee - additional information: No further information provided by the submitter.
Reducing radiation dose and contrast medium volume in paediatric spectral CT imaging: how is the second generation
dual-layer CT? (7 min)
Lichen Ren; Zhengzhou / China
Author Block: L. Ren1, N. Guo2, Y. Zhang1; 1Zhenzhou/CN, 2Beijing/CN

Purpose: The objective of this study was to explore the feasibility of reduced radiation dose and contrast medium using dual-layer
spectral CT in pediatric patients.
Methods or Background: This study has been approved by the ethics committee. One hundred and twelve peadiatric patients were
enrolled in this study and randomly separated into two groups. The study group underwent scanning with a DRI index of 13. The
contrast agent volume was reduced with 0.2 ml/kg compared to the control group. Fifty keV monoenergetic images were
reconstructed. The control group underwent scanning with a DRI index of 16, and the contrast agent volume were 1.2 to 1.8 ml/kg
based on patient’s weight. Hundred kVp polyenergetic images were reconstructed. CT values and standard deviations (SD) of the
abdominal aorta, abdominal aorta branches, hepatic veins, portal vein main trunk, liver parenchyma and paraspinal muscles were
measured. Subjective evaluations of overall image quality were performed. Statistical analysis was conducted with student t-tests and
rank sum tests.
Results or Findings: There were no statistically significant differences in age and weight between two groups (P>0.05). The
CTDIvol, SSDE, and DLP of the study group were lower than those of the control group (each P<0.05). The contrast medium volume of
the study group was reduced by approximately 24.0% compared to the control group. CT values for vessels and organs in the study
group were higher than those in the control group (P<0.001). There was no statistically significant difference in subjective image
quality scores between two groups (P>0.05). Both groups had image quality sufficient for diagnostic requirements.
Conclusion: In paediatric patients, dual-layer spectral CT can significantly reduce radiation dose and iodine contrast medium volume
of abdominal and pelvic spectral imaging without compromising image quality.
Limitations: This study is a single-center study.
The effect of arm positioning on radiation dose and image quality during paediatric CT chest examinations (7 min)
Charlotte Emma Kelly; Doha / Qatar
807
Author Block: C. E. Kelly, O. Arthurs, I. C. Simcock; London/UK
Purpose: The aim of this study was to identify the optimal arm position for paediatric CT chest scanning if patients cannot raise both
arms alongside their head, considering image quality and radiation dose.
Methods or Background: Positioning the arms outside of the field of view, (alongside the head) during CT chest scans reduces
radiation dose and artifacts. However, some children are unable to achieve this optimal positioning due to physical disability or an
inability to co-operate. In this study we measured the effect of sub-optimal arm positioning on absorbed patient dose and image
quality. We evaluated 154 chest CT scans performed with arms in the suboptimal position between January 2022 and July 2023 (age
range 4 days and 16 years old), compared to mean age-weighted CTDIvol for patients with arms up.
Results or Findings: The highest doses were seen when one or both arms were positioned perpendicular to the torso and this was
commoner in younger patients. Overall, mean radiation dose was 125% higher with both arms perpendicular, 50% higher with one
arm perpendicular, 88% higher with both arms down, and 63% higher with one arm down. All scans were of diagnostic quality, but
streaking was present in 47% of scans when both arms were perpendicular, 24% of scans with one arm perpendicular and 14% when
both arms were down alongside the body.
Conclusion: Sub-optimal positioning of the arms during paediatric CT chest scans increases patient dose and reduces image quality.
Arms should be placed alongside the body if they cannot be raised, to reduce radiation dose and improve image quality.
Limitations: The range of weights meant absolute CTDIvol values could not be compared but instead were compared as a proportion
of the optimum CTDIvol value.
Ethics committee - additional information: The study utilised retrospective anonymous data analysis.
Ushering into another dimension: diagnostic utility of low-dose four dimensional (4D) dynamic airway CT in paediatric
stridor (7 min)
Shubham Saini; Chandigarh / India
Author Block: S. Saini, K. S. Sodhi, A. Bhatia, A. K. Saxena, J. Mathew; Chandigarh/IN

Purpose: Dynamic airway disorders are an important cause of stridor in children. Fiberoptic bronchoscopy (FOB) is the gold standard
and an invasive method of diagnosis of these disorders. There is lack of objective criteria for diagnosis of paediatric dynamic airway
disorders. The study aims to assess the diagnostic utility of low-dose 4D dynamic airway CT in paediatric stridor on comparison with
FOB.
Methods or Background: We conducted a prospective study in 24 children (age:13 days-1.5 years, boys: 18) with stridor. Low-dose
4D dynamic airway CT was performed with 70kVp and 15 mAs. The number of acquisition cycles varied from five to ten. Stridor during
CT acquisition was also noted. Image analysis was done using volume rendered reconstruction and axial CT images and degree of
collapse was noted. FOB was performed within 72 hours of CT with blinding of both the respective observers. The findings were
compared with FOB as the gold standard.
Results or Findings: The diagnostic accuracy of 4D dynamic airway CT was 91.6%, 95.7%, and 95.5% for the diagnosis of
laryngomalacia, tracheomalacia, and bronchomalacia respectively. The sensitivity and specificity of 4D dynamic airway CT for the
diagnosis of laryngomalacia was 94.7% and 80.0% respectively with substantial agreement between the two modalities (k=0.777).
Sensitivity for detection of tracheomalacia and bronchomalacia was 90.9% and 75% respectively with 100% specificity. There was
near perfect agreement between the two modalities with kappa coefficient of 0.913 for tracheomalacia and 0.831 for bronchomalacia.
Mean radiation dose was 0.827 mSv. The presence of stridor during CT had a positive association with prediction of abnormality
detection on CT.
Conclusion: 4D dynamic airway CT can be used as a robust and non-invasive objective tool in assessing dynamic airway diseases in
children.
Ethics committee - additional information: This study was approved by institutional ethics committee.
Deep learning reconstruction in the paediatric brain DWI: comparison with conventional multi-shot DWI (7 min)
Ha Young Yang; Seoul / Korea, Republic of
808
Author Block: H. Y. Yang; Seoul/KR
Purpose: The purpose of this study is to compare DWI for paediatric brain evaluation using single-shot echo-planar imaging with
deep learning reconstruction (DL) and multi-shot multiplexed sensitivity-encoding (MUSE).
Methods or Background: DL- and MUSE-DWI were performed for 85 paediatric patients (mean age, 91 months), and two datasets
were independently reviewed by two radiologists. Qualitative analyses were performed for perceptive coarseness, image distortion,
susceptibility-related signal change, motion artefacts, ghosting, truncation artefacts, and lesion conspicuity using a 5-point Likert
scale. Quantitative analyses were conducted for spatial distortion (measured using eyeball diameter) and signal uniformity (at lateral
ventricle, centrum semiovale, and pons) of each sequence. Qualitative scores were compared using Wilcoxon signed-rank test and
quantitative values were compared using a paired t-test. Interobserver agreement between the two radiologists was evaluated using
the weighted Cohen kappa test.
Results or Findings: DL-DWI showed more ghosting (p-value <0.001) and less truncation artefacts (p-value <0.001) for both
readers (interobserver agreement 0.898 and 0.905 respectively). Reader 2 gave DL-DWI a higher score for image coarseness (p-value
0.034) and susceptibility-related signal change (p-value 0.046). Other qualitative parameters did not differ significantly between two
sequences. Both spatial distortion and signal uniformity did not differ significantly between two sequences.
Conclusion: Single-shot DWI combined with a parallel imaging and a deep learning reconstruction (DL-DWI) can provide image
quality comparable to multi-shot DWI in a shorter scan time. Therefore, DL-DWI can be recommended for children who need to be
imaged as quickly as possible.
Limitations: The acquisition order of two sequences was not randomised due to the retrospective nature of the study. Also, the
study population is relatively small and only 17 of 85 patients had visible abnormalities, hence further studies are needed to solidify
clinical relevance.
Ethics committee - additional information: The study was approved by Seoul National University Hospital.
809
BS 19 - Normal or not? A neuroradiological roadmap
Categories: Neuro
Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Brain (15 min)

Majda M. Thurnher; Vienna / Austria
1. To recognise the most common anatomic variants in brain imaging.

2. To learn pitfalls to avoid when interpreting brain MRI.
3. To be aware of a failure in not suggesting the next appropriate procedure.
Head and neck (15 min)

Katarina Šurlan Popović; Ljubljana / Slovenia
1. To become familiar with normal variants and pitfalls in head and neck imaging.
2. To learn how to avoid unnecessary tests or follow-up imaging.
3. To recognise the most common anatomic variants that are relevant for head and neck surgical planning.
Spine (15 min)

Johan Van Goethem; Antwerpen / Belgium
1. To familiarise oneself with common normal variants that simulate disease in spinal imaging.
2. To learn how to differentiate normal variants from pathology in spinal imaging.
3. To recognise the most common anatomic variants that are relevant for head and neck surgical planning.
Panel discussion: To follow up or not to follow up? That is the question. (10 min)
810
EDiR 19 - EDiR simulation session: real exam cases with feedback from the examiners, win an EDiR
examination place

ETC Level: LEVEL II
Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderators:
Carmel J. Caruana; Msida / Malta

EDiR simulation (40 min)

1. To experience a slightly different kind of EDiR: multiple response questions, short cases and clinically oriented reasoning evaluation
(CORE) cases as you would find in an actual exam but focussed on hybrid imaging and physics.
2. To analyse and solve theoretical and hands-on cases.
3. To become familiar with the clue patterns of the EDiR questions.
4. To gain knowledge about the most challenging cases of this simulation.
5. To understand and assimilate the fundamentals of these subspecialties.
6. To learn the most relevant EDiR educational tools.
Scientific discussion and Q&A (15 min)

*Win your place in an EDiR examination. Please note that there can only be one winner per session.
811

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1


812
HW 19Sd - Mastering stroke perfusion imaging and MR, CT and AI applications

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1

1. To become familiar with the use of MRI for the purpose of using perfusion MRI in assessing stroke patients.
2. To learn about stroke and treatment options by examining the role of CT scans in guiding treatment decisions for strokes.
3. To develop practical skills in perfusion techniques and post-processing for patient diagnosis, triage, and treatment orientations.
4. To understand available public and commercial platforms in stroke imaging and artificial intelligence-driven applications for acute
stroke.

Abderrahmane Hedjoudje; Cergy / France
813
HW 19Mb - Ultrasound of the muscles and nerves of the lower limb

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1

1. To become familiar with muscle and nerve anatomy of the lower limb with the usual landmarks.

Tutors
Guillaume Lefebvre; Lille / France
Alexander Talaska; Vienna / Austria
814
RPS 1909 - IR in the management of benign and malignant disease in male patients
Categories: Genitourinary, Interventional Radiology, Vascular

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
A single-centre retrospective study of 300 men with localised primary prostate cancer treated using MRI-guided
transurethral ultrasound ablation (TULSA) (7 min)
Lucas Engelage; Berlin / Germany
Author Block: L. Engelage1, N. Behnel1, P. Doerwald2, L. Steinmeister1, R. Muschter1, A. Lumiani1; 1Berlin/DE, 2Hamburg/DE
Purpose: The objective of this study was to report a retrospective clinical service evaluation of TULSA for patients with organ-
confined primary prostate cancer treated at a single centre.
Methods or Background: 300 men with primary localised PCa visible on MRI and confirmed by biopsy were treated under clinical
routine. Based on patients’ preferences and disease characteristics, 163 men received whole-gland treatments and 137 focal TULSA.
Adverse events (AEs), prostate-specific antigen (PSA), and functional and oncological outcomes were assessed using mp-MRI and
biopsy, if necessary.
Results or Findings: Baseline characteristics include median [IQR] age 66 [60-73], PSA 6.85 ng/ml [4.64-9.5], overall cancer length
7.6 mm [4.35-10], prostate volume 49.2 cc [min 11–max 180], baseline risk stratifications include ISUP 1 (52/300), 2 (191/300), 3
(46/300), 4 (9/300), and 5 (2/300). 57 patients experienced Grade 1 and 2 complications, mostly resolving within 4 weeks through
prolonged catheterisation and antibiotics. Grade 3 adverse events occurred in seven patients, resolving within 3 months. No grade 4
or higher AEs and no bowel-related complications were observed. Median [IQR] IIEF score of 24 [14-29] at baseline remained stable at
the 48-mo follow-up with a value of 25 [13-30]. IPSS score immediately post-treatment increased from its baseline value of 8 [4-15] to
12 [5-19] and improved to 8 [2.5-10.5] by 12 months and further improved to 48-mo with 6.5 [3.5-8]. To the last follow-up, 26 men
experienced biochemical failure (Phoenix≥2); MRI findings showed suspicion of residual cancer in 26 men, 54% (14/26) of which were
confirmed by positive biopsy. 15 patients received salvage therapy (3 surgery, 12 other), 18 patients remained under active
surveillance, and 13 patients underwent a single repeat TULSA.
Conclusion: TULSA offers promising safety, functional, and oncological results for the treatment of organ-confined prostate cancer.
Limitations: No systematic follow-up protocol limits this study.
Funding for this study: Study is a clinical service evaluation and so no funding was obtained for this study.
Twelve-month outcomes of salvage transurethral ultrasound ablation (TULSA) for the treatment of localised radio-
recurrent prostate cancer (7 min)
Pouya Doerwald; Hamburg / Germany
815
Author Block: M. Anttinen1, P. Mäkelä1, P. Doerwald2, P. Nurminen1, T. Sainio1, H. E. Pärssinen1, R. Blanco Sequeiros1, P. J. Boström1;
1
Turku/FI, 2Hamburg/DE
Purpose: The purpose of this study was to report the 12-month safety and oncological outcomes of salvage TULSA (sTULSA) for men
with localised radio-recurrent PCa.
Methods or Background: Men with biopsy-proven localised PCa recurrence after RT were enrolled. mp-MRI and 18-F-PSMA-PET-CT
were used to confirm organ-confined disease. Patients were followed every three months for adverse events (AEs, Clavien-Dindo),
functional status, and PSA. Disease control was evaluated at 12-month by mpMRI, PSMA PET-CT, and prostate biopsy targeting the
treatment area plus areas suspicious in imaging.
Results or Findings: Forty-one men underwent sTULSA (25 whole-gland and 16 focal ablation); one patient withdrew from the study
due to frequent follow-up protocol three months after his treatment with undetectable PSA (<0.006 ng/ml). Baseline characteristics
included median [IQR] age 73 [69-77], PSA 3.3 ng/ml [2.3-7], and an interval of 11 years [8-13] between RT and sTULSA. 12-month
follow-ups are available for 31 men. AEs included two grade 3 events and 17 grade 2 events. One patient with castration-resistant
PCa experienced a pubo-prostatic fistula and osteitis after sTULSA, treated with prolonged suprapubic catheterization and oral
antibiotics. Median PSA at 12-month was 0.19 ng/ml (IQR 0.07–0.57) and was undetectable (<0.1) in 13 patients. Two patients were
diagnosed with biochemical recurrence (PSA≥nadir+2) at 12-month follow-up, correlating with the extraprostatic disease on imaging.
At 12-month MRI and PSMA PET-CT, 28/31 (90%) men had no visible cancer in the prostate, and four patients had seminal vesicle
invasion. Biopsy outcomes at one year revealed that 26/30 (87%) were free of any PCa in the treatment region, while four patients
had positive out-of-field biopsies.
Conclusion: Salvage therapy options for men with radio-recurrent PCa are limited. Twelve-month clinical outcomes of sTULSA show
remarkable oncological outcomes with a favorable safety profile for treating localised radio-recurrent PCa.
Limitations: Small population
Funding for this study: This study was investigator-initiated so funding was received for it.
Ethics committee - additional information: Ethics committee approval was granted and informed consent was obtained from all
study participants.
MRI-guided transrectal focal laser ablation for localised low and intermediate risk prostate cancer: initial experience
using an integrated laser ablation system (7 min)
Yvonne Wimper; Nijmegen / Netherlands
Author Block: Y. Wimper, L. te Molder, M. Sedelaar, J. Bomers, C. G. Overduin, J. Fütterer; Nijmegen/NL

Purpose: The aim of this study was to investigate the feasibility and safety of an integrated system for MRI-guided FLA in localized
PCa. Current active treatment of localised prostate cancer (PCa) is directed towards whole-gland radical treatment. MRI-guided focal
laser ablation (FLA) presents a minimally invasive treatment alternative for selected patients offering targeted tumour treatment at
reduced morbidity. It relies on a laser applicator combined with real-time MR thermometry feedback for accurate ablation control.
However, in current systems these are often non-integrated. Our study will enable a more precise tissue ablation without damaging
vital adjacent structures.
Methods or Background: Between April 2022 and May 2023, 10 patients (mean age: 66±7.2 years) with low- and intermediate risk
PCa were prospectively included and treated with MRI-guided FLA using an integrated FLA system (Tranberg,CLS). Primary endpoints
were technical success, procedure-related adverse events (AEs) and tumour response on MRI at 6 months. Secondary endpoints were
prostate-specific antigen (PSA) levels, sexual and urinary function response at 6 months.
Results or Findings: Technical success was achieved in 10/10 (100%) patients. Two AEs were observed, acute urinary retention
(classification D, n=1) and urinary tract infection (classification B, n=1). Six-month follow-up was available in 6/10 patients. In all six
patients, follow-up MRI indicated no evidence of residual tumor. At six months, mean PSA level was reduced (5.9 vs. 3.4 ng/mL;
p=0.59), urinary function scores (10.0 vs. 7.8; p=0.37) and sexual function scores improved (10.3 vs. 15.4; p=0.47) compared to
baseline, all being non-statistically significant.
Conclusion: Our study demonstrates feasibility and safety of an integrated MRI-guided FLA system in patients with low-to-
intermediate risk PCa. Initial experience indicates good oncological response and potential improvement in urinary and sexual
function.
Limitations: This feasibility study is limited by a small, single-institution cohort and short-term follow-up.
Funding for this study: Clinical Laserthermia Systems has provided funding for this study.
Ethics committee - additional information: The ethical approval was obtained by the Institutional Review Board.
CTA and angiographic prostatic artery anatomy in a Korean population (7 min)

Seung Hyun Lee; Incheon / Korea, Republic of
816
Author Block: S. H. Lee, D. Shim, J. Yoo, B. Park, C. S. Park, S. K. Kim, J. W. Lee; Incheon/KR
Purpose: This study aims to analyse the origins of prostatic arteries (PAs) in Korean population with LUTS or malignant hematuria
and to compare with literatures. Prostatic artery embolisation (PAE) is a safe and effective treatment for lower urinary tract symptoms
(LUTS) and hematuria from malignancy. Nevertheless, PAE remains challenging due to numerous anatomic variations.
Methods or Background: From April 2018 to August 2023, 73 patients (mean age=68.9 ± 9.6) with LUTS (n=69) or malignant
hematuria (n=4) underwent PAE. CTA and transcatheter angiography images were retrospectively reviewed for arterial anatomy
evaluation. The branching pattern of internal iliac artery (IIA) and the origin of PA were categorised, and the incidence of each type
was recorded.
Results or Findings: PAE was successfully implemented in 143 of 146 pelvic sidewalls. PA cannulation failed in three sidewalls due
to total IIA occlusion. Most common IIA was A type (dividing into superior gluteal and gluteal-pudendal trunk, 72%). Eleven of 143
sidewalls exhibited dual PAs. A total of 154 PA origin was investigated. Internal pudendal artery origin (IPA, type IV) was most
common, 40%, followed by superior vesical artery origin (type I, 25%) and obturator origin (type III) in 19%. Less commonly, anterior
division of IIA origin (type II) was 7%. Type V (uncommon origins) was 8% including three distal IPA, two inferior gluteal, two medial
femoral circumflex, and two rectal arteries and two quadfurcation and one trifurcation patterns.
Conclusion: In Korean population, the most common IIA pattern and PA origin was type A and IV, respectively, which aligns with
previous reports.
Limitations: The limitations of the study include its retrospective nature, small sample size, and single centre study.
Ethics committee - additional information: This study is a retrospective study hence no ethical approval was required.
Six years of prostatic artery embolisation (PAE): a comprehensive and critical single-centre, single-operator
retrospective analysis of 551 cases (7 min)
Leona Soraja Alizadeh; Frankfurt a. Main / Germany
Author Block: L. S. Alizadeh, C. Booz, I. Yel, D. Radek, T. Vogl; Frankfurt a. Main/DE

Purpose: The objective of this study was to evaluate the efficacy, safety, and outcomes of prostatic artery embolisation (PAE) in
patients with benign prostatic hyperplasia (BPH) over a six-year period at a single centre and to discern its impact on clinical
symptoms, prostate volume, and procedural specifics.
Methods or Background: A single-centre retrospective study was conducted, analysing 551 PAE interventions performed by a
single interventional radiologist from January 2017 to July 2023. The study reviewed technical specifics, changes in prostate volume,
PSA levels, International Prostate Symptom Score (IPSS), Quality of Life (QoL) scores, International Index of Erectile Function (IIEF),
and procedural radiological metrics.
Results or Findings: The average age of patients was 68.81 years. A significant majority underwent bilateral embolisations in a
single session (435 out of 551) with particles predominantly sized 100-300 μm (441 out of 551). Post-intervention, there was a
notable reduction in prostatic volume (mean decrease of 9.66 ml) and PSA levels (mean decrease of 0.32 ng/ml). Symptomatic relief
was evident with a substantial drop in IPSS (mean decrease of 9.17) and improved QoL scores (mean decrease of 1.74). IIEF scores
showed a slight increase post-procedure. The average fluoroscopy time was 25.23 minutes, with an average dose area product (DAP)
of 25452.08 μGym².
Conclusion: PAE emerges as a potent therapeutic intervention for BPH patients, offering symptomatic relief, reduction in prostate
volume, and a favorable safety profile. We reported increasing patient numbers since the implementation of the technique. This
single-center experience provides a comprehensive insight into the real-world applicability and efficacy of PAE.
Limitations: Single centre retrospective trial limits the scope of this study.
Ethics committee - additional information: The ethical approval was received from the Institutional ethics committee with a
waiver for informed consent.
12-month outcomes of MRI-guided transurethral ultrasound ablation (TULSA) for the treatment of benign prostatic
hyperplasia (7 min)
Pouya Doerwald; Hamburg / Germany
817
Author Block: A. Viitala1, M. Anttinen1, P. Mäkelä1, P. Doerwald2, P. Nurminen1, H. E. Pärssinen1, T. Sainio1, R. Blanco Sequeiros1, P. J.
Boström1; 1Turku/FI, 2Hamburg/DE
Purpose: The aim of this study was to assess clinical outcomes of MRI-guided transurethral ultrasound ablation (TULSA) in treating
benign prostatic hyperplasia (BPH).
Methods or Background: Men with BPH scheduled for TURP were enrolled. EPIC-26, IPSS, IIEF-5, uroflowmetry, PSA, MRI, and
complications in Clavien-Dindo were recorded.
Results or Findings: Thirty patients received TULSA, with a median follow-up of 16-mo (max 48-mo). At baseline, median [IQR] age
was 67 years [64–72], PSA 3.1 ug/l [2.2–6.9], prostate volume 51.5 ml [min 29-max 107], average flow rate 4.1 ml/s [3.4–7],
maximum flow rate 11 ml/s [8–15], voided volume 211 ml [139–336], and post-void residual(PVR) 71 ml [40–247]. Median sonication,
hospitalization, and catheterization times were 39 min, 24 h, and 16 d, respectively. Complications included one grade 1, 8 grade 2,
and one grade 3 events, all resolving within three-months. 24/30 patients have completed their 12-month follow-up with two men
dropping off due to cancer diagnosis. Between baseline and 12 months, median values for measured parameters changed as follows:
prostate volume, PSA, PVR reduced from 51.5 to 31 ml, 3.1 to 1.5 ug/l, and 71 to 41.5 ml, respectively. Average flow rate, Qmax, and
voided volume increased from 4.1 to 8.7 ml/s, 11 to 18 ml/s, 211 to 301 ml, respectively. IPSS, IPSS QoL, IIEF-5 scores improved from
16.5 to 4, 4 to 1, and 15 to 19, respectively. EPIC-26 urinary incontinence, irritative/obstructive, bowel, sexual, and hormonal domains
improved from 85.5 to 100, 65.6 to 94, 87.5 to 100, 54 to 67, and 95 to 100, respectively. Results of uroflowmetry, functional, and
QOL questionnaires all improved at 12-month despite discontinuation of LUTS medication in 26/30 patients.
Conclusion: 12-mo clinical-outcomes demonstrate safety and efficacy of TULSA for treating BPH.
Limitations: Small population
Funding for this study: Study was funded by Profound Medical.
Ethics committee - additional information: Ethics committee approval was granted and informed consent was obtained from all
study participants.
Recurrence rates of male varicocele after endovascular treatment using NBCA-MS (Glubran2): a retrospective study (7
min)
Alessandra Mininni; Milan / Italy
Author Block: A. Mininni, L. Tomasino, E. Lanza, G. Ferrillo, D. Poretti, M. Francone, V. Pedicini; Milan/IT
Purpose: Male varicocele is characterised by high prevalence in young individuals, diminished fertility rates and adverse impact on
the quality of life. The aim of this study is to evaluate the recurrence rate of varicocele in patients who had undergone endovascular
treatment using N-butyl 2 cyanoacrylate plus methacryloxy sulfolane (NBCA-MS, Glubran2) glue.
Methods or Background: A retrospective analysis of varicocele embolisation procedures completed between January 2016 and
December 2022 was performed. A total of 102 patients were recruited (mean age 27.7 years) through phone interview. Before the
procedure 10.8% patients had stage two disease severity; 61.8% had stage three and 27.5% had stage four. All patients manifested
typical varicocele symptoms such as testicular heaviness, pain or spermiogram abnormalities. Improvement in spermiogram,
recurrence rate of disease and need for new treatment were evaluated.
Results or Findings: Spermiogram improvement was observed in 81.3% of patients (49 patients underwent semen analysis 6-8
months after the procedure). Recurrence rate of disease was 10.8% (11 patients) and among those, patients requiring new treatment
were 45.5% (three patients needed new endovascular embolisation treatment using NBCA-MS, and two patients needed classic
surgery). No significant post-procedural complications were reported.
Conclusion: Endovascular embolisation using NBCA-MS glue is an effective treatment for male varicocele, offering very low
recurrence rates and negligible complications.
Limitations: The single-centre and retrospective nature of the study were identified limitations.
Ethics committee - additional information: IRB approval was waived considering the retrospective non-interventional study
design.
Endovascular embolisation for high-flow priapism in paediatric patients (7 min)

Edanur Karapinar; Istanbul / Turkey
818
Author Block: E. Karapinar, C. C. Ercan, M. S. Çakır, A. B. Aydin; Istanbul/TR
Purpose: High-flow priapism is a persistent penile tumescence, which is relatively uncommon compared to low-flow priapism. It is
typically associated with a history of trauma leading to lacerations in the cavernosal arteries within the corpora cavernosa. Super-
selective arterial embolisation is a well-established therapeutic technique; however, conservative management is usually the initial
approach in paediatric population. As a tertiary medical centre, we are compelled to provide treatment to those who do not respond
to conservative measures.
Methods or Background: We observed a total of four paediatric patients with high-flow priapism between 2020-2022. Three of
these patients had documented trauma histories, while the cause in one patient remained uncertain due to their mental status. All
patients underwent conservative treatment for up to two weeks, with no observed response. Initially, we conducted Doppler
ultrasound examinations to assess arterial waveforms and identify fistula tracts in the corpora cavernosa. Subsequently, we
performed conventional angiography and proceeded with an embolisation plan.
Results or Findings: Among the patients, three exhibited angiographic evidence of arteriocavernosal fistulas. Super-selective
embolisation with gel-foam was successfully completed in all patients, and post-procedural angiography revealed the absence of
residual pathological flow. In long-term follow-up, all of these patients exhibited recovery with no reported complications overtime.
Early postoperative and subsequent intermittent follow-up Doppler studies have consistently demonstrated normal flow. The fourth
patient, who had an uncertain history of trauma, exhibited normal angiographic features and was deemed ineligible for embolisation.
Conclusion: Super-selective embolisation emerges as a promising technique for managing high-flow priapism not only in adults but
also in paediatric patients.
Limitations: Given the rarity of this condition, our study was limited by the small number of patients.
Ethics committee - additional information: We adapted a treatment modality with proven successful outcomes in the literature
to our own patients under the most suitable conditions. We did not employ any additional medications or medical interventions
beyond the methods documented in the literature.
819
RPS 1910 - CT applications in musculoskeletal imaging
Categories: Imaging Methods, Musculoskeletal, Physics in Medical Imaging

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:
Simone Waldt; Essen / Germany
Value-added opportunistic chest CT screening for osteoporosis using bone mineral density estimation computed by
multi-view semi-supervised learning (7 min)
Jinrong Yang; Wuhan / China
Author Block: J. Yang1, H. Guo2, L. Le2, M. Xu2, H. Shuishi1, F. Yang1; 1Wuhan/CN, 2Hangzhou/CN
Purpose: The aim of the study was to evaluate the value of bone mineral density estimation computed by multi-view semi-
supervised learning on chest CT for opportunistic osteoporosis screening.
Methods or Background: A total of 1295 patients who underwent chest CT for lung cancer screening and received dual-energy X-
ray absorptiometry (DXA) during the same period were collected. They were divided into three groups: normal group, osteopenia
group and osteoporosis group based on bone mineral density measured by DXA. And a new proposed indicator—bone mineral density
estimation of all thoracic vertebrae and the first lumbar vertebra were automatically computed by multi-view semi-supervised
learning on chest CT images. The differences of bone mineral density estimation were compared among the three groups. And the
diagnostic efficacy for distinguishing osteopenia and osteoporosis from normal group were also evaluated.
Results or Findings: The bone mineral density estimation of thoracic and first lumbar vertebrae was significantly different among
three groups and between any two groups (P<0.001), and decreased successively in normal group, osteopenia group and
osteoporosis group. They were positively correlated with t-values of the first lumbar vertebrae (R = 0.58-0.77, P<0.01). And they had
high diagnostic efficacy for distinguishing osteopenia and osteoporosis from normal group (AUC = 0.777~0.824).
Conclusion: The bone mineral density estimation of thoracic and first lumbar vertebrae, which computed by multi-view semi-
supervised learning on chest CT, is of great value in the opportunistic screening of osteopenia and osteoporosis.
Limitations: The new marker—bone mineral density estimation can help patients realize the value-added value of chest CT, that is,
during a chest CT examination, not only lung cancer can be screened, but also osteoporosis opportunistic screening can be realized.
However, it was a single-center, retrospective study.
Ethics committee - additional information: The ethical approval was obtained by the Ethics Committee of Wuhan Union Hospital
([2021]0853).
Multiparametric contrast-enhanced dual-energy CT as a one-stop shop approach in arthritis imaging-first results of the
HAI-DECT study (7 min)
Sevtap Tugce Ulas; Berlin / Germany
820
Author Block: S. T. Ulas, K. Ziegeler, J. Mews, U. Schneider, S. Ohrndorf, R. Biesen, D. Poddubnyy, F. N. Proft, T. Diekhoff; Berlin/DE
Purpose: This study aimed to evaluate the efficacy and diagnostic utility of contrast-enhanced dual-energy computed tomography
(CE-DECT) in detecting and differentiating rheumatic joint diseases.
Methods or Background: Patients with suspected hand arthritis were prospectively enrolled. All underwent CE-DECT imaging
alongside the standard clinical evaluation. CE-DECT (80/135 kVp) was captured pre and three-minutes post weight-adapted contrast
agent application. Reconstructions included two-material decomposition for tophus imaging, virtual non-calcium for bone marrow
edema, and CT-subtraction for soft-tissue inflammation. Diagnoses using CE-DECT were juxtaposed with the initial and final
evaluations by referring rheumatologists. A subsequent survey assessed CE-DECT‘s diagnostic utility and impact on patient
management, rated on a 1-10 scale. Descriptive statistics were employed.
Results or Findings: Out of 67 patients, diagnoses included rheumatoid arthritis (29), psoriatic arthritis (8), crystal disease (13) and
others, for example osteoarthritis and peripheral spondyloarthritis (17). In 57 cases (85%), CE-DECT diagnoses aligned with final
clinical evaluations. CE-DECT altered clinical diagnoses for 21 patients (31%). Rheumatologists rated CE-DECT‘s diagnostic utility at
an average of 8.31 (SD 2.49) and its contribution to patient management at 8.61 (SD 2.19).
Conclusion: CE-DECT offers standardized arthritis imaging. Its diverse diagnostic capabilities make it a potent tool for early diagnosis
in rheumatic joint diseases.
Ethics committee - additional information: The study was approved by the local ethics committee (EA1/183/21) and the Federal
Office for Radiation Protection (ZD 3-22464/2021-285-G).
Risk factors for compression fractures in patients treated with gastrectomy: emphasis on bone mineral density
measured on opportunistic CT (7 min)
Suiji Lee; Seoul / Korea, Republic of
Author Block: S. Lee, K. W. Kim; Seoul/KR

Purpose: The aim of this study was to assess apply osteosarcopenia measurement methods in CT patients. Osteosarcopenia is the
coexistence of osteoporosis and sarcopenia, which has a significant impact on the survival outcomes of critically ill patients. Many
studies tried to diagnose osteosarcopenia using opportunistic CT. However, in case of patients treated with gastrectomy, the loss of
bone mineral density and muscle mass is not routinely evaluated, even though these patients often experience bone and muscle
shrinkage and subsequent compression fractures after surgery. Therefore, we assessed bone mineral density (BMD) and skeletal
muscle area (SMA) in these patients by applying osteosarcopenia measurement methods in opportunistic abdominal CT.
Methods or Background: Among 2457 patients who had undergone gastrectomy between January 2003 and December 2010, we
included 100 patients (mean age 58 years). Preoperative and 1-year postoperative abdominal CT scans were evaluated for BMD and
SMA. Region of interest was situated in L1 vertebral body to estimate trabecular attenuation value to analyze BMD and SMA. To
determine predictive risk factors for compression fractures, we implemented univariate logistic regression analysis.
Results or Findings: Between preoperative and postoperative CT, all patients showed a decline in the SMA (116 ± 27 cm2 vs. 110 ±
25 cm2, p<0.0001) and the BMD (177 ± 50 HU vs. 166 ± 50 HU, p<0.0001). Body weight (60 ± 9 kg vs. 54 ± 9 kg, p<0.0001) and
BMI (24 ± 3 vs. 21 ± 3, p<0.0001) were also decreased. During the five-year follow-up record, five patients experienced osteoporotic
vertebral compression fractures. Univariate analysis showed preoperative and postoperative BMD were significant risk factors for
compression fractures, apart from age, sex, preoperative and postoperative SMA.
Conclusion: Preoperative and postoperative BMD measured on opportunistic CT are significant risk factors for compression fractures
in patients treated with gastrectomy.
Limitations: Single center study and retrospective design.
Ethics committee - additional information: The study was approved by the IRB with the approval code: 2023-0976.
Photon-counting detector computed tomography (PCD-CT) for quantitative microstructural imaging of bone (7 min)
821
Author Block: A. Burghardt1, A. Korajac2, A. E. Strassl2, S. Sadoughi1, G. Kazakia1, J. M. Patsch2; 1San Francisco, CA/US, 2Vienna/AT
Purpose: The goal of this study was to evaluate feasibility, performance and accuracy of PCD-CT for bone structure quantification.
The quantification of bone quality by high-resolution research CT scanners (built for the peripheral skeleton) has been shown to
improve fracture risk prediction. Now, photon-counting detector CT (PCD-CT) has been introduced as a clinical imaging tool with a
generational advance in imaging performance.
Methods or Background: Cadaveric phantoms (n= 20 1-cm metaphyseal sections of distal radius and tibia) were imaged by μCT,
1st- and 2nd-generation HR-pQCT, and PCD-CT. PCD-CT scans were acquired at 120kVp in Ultra-High Resolution (UHR) mode
(200mAs, 16.1mGy). A 15mm FOV was reconstructed across 1024x1024 matrix using the Br92 kernel (146μm pixel size, 150μm slice
thickness). Spatial resolution (10% MTF) and noise performance (SNR) were measured in idealized density and wire phantoms.
Standard density, geometry, and structure measures were quantified and evaluated for accuracy by linear regression & Bland-Altman
analysis.
Results or Findings: PCD-CT provided an exquisite depiction of cortical and trabecular microstructure. PCD-CT noise performance
(SNR=20.5) was superior to 1st- and 2nd-gen HR-pQCT research scanners (SNR=15.4 and 7.1). The spatial resolution of PCD-CT (10%
MTF=132 μm) was comparable to 1st-gen HR-pQCT (10% MTF=138 μm), and inferior to 2nd-gen HR-pQCT (10% MTF=95 μm).
Accuracy of PCD-CT was high (R2>0.9) for trabecular bone volume and trabecular number, and comparable to 2nd-gen HR-pQCT.
Conclusion: PCD-CT provides significant advantages for measuring human bone structure on a clinical device, including superior
noise performance and scan time, while spatial resolution is comparable to 1st-gen HR-pQCT and accuracy was comparable to 2nd-
gen HR-pQCT. PCD-CT represents an exciting opportunity to translate quantitative microstructural skeletal imaging to a clinical
platform.
Limitations: Actual in vivo-translation (= in-vivo scanning) is work in progress.
Ethics committee - additional information: Approval has been obtained for specimen donation and use.
The application value of spectral multi-parameter model in bone islands and osteoblastic metastatic tumours (7 min)
Author Block: Y. Zhou, L. Lei, Z. Wang, W. Cao, S. Dong, S. W. Yue; Zhengzhou/CN

Purpose: This study aims to investigate the diagnostic value of spectral CT multi-parameter imaging in the differentiation between
bone islands (BIs) and osteoblastic metastases (OBMs).
Methods or Background: We retrospectively collected data from a total of 119 patients (60 with BIs and 59 with OBMs) who
underwent chest and abdomen plain scans plus dual-phase enhanced scans using a dual-layer spectral detector CT. The spectral data
obtained from plain scans were processed to generate images of mixed energy (CI), effective atomic number (Z-eff), electron density
(DE), virtual monochromatic images (VMI) at 40/100 keV, and X-ray attenuation curve slope (λHU). The spectral data obtained from
dual-phase enhanced scans were used to generate CI, iodine-water map (IoW), Z-eff, DE, VMI at 40/100 keV, and λHU.
Results or Findings: The spectral parameters from plain and enhanced scans showed significant differences between the BIs and
OBMs groups (P < 0.001). Multivariate logistic regression analysis revealed that Z-eff in the plain phase, Z-eff and 40/100 keV VMI in
the arterial phase, λHU in the arterial phase, IoW and Z-eff in the venous phase can be used as independent factors in differentiating
BIs from OBMs. The established three-phase spectral multi-parameter joint model for the differential diagnosis of bone islands and
osteoblastic metastases had an AUC of 0.953, which was superior to that of three-phase CI images (AUC = 0.833-0.837) and various
phase joint models (AUC = 0.884-0.928).
Conclusion: The combined spectral multi-parameter model provides higher diagnostic value in differentiating between BIs and OBMs
compared to conventional CT values and individual spectral parameters. It offers more imaging support for the early clinical
diagnosis, treatment strategy formulation, and prognosis prediction of osteoblastic metastases in cancer patients.
Limitations: The sample size in this research is relatively limited.
Prediction of osteoporosis associated fractures: evaluation of dual-energy CT-derived metrics of the lumbar spine (7
min)
822
Author Block: L. D. Grünewald, V. Koch, S. Mahmoudi, P. Reschke, S. Martin, C. Booz, I. Yel, T. Vogl; Frankfurt a. Main/DE
Purpose: To assess the association of dual-energy CT (DECT)-derived bone mineral density (BMD) assessment with the occurrence of
acute insufficiency fractures of the spine and follow-up fractures in a 2-year period.
Methods or Background: L1 of 160 patients (77 men, 83 women; mean age, 64.1 years, range, 19-94 years) who underwent dual-
source DECT between 01/2016 and 12/2020 was retrospectively analyzed. For phantomless BMD assessment, a dedicated DECT
postprocessing software based on material decomposition was manually applied. All depicted vertebrae were examined for signs of
recent insufficiency fractures, and electronic health records were examined to obtain the incidence of osteoporotic fractures for a
follow-up of 2 years after DECT. Receiver-operating characteristic (ROC) analysis was used to calculate AUC values, and logistic
regression models were used to determine the associations of BMD, sex, and age with the occurrence of insufficiency fractures and
follow-up fractures.
Results or Findings: A DECT-derived BMD threshold of 120.40 mg/cm³ yielded an AUC of 0.82 (p < .0001) to identify patients with
one or more insufficiency fractures of the spine from patients without fracture, and a DECT-derived BMD cut-off of 93.70 mg/cm³
yielded an AUC of 0.9373 (CI, 0.867-0.977, p < .001) for the prediction of follow-up fractures within 2
two years after DECT. A lower DECT-derived BMD was associated with an increased risk to sustain insufficiency (Odds ratio of 0.93,
95% CI, 0.91-0.96, p < .001), and follow-up fractures (Odds ratio of 0.8710, 95% CI, 0.091-0.9375, p < .001), indicating a protective
effect of increased DECT-derived BMD.
Conclusion: Dual-energy CT-derived bone mineral density assessment from routine examinations can be used to stratify the risk of
sustaining osteoporosis-associated fractures following acute trauma and during a follow-up period of 2 years.
Limitations: Preselection Bias, Limited to Dual-Energy-CT, Retrospective Study
Ethics committee - additional information: Informed consent waived due to retrospective nature.
Cone beam CT in the emergency department: revising incidence rates for carpal bone fractures (7 min)
Sophie Murphy; Dublin / Ireland
Author Block: S. Murphy, E. McDermott, B. Gibney, P. J. Macmahon; Dublin/IE

Purpose: The aim of this study was to evaluate and quantify the incidence of acute radio-carpal fractures when routinely employing
CBCT in a major trauma centre.
Methods or Background: In our institution Cone beam computed tomography (CBCT) is routinely utilised in the Emergency
Department (ED) to investigate patients who have suspected acute radio-carpal fractures but negative x-rays. We have noted that the
frequency of wrist fractures as published in the literature does not match the frequency encountered in clinical practice when using
CBCT to investigate these patients. A retrospective analysis was conducted on all CBCT scans performed for wrist trauma in the ED
between January 2018 and August 2023. Fracture incidence, type, and specific carpal bone locations were characterized.
Results or Findings: A total of 690 wrist CTCBs were performed on ED patients across the study period.
52% were female with a mean age of 44.5 years. Forty four percent (306/690) were positive for a fracture even though x-rays were
negative, and 24% (72/306) of positive studies had more than one fracture identified. A total of 386 discrete fractures were identified
and characterised like this: 62% Carpal bones (240), 24% Distal radius (93), 10%Metacarpal bone (38), 4% Distal Ulna (15). Of the
240 carpal bone fractures: 29%Trapezium (70), 29% Scaphoid (69), 22% Triquetral/triquetrum (53), 8% Hamate (18), 4% Pisiform (9),
3% Trapezoid (8), 3% Capitate (7), 3% Lunate (6).
Conclusion: Our data indicates that a revision in the conventional carpal bone fracture incidence estimates is needed, suggesting a
much higher frequency of certain carpal bone fractures (esp trapezium) when CBCT is used as the imaging modality in wrist trauma
assessment.
Limitations: Single centre study.
Potential of dual-energy CT-based collagen maps for the assessment of thoracic disc degeneration (7 min)
823
Author Block: S. Mahmoudi, L. D. Grünewald, V. Koch, J-E. Scholtz, S. Martin, D. Pinto Dos Santos, C. Booz, T. Vogl, I. Yel; Frankfurt a.
Main/DE
Purpose: The objective of this study was to assess the potential of Dual-Energy CT (DECT)-derived collagen maps for the assessment
of disc degeneration in the thoracic spine.
Methods or Background: We retrospectively included 51 patients who received both dual-source DECT (third generation dual-
source DECT; Somatom Force; Siemens Healthineers) and MRI of the thoracic spine within two weeks between July 2020 and October
2022. Disc degeneration was independently evaluated by two blinded readers in DECT-derived collagen maps. Based on the Pfirmann
grading system, thoracic disc generation was classified into no/mild (Pfirmann Grade 1&2), moderate (Pfirmann Grade 3&4), and
severe (Pfirmann Grade 5). To assess the diagnostic value of DECT-derived color-coded collagen maps for the evaluation of thoracic
disc degeneration, we calculated diagnostic accuracy, sensitivity and specificity using MRI as reference standard. 5-point Likert scales
were used to evaluate diagnostic confidence (1=insufficient,2=low,3=moderate,4=high,5=excellent) and image quality (1=non-
diagnostic,2=weak,3=moderate,4=good,5=excellent) subjectively.
Results or Findings: In total, 612 intervertebral discs from 51 patients were evaluated (mean age, 68±16 years, 23 female). MRI
depicted 135 non/mildly degenerated discs (22.1%), 470 moderately degenerated discs (76.8%), and seven severely degenerated
discs (1.1%). Collagen maps achieved a high level of sensitivity (792 of 954 [83.0 %], specificity (221 of 270 [81.9 %]), and diagnostic
accuracy (1013 of 1215 [83.4 %]) to discriminate between non/mild and moderate/severe thoracic disc degeneration. Inter-reader
reliability was substantial (κ=0.82) for DECT collagen maps (p<.001). The subjective evaluation of disc degeneration using DECT-
derived collagen maps showed moderate to high diagnostic confidence (median 3.5, interquartile range (IQR) 3-4) and moderate to
good image quality (median 3.5, IQR 3.5-4).
Conclusion: The ability of DECT-derived collagen maps to evaluate the degree of disc degeneration may facilitate assessment of
thoracic disc degeneration in patients with disc-related pathologies when MRI is not available.
Ethics committee - additional information: The ethical review board of our institution approved this retrospective study and
waived written informed consent.
824
E³ 1926 - Imaging of stroke: what should you know in 2024?
Categories: Artificial Intelligence & Machine Learning, Emergency Imaging, Interventional Radiology, Neuro, Vascular
Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

State-of-the-art diagnosis: role of MRI combined with AI (15 min)

Georgina Gáti; Pécel / Hungary
1. To summarise currently available clinical application tools for stroke imaging workflow in MRI.
2. To discuss AI tools in the research phase for stroke imaging in MRI.
3. To discuss the ASPECTS calculation using DWI.
4. To discuss patient management in stroke imaging.
State-of-the-art diagnosis: role of CTA combined with AI (15 min)

Paul M. Parizel; Perth / Australia
1. To summarise currently available commercial AI tools for stroke imaging workflow.

2. To discuss AI tools in the research phase for stroke imaging.
3. To discuss future research and development areas for AI in stroke imaging.
What is new in endovascular management? (15 min)

Roberto Gandini; Rome / Italy
1. To discuss the possibility of treating distal vessels.

2. To discuss the recent devices for the treatment of stroke.
3. To consider the potential role of stents in case of repeated unsuccessful steps.
Panel discussion: Logistical challenges in stroke diagnosis and treatment. Is there room for improvement? (10 min)
825
RC 1913 - Physics of photon-counting CT: game changer or incremental progress?
Categories: EuroSafe Imaging/Radiation Protection, Imaging Methods, Physics in Medical Imaging

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Technical possibilities of photon-counting CT (15 min)

Marc Kachelrieß; Heidelberg / Germany
1. To learn about the technologies behind photon-counting CT.

2. To learn about new developments.
3. To become familiar with how new protocols can be implemented in a practical manner.
Image quality and dose in photon-counting CT (15 min)

1. To learn about image quality implications in photon-counting CT.

2. To understand if the potential dose reductions are real or not.
3. To become familiar with how to optimise photon-counting CT in a practical manner.
Clinical practice in photon-counting CT (15 min)

Dominik Juskanic; Nitra / Slovakia
1. To describe the implementation of optimised protocols for the new clinical uses of photon-counting CT.
2. To understand the effect that this has on clinical practice.
3. To look toward future trends in this area.
Panel discussion: Photon-counting CT: game changer or incremental progress? (10 min)
826
RC 1901 - Emergency! Two patients, not one: abdominal pain in pregnancy
Categories: Abdominal Viscera, Emergency Imaging, GI Tract

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Liver, gallbladder, pancreas and bile duct disorders (15 min)

Marco Dioguardi Burgio; Levallois Perret / France
1. To know the imaging features of the most typical disorders affecting the hepatopancreatobiliary system in pregnancy.
2. To understand the relevant limitations of imaging techniques (US, CT, MRI)
3. To understand how to optimise imaging in hepatopancreatobiliary disease in pregnancy.
Appendix and bowel disorders (15 min)

Vikas Shah; Leicester / United Kingdom
1. To know the imaging features of the most typical disorders affecting the appendix and bowel in pregnancy.
2. To understand the imaging pitfalls when imaging the appendix and bowel during pregnancy (US, CT, MRI)
3. To understand how to optimise imaging in the appendix and bowel disease in pregnancy.
Quantifying the risks of cross-sectional imaging in an emergency (15 min)

Jeannette Kraft; Leeds / United Kingdom
1. To understand the risks of radiation to the foetus during pregnancy.

2. To understand scenarios where emergency CT imaging is of greatest benefit in pregnancy.
3. To know the key information to include in discussions with clinicians and patients.
Panel discussion: Imaging as a problem-solving tool? (10 min)
827
ESR Research - Next-generation researchers: here's how to get started!
Categories: Management/Leadership, Multidisciplinary, Professional Issues, Research, Students

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Getting funded (5 min)

Johannes Uhlig; Göttingen / Germany
1. To learn how to apply for funding.

2. To appreciate the value of seed funding.
3. To understand what funding scheme is good for you.
Finding a mentor (5 min)

1. To learn what to look for in a mentor.

2. To appreciate the value of a mentor.
3. To understand how to get the most out of a mentor.
Going abroad (5 min)

Oğuz Lafcı; Ankara / Turkey
1. To learn how and where to find opportunities for doing a research fellowship/visit.
2. To appreciate the value of a research fellowship/visit.
3. To understand how to approach a foreign institution and build a connection.
Building a research line (5 min)

Alejandro Rovira Cañellas; Barcelona / Spain
1. To learn how to find your niche topic and make it yours.

2. To appreciate the value of multidisciplinary collaboration.
3. To understand the ingredients of a successful research line.
Involving patients in research projects (5 min)

1. To appreciate the importance of involving patients in your research.

2. To learn how to involve patients in the various stages of research.
3. To understand what patients can add to the various aspects of a research project.
828
Interactive panel discussion: How do/did you take your first steps in the research domain? (33 min)
829
ESR/ESSO 19 - From imaging to cancer surgery
Categories: Abdominal Viscera, Breast, GI Tract, Multidisciplinary, Oncologic Imaging

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderators:

Imaging to help decisions about colorectal cancer surgery: impact on surgical options (28 min)
Per J. Nilsson; Stockholm / Sweden
1. To know the relevant clinical questions in the management of colorectal cancer patients.
2. To understand which and how imaging findings impact treatment decisions.
3. To know how imaging guides the surgical approach.
Screening and early detection in breast cancer: impact on non-palpable breast-guided surgery (28 min)
1. To understand the new guidelines on screening for breast cancer.

2. To plan surgical treatments depending on breast imaging outcomes.
3. To help building a strong relationship with radiologists to improve the management of breast cancer patients.
830
OF 19R - Developing image perception skills for enhanced practice
Categories: Education, Imaging Informatics, Imaging Methods, Radiographers, Research

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
The aim of this session is to offer a comprehensive exploration of the critical aspects of image perception in the field of radiography.
This session comprises three insightful talks that collectively provide attendees with a deep understanding of the fundamental
concepts, practical skills, and the role of research in advancing image perception proficiency. The multifaceted aspects of developing
image perception skills will be explored and explained, which should be appealing to both novice and more experienced
radiographers and health professionals. Indeed, this session will offer valuable knowledge and practical insights to enhance one's
proficiency in image perception, ultimately contributing to improved patient care.
Moderator:

Medical image perception: key concepts for radiographers (16 min)

Rachel Toomey; Dublin / Ireland
Developing skills in image interpretation: insights and tips from a reporting radiographer (16 min)
Jarno Tapani Huhtanen; Raisio / Finland
The importance of research in understanding medical image perception (16 min)

Tim Donovan; Lancaster / United Kingdom
831
ESR/EORTC - Next-generation imaging: clinical trials and AI
Categories: Artificial Intelligence & Machine Learning

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderators:
Luc Bidaut; Tayport / United Kingdom

Luc Bidaut; Tayport / United Kingdom
Large language models in radiology and beyond (15 min)

Christian Blüthgen; Zürich / Switzerland
1. To understand the foundations of language modelling.

2. To learn about application scenarios of large language models in radiology.
3. To learn about current and future developments in multimodal deep learning.
How to validate AI in healthcare? (15 min)

1. To understand the importance of careful validation of AI tools in the context of data drift and data shift.
2. To discuss potential approaches to post-market surveillance of AI tools.
3. To learn about potential pitfalls in human-machine interaction and their impact on AI tools.
AI in hybrid imaging (15 min)

1. To learn about the potentials and caveats of AI in clinical HI.

2. To appreciate the promise of AI and the responsibilities of the HI user community.
3. To understand the role of AI in the entire workflow of HI (from data acquisition to data corrections and quantification, as well as
data handling and prediction models building on HI).
AI for oncological follow-up (15 min)

Andrea Schenk; Bremen / Germany
1. To learn about the potential benefits of AI for oncological follow-up.

2. To understand the role of AI in the workflow of tumour monitoring.
3. To discuss challenges and potential pitfalls when using AI in oncological patients.
Patient's perspective: collecting high-quality data to benefit the patient (15 min)
832
1. To acknowledge why the quality of data collected plays a major role for patients on the one hand but for all stakeholders on the
other hand regarding environment, costs, optimisation, evaluation, usability and acceptance, safety, and treatment progress.
2. To understand what we can learn and copy from other industries successfully implementing AI and deep learning.
3. To address misleading AI use of patients like Google and ChatGPT and to use tools to build awareness to avoid this.
Panel discussion: Challenges and opportunities of AI in clinical trials (10 min)
833
RC 1904 - Technical advances in thoracic imaging
Categories: Chest, Imaging Methods

Date: March 2, 2024 | 12:30 - 13:30 CET
CME Credits: 1
Moderator:

Protons and more: the rise of lung MRI (15 min)

Mark O. Wielpütz; Heidelberg / Germany
1. To describe the current state-of-the-art of technology and methodology.

2. To integrate MRI in clinical practice for thoracic imaging.
3. To judge the future impact of current translation research.
Positrons and more: the further rise of nuclear medicine (15 min)
Tim van den Wyngaert; Edegem / Belgium
1. To learn about changes in PET-CT and nuclear medicine.

2. To understand how and when to use hybrid imaging to diagnose thoracic diseases.
3. To understand how and when to use hybrid imaging for response assessment of thoracic diseases.
Photons illuminating the dark: potential of spectral thoracic imaging (15 min)
1. To understand the technical principles of spectral imaging, including photon-counting detector CT.
2. To evaluate the capabilities of spectral thoracic imaging.
3. To assess the challenges of implementing spectral thoracic imaging into routine clinical practice.
Panel discussion: How do patients profit from technical advances (10 min)
834
ST 17 - EuroSafe Imaging: celebrating 10 years of success in radiation protection
Categories: EuroSafe Imaging/Radiation Protection

Date: March 2, 2024 | 12:45 - 13:30 CET
In this session, joins us as our speakers discuss EuroSafe Imaging, the ESR’s flagship initiative on radiation protection.
As EuroSafe Imaging is celebrating its 10 year anniversary at ECR 2024, we will take a look back at some of its successes over the
past decade as well as discussing current initiatives, such as its educational actvities that aim to support and strengthen medical
radiation protection across Europe and beyond.
We will also highlight one of the most recent European projects related to radiation protection led by the ESR: the European co-
ordinated action on improving justification of computed tomography (EU-JUST-CT) project.
Moderators:
Interview (45 min)

Reinhard W.R. Loose; Nürnberg / Germany
Deniz Akata; Ankara / Turkey
835
EFRS 9 - Quiz for Radiographers/Technologists not specialized in Nuclear Medicine
Categories: Nuclear Medicine, Professional Issues, Radiographers

ETC Level: LEVEL I
Date: March 2, 2024 | 13:00 - 14:00 CET
This session is an excellent opportunity and time to introduce EFRS Nuclear Medicine Committee, get to know colleagues, and explore
possibilities for committee integration.
Moderator:

Quiz for Radiographers/Technologists not specialized in Nuclear Medicine (50 min)
Closing (5 min)
836
CUBE 22 - Opportunities to improve patient safety during neurological IR procedures

Date: March 2, 2024 | 13:00 - 13:30 CET
Opportunities to improve patient safety during neurological IR procedures (30 min)

Jose Jr Binghay; Dublin / Ireland
1. To explain why patient safety is of paramount importance in neurological IR procedures.

2. To discuss the role of the radiographer in maximising patient safety.
3. To discuss how technology might help improve patient safety neuro IR procedures.
837
HW 20Cc - Cardiac inflammation: imaging insights and reporting strategies
Categories: Cardiac, Imaging Methods

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:


Ausami Abbas; Sherbourne st.john / United Kingdom
Luigia D'Errico; Cambridge / United Kingdom
1. To become familiar with typical and atypical imaging findings of inflammatory diseases of the myocardium and pericardium.
2. To become familiar with imaging criteria and other supporting diagnostic modalities.
3. To discuss the limits and technical drawbacks of cardiac MRI and CT.
4. To learn how to report cardiac MRI and CT using specific templates.
838

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1


Tutors
839
S 20 - Students Session 2
Categories: Cardiac, Chest, Interventional Radiology, Neuro, Oncologic Imaging, Students

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Automated detection of prostate cancer biomarkers with pseudo-diffusion techniques in 3.0 Tesla MR (8 min)
Pablo Baltasar Irusta; Mendoza / Argentina
Author Block: P. B. Irusta1, T. Gonzalez Padin1, R. N. Alcalá Marañón1, M. Noceti1, F. Gonzalez Nicolini1, T. Doring2, E. Figueiredo2, D. A.
Romero3; 1Mendoza/AR, 2Rio de Janeiro/BR, 3San Carlos de Bariloche/AR
Purpose: Prostate cancer (PCa) diagnosis by multiparametric MRI (mpMRI) encompasses wT2, DWI and DCE sequences. The aim of
this work was to obtain significant perfusion biomarkers through IVIM-KURTOSIS (IK) with an automated pseudo-diffusion analysis
achieved by post-processing DWI images employing an MR system.
Methods or Background: The protocol involved 36 male subjects using a 3.0 Tesla PET/MR scanner. For the pseudo-diffusion
analysis, a DWI sequence with 19 b-values (between zero and 2400 s/mm2) was implemented. Employing an in-house Python
algorithm (v3.7.3), preprocessing and fitting were performed. Post-processing consisted of a three-step biexponential function fit, with
automatic boundary recognition for IVIM/ADC/KURTOSIS regions and b-value thresholds for SNR. Images of DCEs (9-second temporal
resolution) underwent processing using the manufacturer's workstation; IK and DCE maps were compared using the Pearson test
(r>0.5).
Results or Findings: Maximum b-values were constrained due to noise presence, ranging between 1900 and 2100 s/mm2 (patient
dependent). The automated algorithm established region limits at 300 and 1000 s/mm2. Regarding Pearson’s coefficient, the pseudo-
diffusion coefficient (D*) exhibited values of 0.88 and 0.6 for Ktrans and IAUGC, respectively, while the perfusion fraction coefficient
(f) yielded 0.53 and 0.55 values. Comparing diffusion coefficient (D) and KURTOSIS (K) to IAUGC resulted in 0.58 and 0.26,
respectively. Furthermore, f·D*, in comparison to Ktrans and IAUGC, showed results of 0.93 and 0.73.
Conclusion: The b-value threshold for SNR was organ-proportional, as in the bladder, it decreased for the KURTOSIS region.
Pearson’s coefficients indicate that IK coefficients could be considered as relevant biomarkers for the carcinoma. Obtained IK
biomarkers were comparable to DCE-acquired information, given its potential inconveniences (renal dysfunction, elevated cost and
vascular problems), minimising its impact.
Funding for this study: This study was performed under a research agreement with General Electric Healthcare.
Ethics committee - additional information: The study was approved with the project code: 23FFUA-A09.
Spectral computed tomography differentiates airway wall enhancement patterns in typical and atypical pneumonia (8
min)
Johanna Thomä; Heidelberg / Germany
840
Author Block: J. Thomä, P. Konietzke, A. Bodenberger, T. F. Weber, T. D. Do, H-U. Kauczor, W. Stiller, O. Weinheimer, M. Wielpütz;
Heidelberg/DE
Purpose: Previous studies in healthy individuals demonstrated that airway wall enhancement can be quantified on virtual
monochromatic reconstructions from spectral computed tomography (CT) by calculating the spectral attenuation curve’s slope based
on Hounsfield Units (λHU) for wall density in segmented airways. This technique has the potential to detect airway wall inflammation
in airway diseases. Thus, we aimed to differentiate patients with typical and atypical pneumonia as well as COVID-19 pneumonia from
healthy individuals based on λHU.
Methods or Background: 432 patients (mean age 58.9±17.2yrs) who underwent contrast-enhanced dual-layer detector spectral CT
(1st and 2nd generation prototype by Philips; 303 arterial and 129 venous phase) were retrospectively recruited: 106 with typical, 107
atypical, as well as 60 with COVID-19 pneumonia, and 161 lung-healthy controls. Well-evaluated scientific software (YACTA) was used
for segmenting and measuring airway wall attenuation. λHU was calculated as the median maximum airway wall attenuation at
40keV-100keV display energy and aggregated for subsegmental airway generations 5-10 (λHUG5-10).
Results or Findings: λHUG5-10 was significantly higher in the arterial phase in patients with COVID-19 pneumonia (3.2±2.3HU/keV)
vs patients without pneumonia (2.1±1.2HU/keV, p<0.01), typical (2.2±1.8HU/keV, p<0.05) and atypical pneumonia (2.2±1.4HU/keV,
p<0.01), but not in the venous phase. In a multivariate analysis, contrast phase (p<0.01) and pneumonia subtype were significantly
correlated with λHUG5-10 (p<0.05), whereas intubation status, presence of pulmonary emboli and blood CRP were not.
Conclusion: The spectral attenuation curve’s slope may differentiate airway contrast enhancement between different types of
pneumonia in arterial and venous phases. It is a feasible measurement potentially representing active airway wall inflammation in
pneumonia. Further studies should assess its applicability in other airway diseases such as COPD or bronchiectasis.
Funding for this study: Funding was provided by grants from the German Federal Ministry of Education and Research
(82DZL00401, 82DZL004A1). Airway analysis technology is licensed to Imbio, L.L.C.. The funders and industries had no role in study
design, data collection and analysis, decision to publish or manuscript preparation. The spectral CT prototype used for this study was
provided by the manufacturer prior to commercial availability.
Ethics committee - additional information: This retrospective study was approved by the institutional ethics committee
(S-781/2018, S-924/2019, S-293/2020).
Diagnostic performance of vascular reconstruction kernel in photon-counting CT angiography for lower extremity
peripheral artery disease: a comparative study with invasive angiography (8 min)
Paul Simeon Friemel; Freiburg / Germany
Author Block: P. S. Friemel, N. Verloh, J. Weiß, F. Bamberg, M. Doppler, W. Uller, S. Faby, T. Stein, J. Neubauer; Freiburg/DE
Purpose: CT angiography (CTA) is vital for evaluating peripheral artery disease (PAD), but assessing lower leg vessels remains
challenging due to small diameters, and impaired image quality due to calcium blooming. We aim to determine if photon-counting
detector CT (PCD-CT) technology can improve the diagnosis by applying invasive digital subtraction angiography (DSA) as the
reference standard.
Methods or Background: In this IRB-approved study, consecutive patients with suspected lower leg PAD underwent both PCD-CT
and DSA within 48 hours. PCD-CT data were reconstructed into five series using specific vascular kernels (Bv40, Bv44, Bv48, Bv56,
and Bv60). DSA was performed in two orthogonal orientations. Two interventional radiologists independently assessed all PCD-CT and
DSA data, randomly ordered and blinded to the reconstruction type. They rated overall image quality on a 5-point Likert scale
(5=excellent) and assessed the presence and diagnostic confidence (again, 5-point Likert scale; 5=excellent) of potentially
haemodynamically relevant stenosis (≥50%).
Results or Findings: In the final analysis of 24 patients (age 70±11, 39% female), six ≥50% potentially haemodynamically relevant
stenoses were detected using DSA. The Bv56 and Bv60 kernels provided the best overall image quality (4 [4-4]; 4 [3-5]; p≤0.001),
followed by softer kernels. The Bv56 kernel yielded the highest sensitivity (83.33%) and specificity (94.12%) for detecting potentially
relevant stenosis with the highest diagnostic confidence (4 [3-5]; p≤0.001) and inter-reader agreement (k=0.7), followed by Bv60 and
softer kernels.
Conclusion: PCD-CT CTA with a sharp vascular kernel (Bv56) effectively detects lower leg vasculature stenosis, offering high
diagnostic accuracy and confidence. These results may enhance CTA's role in evaluating patients with (suspected) PAD, potentially
reducing the need for invasive DSA.
Limitations: The limitations for CTA are due to small vessel calibre and potential blooming artifacts.
Ethics committee - additional information: The study was approved by the ethics committee application number: 21-1469.
Endovascular thrombectomy in patients aged 80 years and older with acute ischaemic stroke (AIS): technical and
clinical outcomes (8 min)
Anastasija Solodjankina; Riga / Latvia
841
Author Block: A. Solodjankina, K. Kupcs, J. Vetra; Riga/LV
Purpose: This study aims to investigate the technical and clinical outcomes of endovascular thrombectomy using the thrombolysis in
cerebral infarction (TICI) scale, Modified Rankin Scale (mRS) and NIH stroke scale scores in patients over 80 years with acute
ischaemic stroke compared to patients in the age group under 80 years.
Methods or Background: A total of 318 patients with AIS who underwent endovascular thrombectomy in Pauls Stradiņš Clinical
University Hospital during the period from 2020 to 2021 were included in this retrospective cohort study. All patients were divided
into two groups based on their age: those aged under 80 (235) and those aged 80 and above (83). In this study, the statistical
analysis of clinical and technical outcomes of both groups measured by three scales (HIHSS, mRS, TICI) was performed using IBM
SPSS.
Results or Findings: There was a significant difference between the two groups in terms of pre- and postoperative NIHSS values
(p=0.045) and no significant difference between the two groups in terms of pre- and postoperative mRS scores (p=0.113). The TICI
score was not significantly different between the two groups (p=0.241). The mortality rate after endovascular thrombectomy is higher
in elderly patients (>80 years), 19.28%, compared to younger patients (<80 years), 11.91%.
Conclusion: The findings of this study show that the impairment caused by an AIS, which is quantified with NIHSS, is more severe
after endovascular thrombectomy in elderly patients (>80 years) than in younger patients (<80 years). The technical outcomes
characterised by the TICI score are not significantly different between the two age groups. Disability in patients who have suffered an
AIS measured by mRS is not significantly different between the two age groups.
Limitations: The main limitation of this study is the multimorbidity of patients, which could cause difficulties in the interpretation of
the results.
Ethics committee - additional information: The Riga Stradiņš University's ethics committee has approved this study.
Clinical value of using heart rate variability biofeedback before elective CT coronary angiography to reduce the heart
rate and the need for beta‑blockers (8 min)
Carmen Wolf; Zürich / Switzerland
Author Block: C. Wolf; Kiel/DE

Purpose: Coronary computed tomography angiography (CCTA) is a commonly used method to assess coronary artery disease.
Therefore, patients’ heart rate (HR) must be low and regular to minimise artifacts and radiation dose. In our prospective study, we
investigated the value of heart rate variability (HRV) biofeedback before elective CCTA to reduce patients’ HR and to avoid the
application of beta-blockers.
Methods or Background: Sixty patients who received CCTA were included in our study and separated into two groups: with
biofeedback (W-BF) and without biofeedback (WO-BF). The W-BF group used an HRV biofeedback device for 15 min before CCTA. HR
was determined in each patient at four measurement time points (MTP): during the pre-examination interview (MTP1), positioning on
the CT patient table before CCTA (MTP2), during CCTA image acquisition (MTP3), and after completing CCTA (MTP4). If necessary,
beta-blockers were administered in both groups after MTP2 until an HR of less than 65 bpm was achieved. Two board-certified
radiologists assessed the image quality and analysed the findings.
Results or Findings: Overall, the need for beta-blockers was significantly lower in patients in the W-BF group than in the WO-BF
group (p=0.032). The amount of HR reduction between MTP1 and MTP2 was significantly higher in the W-BF group than in the WO-BF
group (p=0.028). There was no significant difference between the W-BF and WO-BF groups regarding image quality (p=0.179).
Conclusion: Application of HRV biofeedback prior to elective CCTA could decrease HR and beta-blocker use without compromising CT
image quality and analysis.
Limitations: Biofeedback was only performed until the patient was positioned on the CT table. We could possibly not achieve the
maximal effect of HRV biofeedback during the CCTA examination itself. Integrating visual feedback into the CT scanner might further
enhance and stabilise HRV biofeedback effects.
Ethics committee - additional information: The study was approved by the local institutional ethics board of Kiel University (No.
D 549/19).
Optimisation of mesenteric vascular CT mapping (8 min)

Julia Sophie Heeb; Sevelen / Switzerland
842
Author Block: J. S. Heeb, H. Thoeny, L. Widmer; Fribourg/CH
Purpose: The study aims to compare contrast opacification in CT angiography of mesenteric vessels between three cohorts using a
fixed trigger delay, a patient-individualised split-bolus injection, and a slow contrast media injection.
Methods or Background: Complete mesorectal excision in colorectal cancer is technically demanding due to frequent variations in
the vascularisation of the colon. Mapping of the mesenteric anatomy could decrease surgery length and vascular lesions. We aim to
optimise this preoperative mesenteric vascular mapping. In this IRB-approved single-centre prospective study, 162 consecutive
patients were prospectively recruited into three cohorts with distinct CT vascular mapping protocols. The inclusion of 54 participants
per cohort was determined from a-priori statistical power analyses. One reader assessed objective image quality; two readers
assessed subjective image quality. The proportion of unidentified vessels was analysed using beta regression (chi-squared test). A
one-way between-subject ANOVA was computed to test for differences in mesenteric vessel attenuation values.
Results or Findings: Of the three protocols, the split-bolus protocol had the most extreme attenuation values. The fixed trigger
delay and the slow injection protocols had more similar mean values, with the slow injection protocol showing higher attenuation
values for both arteries and veins. This was expressed by significantly lower artery attenuation values in the fixed trigger delay
protocol than in the other two protocols (F[2,138] = 35.7, p < 0.001) and significantly lower vein attenuation values in the split-bolus
protocol than in the other two protocols (F[2,138] = 41.9, p < 0.001). The proportion of unidentified vessels was not significantly
different between protocols.
Conclusion: Slow and continuous contrast media injection improves opacification in CT angiography of the mesenteric vessels,
providing enhanced vascular mapping for surgery.
Limitations: Protocols were tested on different cohorts.
Funding for this study: The study has received institutional funding.
Ethics committee - additional information: The study was approved by CER-VD, swissethics.
Predictive value of PET- and CT-based imaging criteria at first follow-up for GEP-NET under PRRT (8 min)
Christian Alexander Alexander Dascalescu; München / Germany
Author Block: C. A. A. Dascalescu, M. Heimer, M. P. Fabritius, L. Unterrainer, J. Rübenthaler, M. Ingrisch, J. Ricke, C. C. Cyran;
Munich/DE
Purpose: The study aims to evaluate and compare the predictive value of different PET- and CT-based imaging criteria on peptide
receptor radionuclide therapy (PRRT) and therapy guidance in gastroenteropancreatic neuroendocrine tumours (GEP-NET) at interim
follow-up.
Methods or Background: In this single-centre retrospective study, we reviewed 178 patients with GEP-NET who were treated with
at least two consecutive cycles of PRRT and underwent somatostatin receptor (SSR-) PET/CT at baseline and after two cycles of PRRT.
We evaluated the agreement between RECIST 1.1 and mCHOI criteria to a multidisciplinary GEP-NET tumour board and assessed the
modified Krenning and SSTR-RADS scores at baseline and interim follow-up.
Results or Findings: The overall concordance between the response criteria was fair (weighted Cohen's Kappa=0.26). The
concordance at the follow-up between RECIST and mCHOI with the tumour-board assessment was comparable with Cohen's
Kappa of 0.42 and 0.39, respectively. The survival analysis showed no significant differences in PFS for RECIST 1.1 or mCHOI
categories SD and PR (log-rank p=0.70 and 0.68, respectively). mChoi-criteria showed a higher sensitivity (sensitivity: 90%,
specificity: 96%, PPV: 44%, NPV: 99%) but a lower positive predictive value in predicting PRRT discontinuation due to a PD in tumour-
board assessment as compared to RECIST 1.1 (sensitivity: 50%, specificity: 100%, PPV: 63%, NPV: 98%). A baseline Krenning score of
3 was associated with a shorter PFS compared to a Krenning score of 4 (721 days vs 1169 days; p=0.01). Assessment and stage
migration of SSTR-RADS scores between baseline and follow-up yielded no significant value for response assessment.
Conclusion: RECIST 1.1 and mCHOI have comparable diagnostic values to predict PFS. At baseline, the Krenning score could
contribute to estimating long-term therapy response. Further research towards integrated bimodal monitoring and assessment tools
in SSR PET/CT is warranted to guide therapy management at the interim follow-up.
Limitations: The primary limitation was limited response assessment time points, including the first interim follow-up only.
Funding for this study: Funding was received from the Wilhelm Vaillant Stiftung.
Ethics committee - additional information: This study was approved by the ethics committee of LMU-Munich, Project number:
20-1077.
Effects of contrast medium injection pressure on angiographic image quality using different microcatheter systems: a
prospective pilot study (8 min)
Sara Abosabie; New Haven, CT / United States
843
Author Block: S. Abosabie, T. Kao, D. Schnapauff, T. A. Auer, M. Jonczyk, W. Lüdemann, A. Frisch, B. Gebauer, L. J. Savic; Berlin/DE
Purpose: This pilot study tested the hypothesis that a higher contrast medium (CM) injection pressure can improve the image quality
of digital subtraction angiography (DSA) and reduce DSA time and radiation exposure.
Methods or Background: This prospective single-centre study included 12 patients with hepatocellular carcinoma (n=11) or liver
metastases (n=1) with intra-arterial therapies (10/2022-03/2023) to systematically compare DSA image quality (primary) and
radiation exposure (secondary endpoint) using two microcatheters with different maximum application pressures (Pmax). IRB
approval and informed consent were obtained. Patients underwent two DSAs (20ml CM, 4ml/s) using Progreat (Terumo, Pmax=750
PSI) and DraKon (Guerbet, Pmax=1200 PSI) microcatheters (both 2.4Frx130cm) placed in the common hepatic artery. Application
pressure, CM flow, volume, and dose area product (DAP) were recorded. The image quality was evaluated by three blinded IR using a
customised questionnaire with 16 questions in 5 categories. Responses were converted to numerical values and compared by paired
t-test.
Results or Findings: The mean maximum application pressure was higher using Drakon (917±94 PSI) than Progreat (731±45 PSI).
In 10 patients with Progreat, the injection was terminated early. Mean DAP with DraKon (376 μGym2) was slightly lower than with
Progreat (388 μGym2). Vessel visualisation was equivocal or superior with DraKon in 72% of the patients. Additionally, in 100% of
patients, tumour blush was demarcated equally or more clearly in DSAs with DraKon (p< 0.001).
Conclusion: Our results suggest potential benefits of standardised CM injections for DSA using higher application pressure to
enhance image contrast and tumour demarcation during IAT. These findings may enable faster and more precise embolisation if
confirmed in a larger cohort.
Limitations: In the future, we intend to conduct a larger study with a much larger patient cohort because our cohort currently only
consists of 12 patients.
Funding for this study: This study was funded by a research grant from Guerbet.
Ethics committee - additional information: This study was approved by the Ethics Committee of Charité Universitätsmedizin
Berlin.
Prognostic machine learning classifiers using radiomics of supratentorial intracerebral haemorrhage and surrounding
oedema on admission non-contrast head CT (8 min)
Fiona Dierksen; Göttingen / Germany
Author Block: F. Dierksen1, S. Haider1, A. T. Tran1, H. Lin1, J. Sommer1, I. Maier2, S. Aneja1, A. I. Qureshai3, S. Payabvash1; 1New Haven,
CT/US, 2Göttingen/DE, 3Columbia, MO/US
Purpose: Following intracerebral haemorrhage (ICH), the haemorrhagic lesion represents the extent of primary brain injury, whereas
surrounding oedema represents secondary brain injury related to blood product degradation. While prior studies linked admission ICH
volume and, more recently, ICH radiomic features with clinical outcomes, there have barely been studies on prognostic correlates of
combined ICH and peri-haematomal oedema (PHE) radiomic features. We developed an improved algorithm for patient outcome
prediction by incorporating oedema radiomics into the model.
Methods or Background: Using the ATACH-2 trial dataset, we extracted 1130 features from manually segmented ICH and PHE
lesions on admission non-contrast head CTs. We split the data from 892 patients into discovery (n=500) and independent validation
(n=352) cohorts. In the discovery cohort, we trained, optimised, and compared the performance of 36 combinations of six machine-
learning classifiers and six feature selection methodologies for the prediction of favourable vs poor outcomes based on a 3-month
modified Rankin score. Separate models were trained and tested for radiomics from ICH vs ICH and PHE features. Finally, we
evaluated the best-performing models using ICH vs ICH and PHE features.
Results or Findings: The best-performing model using ICH features alone was with Naïve-Bayes and RIDGE regression, achieving an
AUC of 0.71 (confidence interval: 0.65-0.77), and the best-performing model using ICH and oedema features was Elastic-Net and
RIDGE achieving an AUC of 0.74 (CI: 0.7-0.79). Although there was no significant difference in AUCs in the validation cohort
(p=0.074), in risk assessment, there was a 17% improvement in the Net Reclassification Index (p<0.001) and a 12% improvement in
the Integrated Discrimination Index (p<0.001).
Conclusion: Incorporating PHE in addition to ICH features from admission head CT can improve the outcome risk assessment of
prognostic machine-learning models in ICH patients.
Limitations: The study was limited by missing 72-hour follow-up images.
Funding for this study: Dr. Payabvash is supported by the National Institutes of Health (K23NS118056) and the Doris Duke
Charitable Foundation (2020097).
Ethics committee - additional information: The ethics approval was ensured by the ATACH-2 investigators (ClinicalTrials.gov
identifier: NCT01176565). Our group performed post-hoc analyses of anonymised data.
844
ST 18 - Funding the Future: A Dialogue with EIBIR's Scientific Director on Navigating EU Research
Opportunities
Date: March 2, 2024 | 14:00 - 14:30 CET
The European Institute for Biomedical Imaging Research (EIBIR) is at the forefront of coordinating research projects to support the
development of biomedical imaging technologies and the dissemination of knowledge. EIBIR actively supports research networking
activities and common initiatives in the field of biomedical imaging research. With the EU-funded projects EIBIR leads and supports,
the institute is working towards the goal of building stronger, more resilient health systems. In this interview, the EIBIR Scientific
Director Prof. Regina Beets-Tan will share their insights on how the organisation navigates the European funding programmes, aligns
its research efforts with its objectives and prioritises the projects it supports. Join us as we delve into the inner workings of EIBIR and
discover how it is contributing to the European Health Union.
Moderator:
Interview (30 min)

845
RPS 2007 - Innovations in prostate imaging
Categories: Artificial Intelligence & Machine Learning, Genitourinary, Oncologic Imaging

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Maarten De Rooij; Nijmegen / Netherlands
Performance of an ultra-fast deep-learning accelerated MRI screening protocol for prostate cancer compared to a
standard multiparametric protocol (7 min)
Benedict Oerther; Freiburg / Germany
Author Block: B. Oerther1, H. Engel1, R. Strecker2, T. Benkert2, E. Weiland2, F. Bamberg1, M. Benndorf1, J. Weiß1, C. Wilpert1;
1
Freiburg/DE, 2Erlangen/DE
Purpose: The aim of this study was to establish and evaluate image quality of an ultra-fast MRI screening protocol for prostate
cancer in biopsy-naïve men regarding PI-RADSv2.1 classification in comparison to the standard multiparametric protocol.
Methods or Background: This prospective mono-institutional study included consecutive patients with suspected prostate cancer
without prior biopsy. A PI-RADSv2.1 conform multiparametric MRI protocol was acquired in a 3 T MRI scanner (scan time: 25min
45sec). Additionally, two deep-learning accelerated sequences were acquired (scan time: 3 min 28 sec). Two readers evaluated image
quality and the presence of prostate cancer. In a first reading session only the screening protocol (DL accelerated axial T2w and
ZOOMit DWI) was available. Subsequently, the full conventional mpMRI protocol was assessed and served as a reference standard.
Diagnostic performance was analyzed with mpMRI serving as the gold standard. Inter- and intra-reader agreement was assessed
using weighted kappa statistics.
Results or Findings: The final cohort consisted of 77 patients with 97 lesions. Diagnostic performance of the screening protocol was
excellent with a sensitivity and specificity of 100%/97% and 98%/83% (cut-off ≥ PI-RADS 3) vs. 100%/100% and 89%/98% (cut-off ≥
PI-RADS 4) for reader 1 and reader 2, respectively. Mean image quality (Likert-scaling) was 3.96 (R1) and 4.35 (R2) for the standard
protocol vs. 4.74 and 4.57 for screening protocol (p < 0.05). Inter-reader agreement was moderate (κ: o.55) for the screening protocol
and substantial (κ: o.61) for the multiparametric protocol. Intra- reader agreement was excellent (κ: o.98) for R1 and substantial (κ:
o.79) for R2.
Conclusion: A DL accelerated screening protocol for prostate cancer in biopsy-naïve men proved similar diagnostic performance and
better imaging quality compared to the conventional mpMRI protocol, requiring less than 15% of scan time.
Limitations: Monocentric study, limited number of patients; no histopathological ground truth
Ethics committee - additional information: This study was ethically approved the Local ethics committee of Freiburg.
Developing a dynamic predictive model for baseline detection and follow-up re-evaluation of the risk of prostate
cancer progression on active surveillance (PROGRESS Prostate) (7 min)
Nikita Sushentsev; Cambridge / United Kingdom
846
Author Block: N. Sushentsev1, L. A. Abrego Rangel2, N. Sanmugalingam3, I. Caglič1, V. Gnanapragasam1, A. Warren1, A. Zaikin2, T.
Barrett1, O. Blyuss2; 1Cambridge/UK, 2London/UK, 3Nowich/UK
Purpose: The aim of this study was to develop a dynamic predictive model for baseline detection and follow-up re-evaluation of the
risk of prostate cancer (PCa) progression on active surveillance (AS).
Methods or Background: Four hundred and twenty two AS patients were included in this study, of whom 82 (19.4%) experienced
either histological PCa progression or radiological stage progression (PRECISE 5) over a median follow-up of 4.5 years. The baseline
model included initial serum prostate-specific antigen (PSA) and PSA density (PSAD), MRI-derived Likert score, tumour diameter, and
tumour grade group. The follow-up model included baseline Likert score along with longitudinal PRECISE scores, PSAD measurements,
and repeat biopsy results. Model training and testing were performed in the 50/50 data split using several neural networks, with
three-year progression as the outcome.
Results or Findings: The best-performing baseline model was a generalised additive model (GAM) including baseline PSAD and
Likert score. With an overall test AUC of 0.65, the model achieved a 21% specificity at 95% sensitivity in the test set, which may be
used to avoid repeat biopsies in a substantial proportion of patients with minimal risk of missing disease progression. The follow-up
model, comprised of a long short-term memory recurrent neural network, included baseline Likert score together with longitudinal
PRECISE and PSAD measurements, with its test AUC of 0.75 being significantly higher compared to that of PRECISE alone (AUC=0.61,
P<0.01).
Conclusion: The development of MRI-driven risk-adapted AS predictive models is a high research priority in the field; this study
shows the promise of the proposed approach to objectively stratify patients at baseline and significantly improve the performance of
current standard-of-care PRECISE assessment for detecting disease progression in the follow-up.
Limitations: Lack of external validation to be addressed in future studies.
Funding for this study: This study was funded by the Cancer Research UK ACED (A095792/EICEDAAP\100009).
Ethics committee - additional information: This study was ethically approved by the HRA and Health and Care Research Wales
(HCRW), IRAS Project ID: 288185.
DWI²-improvement of diffusion-weighted imaging for the detection of clinical significant prostate cancer (7 min)
Birte Valentin; Düsseldorf / Germany
Author Block: B. Valentin, M. Boschheidgen, T. Ullrich, J. P. Radtke, P. Albers, G. Antoch, H-J. Wittsack, L. Schimmöller; Düsseldorf/DE
Purpose: This study aimed to improve the contrast between prostate cancer (PC) and healthy tissue by diffusion-weighted imaging
(DWI) post-processing by using a square method.
Methods or Background: DWI post-processing was retrospectively applied on 40 patients with PC (median 68y and PSA 8.0 ng/ml)
and multiparametric MRI (mpMRI) performed at 3 Tesla (Magnetom PRISMA® Siemens, Erlangen, Germany). In 20 patients a multi-
shot readout segmentation (rs-EPI) plus zoomed single-shot imaging (z-EPI) sequence (Group 1) and in 20 patients a single-shot echo-
planar imaging (ss-EPI) plus rs-EPI sequence (Group 2) was applied. All sequences (b1000 and b1800/2000) were squared and
afterwards evaluated objectively using the SyngoVia Software (Siemens Healthineers, Erlangen) and subjectively by applying a 5-
point scale from unacceptable, poor, moderate, good, to excellent.
Results or Findings: The squared processed DWI sequences showed significantly higher contrast-ratio (CR) for ss-EPI b1800
(p<0.001), rs-EPI b1000 (p<0.001), rs-EPI b1800 (p<0.001), z-EPI b1000 (p=0.002) and for z-EPI b2000 (p<0.001). Following post-
processing of these sequences, a significant improvement in the subjective assessment of image quality was noticeable for ss-EPI
b1000 (p=0.043), ss-EPI b1800 (p=0.030), rs-EPI b1000 (p<0.0001), rs-EPI b1800 (p<0.001) and z-EPI b1000 (p<0.001).
Conclusion: The application of the square post-processing for DWI results in a significant improvement in the CR between PC and
healthy tissue, especially for high b values of ss-EPI or re-EPI. This method can be instrumental in enhancing the detection and
differentiation of PC lesions.
Limitations: High b-values can cause overexposure of the lesion and highlighting of non-specific DWI restrictions. This could result in
overdiagnosis as well as misinterpretation of the tumour lesion margins.
Funding for this study: This study did not receive any funding.
Ethics committee - additional information: The local ethics committee has positively reviewed this retrospective study.
Spectral diffusion analysis to improve the detection and classification of clinical significant prostate cancer (7 min)
Birte Valentin; Düsseldorf / Germany
847
Author Block: B. Valentin, T. Thiel, T. Ullrich, M. Boschheidgen, J. P. Radtke, P. Albers, G. Antoch, H-J. Wittsack, L. Schimmöller;
Düsseldorf/DE
Purpose: In this study, we applied the model-free nonnegative least squares (NNLS) method. The aim of this study was to assess the
number of distinguishable diffusion components (spectral diffusion analysis) within the prostate and to differentiate between healthy
and pathological prostate tissue.
Methods or Background: NNLS imaging was performed at 3 Tesla (Magnetom PRISMA® Siemens, Erlangen, Germany) in 10
patients with prostate cancer (PC) (PI-RADS 5 and subsequent biopsy). The 16 b values used were 0, 50, 100, 150, 200, 300, 400,
500, 600, 700, 800, 1000, 1200, 1400, 1600, 1800 s/mm 2. Relative signal fractions and mean diffusivities of the diffusion
components in the peripheral zone, central zone and PI-RADS 5 lesion were obtained using the regularized NNLS fitting of the
intravoxel incoherent motion data.
Results or Findings: Three different diffusion components (10–4, 10–3, and 10–2 mm2/s) were detected in prostate tissue. In
comparison, the three peaks were significantly different between healthy and diseased tissue. The fraction of the slow component
was significant higher in PC (maximum amplitude of 0.2) compared with the unaffected prostate tissue (maximum amplitude of 0.05).
Conclusion: This pilot study demonstrated the feasibility of spectral diffusion weighted imaging for the differentiation of PC. The
three distinguishable components may be related to slow tissue diffusion caused by higher tissue density of PC lesions, intermediate
fluid flow caused by glandular tissue, and fast blood flow in prostatic vessels. A larger cohort study with a ISUP range is needed to
further evaluate this technique.
Limitations: This study exclusively focused on PI-RADS 5 lesions. Consequently, we were unable to differentiate between lesion sizes
and did not account for variations in high-, intermediate-, and low-risk prostate cancer cases.
Funding for this study: This study was funded by DFG.
Ethics committee - additional information: The local ethics committee has positively reviewed this prospective study.
Assessing the performance of generative pre-trained transformers against radiologists for PI-RADS classification
based on prostate mpMRI text reports (7 min)
Kang Lung Lee; Cambridge / United Kingdom
Author Block: K. L. Lee, D. Kessler, T. Barrett; Cambridge/UK

Purpose: Large language models, such as ChatGPT and Bard, have sparked a wave of enthusiasm for their potential applications in
clinical radiology, including formulating clinical interpretation of reports. This study aims to compare the classification abilities of
ChatGPT, Bard, and two uroradiologists in assigning PI-RADS categories based on clinical text reports.
Methods or Background: Clinical prostate MRI text reports from 100 consecutive treatment-naïve patients undergoing mpMRI
between 06.11.2022 to 28.12.2022 were analysed. Clinical history and concluding remarks were removed from the text reports. Two
uroradiologists with 14 and 3 years of prostate MRI reporting experience, retrospectively independently classified PI-RADS 2.1
categories on the edited text reports. The same reports were inputted manually into online ChatGPT-3.5 and Bard platforms to
generate PI-RADS classifications (without prior training). Original report classifications were considered definitive, and comparisons
were made to compare the original reports, the two radiologists, ChatGPT, and Bard. Agreement rates and Κappa scores were
analysed.
Results or Findings: In the original reports, 52/100 MRIs were classified as PI-RADS 2, 9/100 as PI-RADS 3, 19/100 as PI-RADS 4, and
20/100 as PI-RADS 5, respectively. Compared to the original classifications, the senior and junior radiologists concurred on 95% and
90% of the reports, respectively, while ChatGPT and Bard aligned both on 67 reports. Notably, Bard assigned a non-existent PI-RADS 6
classification to two patients (2%). The interreader agreement (Κ) between the original reports and the senior radiologist, the junior
radiologist, ChatGPT, and BARD were 0.92, 0.85, 0.55, and 0.49, respectively.
Conclusion: Concordance on PI-RADS scoring was high among radiologists, however, ChatGPT and Bard demonstrate poor
performance for the text-based classification task.
Limitations: The limitation of the study is that this a relatively small sample of 100 reports.
Ethics committee - additional information: The study was approved by Cambridge University NHS Foundation Trust (reference
number: 288185).
Sodium MRI quantification of prostate tissue and therapy-naïve primary prostate cancer with internal references (7
min)
Fabian Tollens; Mannheim / Germany
848
Author Block: F. Tollens, F. G. Zöllner, A. Adlung, S. O. Schönberg, D. Nörenberg; Mannheim/DE
Purpose: The aim of this study was to quantify tissue sodium concentration (TSC) in healthy prostate tissue and prostate cancer
regions based on internal references in order to evaluate TSC as a potential quantitative imaging biomarker.
Methods or Background: Thirty-six patients with clinically suspected prostate cancer were included into this prospective study and
underwent clinical multiparametric magnetic resonance imaging (MRI) and additional sodium MRI of the prostate. Imaging was
performed at 3T using a dual-tuned 1H/23Na body-coil to acquire a 3D radial density-adapted 23Na sequence. For the quantification
of tissue sodium concentration, femoral blood vessels were chosen as an internal reference and reported sodium levels of ~81 mM
were assumed. Peripheral zone (PZ), transition zone (TZ) and tumor regions of interest were defined and TSC was extracted for each
segmentation.
Results or Findings: Mean TSC differences between right and left femoral blood vessels was 3.3 ± 2.2 mM. TSC was significantly
higher in the PZ (40.7 ± 6.0 mM) than in the TZ (37.5 ± 5.7 mM). Nine suspicious lesions (PI-RADS 4 and 5) were detected in eight
men that were confirmed with Gleason scores of ≥ 3+3 by biopsy. TSC in prostate cancer (32.2 ± 5.5 mM) was significantly lower
than in contralateral healthy regions (36.1 ± 3.9 mM, p=0.018).
Conclusion: Femoral blood vessels as an internal reference for TSC quantification are less prone to inaccuracies caused by B1
inhomogeneities as opposed to external sodium probes, which enables a robust quantification. TSC was significantly decreased within
prostate cancer compared to healthy prostate tissue. Reduced TSC could represent a quantitative imaging biomarker that could
improve prostate cancer risk stratification.
Limitations: Patients’ blood sodium concentration assumed based on literature; only PI-RADS 4 and 5 lesions with histopathologic
confirmation considered as confirmed prostate cancer lesions.
Funding for this study: This research project is part of the Research Campus M2OLIE and funded by the German Federal Ministry of
Education and Research (BMBF) within the Framework “Forschungscampus: public-private partnership for Innovations” under the
funding code 13GW0388A and 13GW0092D.
Ethics committee - additional information: The study was approved by the local ethics committee.
Qualitative comparison of conventional and deep-learning reconstructed diffusion-weighted magnetic resonance

imaging sequence for prostate cancer imaging (7 min)
Yolène Lefebvre; Anderlecht / Belgium
Author Block: Y. Lefebvre1, E. Venetis1, T. Metens1, T. Benkert2, E. Weiland2, M. A. Bali1, N. Coquelet1; 1Brussels/BE, 2Erlangen/DE
Purpose: The goal of this study was to compares a conventional diffusion-weighted imaging (DWI-STD) to a faster DWI using deep-
learning reconstruction strategies (DWI-DLR) for prostate cancer (PCa) imaging in terms of image quality (IQ) and diagnostic
interpretation.
Methods or Background: From February 2023 to July 2023, we query a retrospective monocentric database of patients with
prostatic lesions who underwent a 3T magnetic resonance exam. Each patient successively underwent the DWI-DLR (acquisition time:
3 min38 s; research application) and the DWI-STD (4 min58s). Image analysis was performed by an expert radiologist in prostate
imaging. Using high acquired b-value DWI images (800 s/mm2) and apparent diffusion coefficient (ADC), each sequence was rated
using a 5-point Likert scale for overall IQ, noise, sharpness, contrast, artifacts and distortion. Lesion characteristics were assessed
based on calculated b-value DWI (1400 s/mm2) and ADC maps, and rated for diagnostic confidence and detectability. Rating
comparisons between the DWI sequences were performed using receiving operating characteristic curves and associated areas under
the curve (AUC). P-values testing the null hypothesis that AUC equal 0.5 were computed and p-values below 0.05 Bonferroni-corrected
for multiple comparisons were deemed significant.
Results or Findings: Twenty patients were included (mean age: 66.9 years, range: 58–80 years). We found that overall IQ (b800
and ADC); noise, sharpness and distortion (b800) were better for DWI-STD. Noise, sharpness and distortion (ADC); contrast and
artifacts (b800 and ADC) were better for DWI-DLR. For lesions, diagnostic confidence and lesion detectability were better for DWI-DLR
compared to DWI-STD. Significance was reached for contrast (b800 and ADC), noise and sharpness (ADC), and lesion detectability
(b1400).
Conclusion: Our preliminary results show that PCa DWI can be acquired more rapidly using deep-learning reconstruction strategies
without loss of IQ and diagnostic interpretation.
MRI radiomic model to predict prostate cancer of the anterior zone (7 min)
Silvia Schirò; Parma / Italy
849
Author Block: S. Schirò, L. Leo, M. Russomando, P. Canale, C. Zilioli, V. Casale, C. Roberti, C. Manna, N. Sverzellati; Parma/IT
Purpose: The purpose of this study was to assess the diagnostic performance of MRI radiomic model in the prediction of anterior
zone prostate cancer.
Methods or Background: This study included a retrospective monocenter dataset of subjects with prostate 1.5 T mpMRI showing an
anterior zone lesion (PIRADS 3-5) and fusion target biopsy within six months. The histopathology results were the standard of
reference. Two radiologists ( four years of experience) independently, reviewed and manually segmented the lesion with an open-
source software at T2-weighted and ADC maps. Overall, 851 radiomics features (FR) were extracted for both T2-weighted and ADC
maps. 100 train:test (0.7:0.3) splits were created and recursive feature elimination with a 5-fold cross-validation was performed on
train partitions using the Random Forest Classifier (RFC). Subsequently, RFC was trained by selecting iteratively an increasing number
of features sorted by their occurrences to evaluate the minimum number of informative features. Finally, means and 95% confidence
intervals of accuracy, sensitivity, specificity, precision, area under the receiver operating characteristic curve (ROC-AUC) were
calculated on the test partitions.
Results or Findings: Overall, 89 males (mean age 68 years; ±SD 8) were included. The anterior zone lesions were scored as follows:
PI-RADS 3 (n=28, 31%), PI-RADS 4 (n=40, 45%) and PI-RADS 5 (n=21, 24%). Of these, 47/89 (53%) showed anterior zone prostate
cancer (Gleason Score ≥ 3+3). The best model on the test set exploited six first-order features of ADC maps and three first-order
features of T2-weighted images reaching accuracy, sensitivity, specificity, precision and ROC-AUC of 0.74 [95% C.I. 0.73-0.75], 0.72
[0.70-0.75], 0.77 [0.74-0.80], 0.80 [0.78-0.82], 0.82 [0.80-0.83], respectively.
Conclusion: The proposed radiomic model reached satisfactory performance in predicting anterior zone prostate cancer and may be
a useful supportive tool in the diagnostic pathway.
Limitations: Monocentric retrospective cohort and no external validation
Ethics committee - additional information: This study was approved by CE AVEN : AVEN873/2018/OSS/UNIPR.
The role of in-bore prostate biopsy in the diagnosis of prostate cancer (7 min)
Caterina Pizzi; Milan / Italy
Author Block: C. Pizzi, C. Sattin, S. Alessi, Q. Nguyen, G. Musi, G. Petralia; Milan/IT

Purpose: The goal of this study was to evaluate the detection rates of prostate cancer (PCa) and clinically significant prostate cancer
(csPCa) with in-bore MRI-guided prostate biopsy.
Methods or Background: We considered in-bore MRI-guided prostate biopsies performed at our institution from November 2014 to
January 2023, using two different systems: one manual and one robotic. Differences in detection rates of PCa and csPCa were
assessed between the two systems and between patients with lesions ≤10 mm and >10 mm.
Results or Findings: Of the 831 biopsies included, 493 were positive for PCa and 281 for csPCa (59.3% and 33.8%). The detection
rate of PCa and csPCa did not differ between manual and robotic systems (55.3% versus 63.2% and 33.2% versus 34.4%,
respectively), neither for lesions ≤10 mm and >10 mm (32.9% versus 25.5% and 18.1% versus 16%, respectively).
Conclusion: The observed detection rates of PCa and csPCa (59.3% and 33.8%) agree with the literature. In the absence of
significant differences in detection rates of PCa and csPCa, MRI-guided bore biopsy can be performed effectively on lesions of any size
using either commercially available system (manual or robotic).
Ethics committee - additional information: This institutional review board approved study was a single-centre retrospective
analysis, and the specific consent was waived for this audit of clinical procedures that had been performed with informed consent.
Ultrasound/ MRI fusion-guided transperineal laser ablation (TPLA) in the treatment of localised radiotherapy-resistant
prostate cancer (7 min)
Beatrice Carreri; Rome / Italy
850
Author Block: B. Carreri, P. E. Gigliotti, F. R. Fraioli, M. Nezzo, F. G. Garaci, G. Manenti; Rome/IT
Purpose: The aim of this prospective interventional study is assessing the feasibility, safety and treatment success of US/ MRI fusion-
guided transperineal laser ablation (TPLA) as salvage treatment for radiation therapy refractory focal PCa, evaluating clinical and
functional outcomes and defining post-procedural imaging findings using 3T multi-parametric MRI.
Methods or Background: A cohort of five patients over 70 years old who had undergone RT as primary treatment, with a single,
Gleason score ≤ 7 (3 + 4), local recurrence for PCa, underwent TPLA in outpatient setting (SoracteLite, ECHOLASER, Elesta). Post-
ablation follow-up included regular PSA sampling and 3T mpMRI at one hour, three, six, 12, and 18 months and systematic and
targeted ultrasound/MR fusion-guided biopsies at 18 months. International Prostate Symptom Score (IPSS) and the five-item version of
the International Index of Erectile Function (IIEF-5) questionnaires were completed at baseline and at 18-month follow-up to
investigate any procedure-related erectile dysfunction or urinary symptoms.
Results or Findings: All procedures were successfully completed with no significant complications (Clavien-Dindo Grade I). The
procedure achieved optimal outcomes, with a statistically relevant reduction of PSA and ablation cavity volume trends at the end of
follow-up (>70%). IIEF-5 and IPSS scores showed no significant difference between pre-procedural and 18 months values.
Ultrasound/MRI fusion-guided biopsies at 18 months also confirmed the absence of recurrence.
Conclusion: Preliminary results demonstrated that TPLA can effectively and safely treat local recurrences of RT refractory PCa over a
medium-term period, without side-effects and functional complications.
Limitations: The limitations of our study are that it’s single-centred and no preliminary sample size calculation was conducted.
Instead, we only included a small cohort of patients who met our specific criteria, such as having received radiation therapy, having a
low-risk score, no extraglandular extension.
Ethics committee - additional information: The study was approved the ethics committee: UID RS 68/19.
Deep-learning-based image reconstruction of T2W images of the prostate and compressed sensing for one half of the
acquisition time (7 min)
Lukas Lambert; Prague / Czechia
Author Block: L. Lambert1, M. Jurka1, M. Wagnerová1, O. Capoun1, R. Jakubicek2, P. Ourednicek2, A. Burgetova1; 1Prague/CZ, 2Brno/CZ
Purpose: Deep-learning-based reconstruction (DLR) is being developed by major MRI vendors in order to improve image quality and
accelerate image acquisition. We used DLR combined with compressed sensing (CS), a method of K-space subsampling, the aim was
to accelerate acquisition of T2-weighted MRI of the prostate gland and maintain original image quality.
Methods or Background: In this prospective study, forty-seven patients underwent biparametric prostate MRI with two T2
acquisitions in the transverse plane- a standard acquisition (4:27 min) and short acquisition (2:18) accelerated by increasing the CS
factor from 1.3 to 2.8. The images were reconstructed with and without DLR. The image quality was rated in six domains, contrast-to-
noise ratio and image sharpness were measured.
Results or Findings: The image quality of short-DLR was rated better in all categories compared to the standard sequence
(p<0.0001 to p=0.0044). DLR images had higher sharpness compared to non-DLR. Both short and short-DLR images had lower
calculated CNR. Subjective evaluation correlated both with prostate volume and image sharpness (p<0.0001).
Conclusion: The combination of DLR and CS results in accelerated acquisition of T2 images of the prostate with maintained
perceived image quality, higher sharpness, and lower contrast-to-noise ratio.
Limitations: Study limitations: single center, single vendor, optimization of one sequence, technical level study
Funding for this study: This study was supported by the Ministry of Health of the Czech Republic (MH CZ-DRO, General University
Hospital in Prague, 00064165) and by the institutional funding of the Charles University in Prague (Cooperatio, Medical Diagnostics
and Basic Medical Sciences).
Ethics committee - additional information: The study was approved by the Ethics Committee of the General University Hospital
in Prague
PSMA PET-based radiomic features for non-invasive discrimination of intraprostatic tumours (7 min)
Liang Luo; Xi’an / China
851
Author Block: L. Luo, R. Chang, Y. Li, Y. Liao, Z. Wang, X. Duan; Xi’an/CN
Purpose: This study aims to investigate the utility of machine learning-based radiomics models derived from PSMA PET/CT images in
differentiating between benign and malignant intraprostatic lesions detected by [18F]-PSMA-1007 PET/CT.
Methods or Background: We retrospectively analyzed consecutive patients with prostate cancer (PCa) who underwent [18F]-
PSMA-1007 PET/CT imaging and biopsy. A total of 1316 radiomic features were extracted from volumes of interest on PET and CT
images respectively. Feature selection was performed using the Max-Relevance and Min-Redundancy (mRMR) algorithm and the least
absolute shrinkage and selection operator (LASSO). Two radiomics models (PET model and PET/CT model) were generated using
logistic regression in the training set. In addition, two baseline models were developed using clinical data including fPSA and tPSA, as
well as prostate cancer molecular imaging evaluation standards (PROMISE), namely Clinical model and PROMISE model. ROC and
Delong test were employed for model evaluation and comparisons.
Results or Findings: A total of 75 patients (50 with PCa and 25 with benign prostate hyperplasia) were included, with 53 patients for
training and 22 for testing. In the training set, the area under curve (AUC) of the PET and PET/CT models were 0.94 (95% CI:
0.88-1.00) and 0.97 (95% CI: 0.92-1.00), respectively. The best-performing model (PET/CT model) demonstrated an accuracy of
92.5%, sensitivity of 100%, and specificity of 77.8%. Although the AUC of the PET/CT model was not significantly better than that of
the PET model, it was significantly outperformed the Clinical model and PROMISE model (P < 0.001 and P = 0.002, respectively.).
Conclusion: Our findings highlight the potential clinical relevance of [18F]-PSMA-1007 PET-based radiomics models in the non-
invasively prediction of intraprostatic lesions in patients with PCa, and show better diagnostic performance compared to baseline
models.
Funding for this study: No funding was obtained for this study,
Ethics committee - additional information: Approval was granted by the Ethics Committee of the First Affiliated Hospital of Xi’an
Jiaotong University.
852
CUBE 23 - Carotid interventions: what to do to avoid disasters!

Date: March 2, 2024 | 14:00 - 14:30 CET
Carotid interventions: what to do to avoid disasters! (28 min)

Grzegorz Marek Karwacki; Luzern / Switzerland
1. To describe the most common complications post carotid interventions.

2. To illustrate the most common tools and techniques to bail out complications.
3. To offer an overview of the outcomes.
853
OF 20T - Making the most of European scholarships and fellowships

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

Tips and tricks for successful applications (10 min)

Katharina Lampichler; Vienna / Austria
1. To describe the application process.

2. To discuss how to maximise chances of being successful.
The practicalities: moving and working in another country (10 min)

Georgios Kalarakis; Stockholm / Sweden
1. To describe the practical steps in sorting out a fellowship.

2. To discuss the challenges and benefits of working in a new country.
Looking back: what I gained from an ESOR scholarship (10 min)

David Castanedo Vázquez; Santander / Spain
1. To describe the personal experience of undertaking an ESOR scholarship.

2. To discuss how ESOR can benefit and shape one's career.
Looking back: what I gained from an ESOR research fellowship (10 min)
Laura Isabel Loebelenz; Bern / Switzerland
1. To describe the personal experience of undertaking a research fellowship.

2. To discuss how ESOR can benefit and shape one's career.
Open forum discussion: Top tips for potential ESOR applicants (15 min)
854
RPS 2003 - Clinical applications of AI and ML in cardiovascular imaging

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Sara Boccalini; Villeurbanne / France
Deep learning denoising in cardiac CT imaging: improved image quality and workflow efficiency (7 min)
Andreas Stefan Brendlin; Tübingen / Germany
Author Block: A. S. Brendlin, D. Wessling, J. Hofmann, J. Mück, B. Stenzl, S. Afat; Tübingen/DE

Purpose: To evaluate the impact of a deep learning denoising (DLD) algorithm on image quality, diagnostic confidence, and
radiological workflows in the context of cardiac computed tomography angiography (CCTA).
Methods or Background: One-hundred patients (mean age 60±11 years) with a CCTA were included. Images were reconstructed
using iterative reconstruction strength 2 (IR2) and the DLD algorithm. Place-consistent noise measurements were used to compare
objective image quality. In addition, two blinded readers independently assessed subjective image quality, diagnostic confidence,
sharpness, and contrast in a forced-choice setup. The results of these assessments were summarised for a semiquantitative overall
quality score. Agatston-score and cardiac age were analysed semi-automatically for both datasets using proprietary software, before
and after manually correcting the initial software output. The time required for manual corrections was measured for each reader to
compare possible workflow benefits. Properly corrected mixed-effects analysis with post hoc subgroup tests was used. Spearman's
correlation coefficient measured inter-reader agreement for the image quality analysis.
Results or Findings: Noise in IR2 was significantly higher than for DLD (22.00±2.32 versus 13.33±2.87 HU; p<0.001). DLD
reconstructions had a significantly higher mean overall quality score than IR2 (3.5±1.0 versus 0.48±1.0, p<0.001) with a good inter-
rater agreement (r≥0.790; p≤0.001). There were no significant differences between cardiac age results (p=0.517) and Agatson score
values (p=0.486) of IR2 and DLD. However, the time required for manual corrections was significantly shorter for DLD than for IR2
(54±44 versus 35±31 seconds, p<0.001).
Conclusion: DLD significantly improves image quality in cardiac CT and substantially reduces the time required for manual
corrections in cardiac age assessment.
Limitations: Identified limitations were: (1) the single-centre design; (2) the retrospective design; (3) the specific hardware and
software setup.
Ethics committee - additional information: The institutional review board approved retrospective eligibility analysis and data
collection for this single-centre study's purpose with a waiver for the need for informed consent (reference number: 167/2022BO2).
Impact of AI-enabled motion compensation algorithm on coronary computed tomography angiography image quality (7
min)
Giuseppe Stancanelli; Rome / Italy
855
Author Block: G. Stancanelli, L. Dominici, L. Conia, G. C. Pambianchi, C. Catalano, N. Galea; Rome/IT
Purpose: Motion artefacts remain a major limitation of coronary computed tomography angiography (CCTA), especially in patients
with high cardiac frequency or rhythm variability and in coronary segments subject to motion, even in relatively quiescent phases of
the cardiac cycle.
The aim of our study is to evaluate the impact on image quality of an AI-enabled interaction-free motion compensation reconstruction
algorithm (MCR) compared to a standard filtered back projection reconstruction (FBP).
Methods or Background: Fifty patients underwent CCTA on a 128-slice scanner (Incisive CT, Philips) with an ECG modulated
retrospective acquisition protocol. Raw datasets were reconstructed during the telediastolic phase of the cardiac cycle using a
standard FBP algorithm and processed on an offline workstation to generate interaction-free MCR images (Precise Cardiac Suite,
Philips).
The two image quality datasets were evaluated side by side by a reader with three years of experience blinded to the reconstruction
technique. Image quality was graded per-segment and per-patient on a 1-4 scale based on the severity of the motion artefacts
("blurring", "winging" or "stairstep").
Results or Findings: Five hundred coronary artery segments were evaluated in both FBP and MCR reconstruction datasets; per
segment coronary segmental image quality scores are reported in Fig 1. We observed a global statistically significant increase of
mean scores after the application of the MCR algorithm (FBP: 2.75±1.04; MCR: 2.82±1.02 [p<0.01]. Overall, 15 out of 75 non-
diagnostic segments were reclassified as diagnostic on MCR images, most of which (8/11) were on midRCA (p=0.02).
Conclusion: The application of MCR algorithm resulted in a global reduction of motion artefacts and an increase in image quality,
with reclassification of non-diagnostic segments most evident on midRCA. This determined a better diagnostic performance of CCTA
on segments most prone to motion artefacts.
Ethics committee - additional information: No additional information provided by the submitter.
Automatic AI-based calcium scoring in cardiac and chest computed tomography: a validation study (7 min)
Iris Hamelink; Zwolle / Netherlands
Author Block: I. Hamelink, Z. Nie, T. Kwee, M. Dorrius, P. M. A. Van Ooijen, R. Vliegenthart; Groningen/NL
Purpose: The extent of coronary artery calcium (CAC), assessed on computed tomography (CT), is a strong predictor of
cardiovascular disease. The aim of this study is to validate the performance of an automatic AI system for quantifying CAC.
Methods or Background: 687 participants (59±4.8 years; 48.8% men) of the population-based ImaLife cohort were analysed for
CAC. The Agatston score (AS) on cardiac and chest CT scans were quantified manually by a radiologist and automatically by an AI
system (AI-Rad Companion Chest prototype, Siemens Healthineers). Agreement of manual and AI measurements was assessed by
sensitivity and accuracy, Bland-Altman analysis and Cohen’s kappa for classification in AS strata (0; 1-99; 100-299; ≥300).
Results or Findings: Three participants were excluded due to incorrect manual measurement or a history of coronary stenting,
resulting in 684 participants for evaluation. In cardiac CT, 200 (29%) participants showed no CAC when evaluated manually. 331
(48.4%) participants showed AS between 1 and 99, 92 (13.5%) participants between 100 and 299 and 61 (8.9%) participants ≥300. AI
software showed a high sensitivity for CAC: 98.1% in cardiac CT (accuracy 97.2%) and 95.4% in chest CT (accuracy 92.1%). Bland-
Altman analysis showed systematic bias of 2.3 and repeatability coefficient of 23.0 for AS on cardiac CT; and -0.3 and 38.0 for AS on
chest CT. Cohen’s kappa for agreement in AS categorisation was 0.94 for cardiac CT and 0.87 for chest CT, with concordance in 96.0
and 91.4% of cases, respectively.
Conclusion: AI-based CAC scoring shows a high detection rate compared to manual evaluation, with excellent performance for risk
classification. Performance is slightly better in cardiac CT than in chest CT.
Limitations: For both scan protocols, a low-dose, high-pitch scan protocol was used; it is unclear how generalisable results are for
other scan protocols.
Funding for this study: The ImaLife study is supported by an institutional research grant from Siemens Healthineers and by the
Ministry of Economic Affairs and Climate Policy by means of the PPP Allowance, made available by the Top Sector Life Sciences &
Health to stimulate public-private partnerships.
Ethics committee - additional information: This study was approved by the Medical Ethics Committee of the University Medical
Center Groningen
AI-precision in cardiovascular risk assessment: non-gated chest CT coronary artery calcium scoring (7 min)
Dan Mu; Nanjing / China
856
Author Block: D. Mu1, K. Yin1, W. Chen1, X. Chen2, B. Zhang1; 1Nanjing/CN, 2Shanghai/CN
Purpose: This study aimed to assess the performance of an artificial intelligence-based coronary artery calcium score (AI-CACS)
algorithm on non-gated chest computed tomography (CT) images in differentiating risk categories for cardiovascular diseases.
Methods or Background: A prospective study enrolled 112 patients who underwent both chest CT and electrocardiogram (ECG)-
gated non-contrast enhanced cardiac CT using the same equipment simultaneously. Chest CT images were reconstructed at three
different thicknesses (1 mm, 3 mm, and 5 mm). The Agatston score, obtained semi-automatically from ECG-gated cardiac CT scans
using a dedicated post-processing workstation, served as the reference. The AI-CACS software automatically derived the Agatston
score from chest CT data. Correlations between AI-CACS and the reference Agatston score were calculated. The AI-CACS's
performance in classifying risk categories, dichotomised at thresholds of 0, 100, and 400, was assessed.
Results or Findings: The AI-CACS showed strong correlations with the reference Agatston score for the three different slice
thicknesses (1 mm: 0.973, 3 mm: 0.941, 5 mm: 0.834; all p < 0.001). Agreement in risk categories, assessed using kappa (κ)
statistics, was substantial (κ = 0.868, p < 0.001), moderate (κ = 0.772, p < 0.001), and fair (κ = 0.412, p < 0.001) for 1 mm, 3 mm,
and 5 mm slice thicknesses, respectively, with concordance rates of 91%, 84.8%, and 62.5%. When dichotomised at thresholds of 0,
100, and 400, the area under the curve for AI-CACS at the three slice thicknesses ranged from 0.785 to 0.996, 0.975 to 0.995, and
0.981 to 1.000, respectively.
Conclusion: The AI-CACS algorithm applied to chest CT images demonstrates promising performance in assessing cardiovascular
disease risk. Using a 1 mm slice thickness for image reconstruction may yield the best results.
Limitations: A larger multi-centred, multi-vendor cohort study shall be conducted.
Ethics committee - additional information: This study was ethically approved by the IRB number: 2022-547-01.
AI-based quantification of total and vessel-specific coronary artery calcifications in calcium scoring CT (7 min)
Lilian Henriksson; Linköping / Sweden
Author Block: L. Henriksson, M. Sandstedt, A. Persson; Linköping/SE

Purpose: The objective of this study was to assess the correlation and agreement between fully automatic AI and standard semi-
automatic evaluations by radiologists in calcium scoring CT (CSCT) examinations within the Swedish CardioPulmonary bio-image
study (SCAPIS).
Methods or Background: A retrospective observational study involving 4567 CSCT exams carried out on a dual-source 128x128
slice CT scanner (Somatom Definition Flash, Siemens Healthineers, Forchheim, Germany) was conducted. Automatic AI measurements
and semi-automatic measurements included Agatston score (AS), volume score (VS), mass score (MS), and number of lesions in the
coronary arteries (LM+LAD, CX, RCA).
Results or Findings: Pearson correlation coefficients (r) for total AS, VS, and MS were r=0.989, 0.988, and 0.988 respectively. Intra-
class correlation coefficients (ICCs) showed high levels of agreement: 0.994 for total AS, 0.998 for VS, 0.998 for MS, and 0.960 for the
number of lesions. Bland-Altman analysis indicated minimal bias across all metrics. Weighted kappa for risk category classifications
was = 0.919, and the overall accuracy was 91.2%.
Conclusion: The study demonstrates excellent correlation and agreement between fully automated AI and radiologists' semi-
automatic evaluations in CSCT examinations. The findings suggest that AI could be a reliable tool for calcium scoring, with
implications for improving efficiency and standardisation in radiological assessments.
Limitations: Even though this is the largest study made so far regarding the accuracy of AI CSCT evaluations the study is limited by
only including CSCT examinations performed on one type of CT scanner.
Ethics committee - additional information: This study was ethically approved by the Swedish ethical review authority, regional
ethical review board in Göteborg: DNR 2021-00441.
A deep learning algorithm for fully-automated myocardial infarct scar segmentation (7 min)
Matthias Schwab; Innsbruck / Austria
857
Author Block: M. Schwab, M. Pamminger, C. Kremser, M. Haltmeier, A. Mayr; Innsbruck/AT
Purpose: Late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) imaging is considered the in vivo reference
standard for assessing infarct size and microvascular obstruction (MVO) in ST-elevation myocardial infarction patients. As LGE
distribution patterns can be quite complex and hard to delineate from the blood pool or epicardial fat, automatic segmentation of LGE
CMR images is challenging. The aim of this work is to develop and evaluate a deep learning-based method that allows to perform LGE
segmentation in a fully-automated way.
Methods or Background: A cascaded framework of two-dimensional and three-dimensional convolutional neural networks (CNNs),
specialized on identifying myocardial scars, was trained on a training data set consisting of 224 patients. On a test data set including
LGE CMR images from 152 examinations AI-based segmentations were compared to manual segmentations, which were performed
according to the +5-SD method. Further, on a big data set of 1012 patients automatically calculated infarct volumes were correlated
with maximum levels of creatine kinase and cardiac troponin obtained after acute myocardial infarction and successful primary
percutaneous coronary interventions.
Results or Findings: Mean Dice coefficients between manual and CNN segmentations were 64.1% for LGE and 85.3% for MVO,
respectively. Further, linear correlation between manually and automatically calculated infarct sizes was very strong (R=0.95,
p<0.001). Good correlation between AI measured LGE volumes and biochemical measurements could also be found (creatine kinase:
R=0.72, p<0.001; cardiac troponin: R=0.67 p<0.001).
Conclusion: Our fully-automated framework for LGE segmentation provides measurements that can compete with the very time-
consuming manual segmentations.
Limitations: The limitations of the study are that for evaluation of the method, data from only one hospital (University Hospital
Innsbruck) was used.
Funding for this study: Funding was provided by the Austrian Science Fund (FWF): DOC 110.
Ethics committee - additional information: The study utilised retrospective data analysis.
Artificial intelligence-based CT-derived FFR for the detection of haemodynamically significant coronary artery disease:
a comparative study with dynamic stress CT myocardial perfusion imaging (7 min)
Antoine Andary; Lille / France
Author Block: A. Andary, A. Rodriguez Musso, C. V. Gkizas, P. Carpentier, N. Abassebay, C. Lardemelle, B. Longere, F. Pontana;
Lille/FR
Purpose: CT myocardial perfusion imaging (CT-MPI) combined with coronary CTA integrates coronary artery anatomy with inducible
ischaemia at the cost of a higher radiation and contrast reinjection. The aim of this study was to evaluate the diagnostic performance
of a deep-learning model of CT-derived fractional flow reserve (FFR-AI) for the detection of haemodynamically significant coronary
artery disease (CAD) compared to CT-MPI.
Methods or Background: This retrospective study included 36 patients who underwent coronary CTA and dynamic stress CT-MPI on
a third-generation dual-source CT system (SOMATOM Force, Siemens Healthineers). CT-MPI was performed when the maximal
coronary stenosis was ≥50% (CAD-RADS≥3) or in the presence of stent, according to our centre’s practice after injection of
regadenoson. Perfusion maps were interpreted by two radiologists by consensus. A perfusion defect was defined as a visually
significant anomaly on the myocardial blood flow (MBF) map, in a coronary territory (windowing at 100 mL/100 mL/min by default).
Curvilinear images of the main coronary arteries were then exported to CorEx model (version 1.0; Spimed-AI), which classified each of
these arteries into two categories: FFR ≤0.8 or FFR >0.8.
Results or Findings: CT-MPI detected perfusion defects in 16 of 36 patients (44%). FFR-AI demonstrated a per-patient sensitivity,
specificity, PPV, NPV and accuracy for the detection of hemodynamically significant stenosis of 100% (95% CI: 79%-100%), 50% (95%
CI: 27%-73%), 61.5% (95% CI: 51%-71%), 100% and 72% (95% CI: 55%-86%), respectively. The areas under the ROC curve of FFR-AI
were 0.75 (95% CI: 0.78-0.88).
Conclusion: FFR-AI provides high sensitivity and NPV for identifying haemodynamically significant CAD among patients with coronary
stenosis ≥50%. FFR-AI could be used as a filter to avoid a subsequent CT-MPI and reduce radiation exposure and contrast reinjection.
Machine learning explainable analysis for prediction of atrial fibrillation recurrence after catheter ablation using
clinical and radiological variables (7 min)
Alvaro Palazón Ruiz De Temiño; Alicante / Spain
858
Author Block: A. P. Ruiz De Temiño, J. M. Castro, M. J. Garfias, H. Trigueros Buil, A. Adarve Castro, D. F. Ferrández, B. Martínez-López;
Alicante/ES
Purpose: Atrial fibrillation (AF) is a common arrhythmia with increasing prevalence and significant clinical impact. Catheter ablation
has emerged as a treatment option for drug-resistant AF, with variable success rates. This study aimed to develop a machine
learning-based model to predict AF recurrence after pulmonary vein ablation.
Methods or Background: A retrospective case-control study included patients who underwent first radiofrequency or cryoablation
between 2017 and 2022. CT scans were used to measure left atrial volume (LAV), periatrial adipose tissue (PAT), interatrial adipose
tissue (IAT), and (EAT) epicardial adipose tissue volume. Demographic, clinical, and recurrence data were collected. Feature selection
and data preprocessing were conducted, followed by model training using three machine learning techniques. Model evaluation
included accuracy, precision, recall, F1-score, and ROC/AUC. SHAP analysis was performed to interpret feature importance.
Results or Findings: Sixty nine patients were included. Recurrence occurred in 29% of patients. Persistent AF exhibited a higher risk
of recurrence (OR 1.99, p<0.05). Radiological variables like left atrial, PAT and IAT volumes were significantly higher in recurrence
cases. The logistic regression model including clinical and radiological variables (model A) achieved the highest average precision,
accuracy, f1-score, and recall during cross-validation. Model A's accuracy in the testing group was 0.86, 0.66, and 0.86 and the AUC
were 0.91, 0.87, and 0.92 using NN, NB, and LR respectively. SHAP analysis revealed varying feature importance across techniques in
model A emphasizing the LAV, PAT and AF type.
Conclusion: This study presents two models incorporating adipose tissue measurements for predicting AF recurrence after
pulmonary vein ablation with the potential of utilizing multimodal data in predicting post-ablation outcomes for AF patients.
Limitations: Sample size is limited, which might lead to overfitting. However, undersampling, scaling and cross-validation were
employed as methods to mitigate this.
Ethics committee - additional information: Patient data was handled confidentially and informed consent was waived by the
ethics committee due to the retrospective nature and reliance on medical record review. Protocol registration number was
PI2023-045.
Impact of super resolution deep learning reconstruction with 1024 matrix in inter and intra reader reproducibility of
pre-TAVR CT measurements (7 min)
Mickaël Ohana; Strasbourg / France
Author Block: A. Walch1, F. Tatsugami2, W. Fukumoto2, D. Touitou-Gottenberg1, A. Taniguchi3, K. Haioun3, K. Awai2, C. Roy1, M.
Ohana1; 1Strasbourg/FR, 2Hiroshima/JP, 3Otawara/JP
Purpose: Reproducibility of aortic annulus sizing and aortic valve opening area planimetry on pre-TAVR cardiac CT is essential.
Whether the use of a Super Resolution Deep Learning Reconstruction (SR-DLR) algorithm with increased matrix size could modify the
inter and intra reader correlation of these measurements, particularly in case of heavily calcified aortic cusps, is unknown. Our
primary objective is therefore to compare inter and intra reader reproducibility of aortic annulus and aortic valve area planimetry
measurements between DLR and SR-DLR.
Methods or Background: Forty pre-TAVR CT with excellent image quality were retrospectively selected from two tertiary centers.
Systolic phase was reconstructed with DLR in 512 and SR-DLR in 1024 matrixes. Four radiologists with different levels of expertise
independently and randomly reviewed all 80 datasets to assess aortic annulus area and aortic valve planimetry. Two readers redid all
measurements following a four week delay. Statistical analysis was performed using Bland-Altman plots and intraclass correlation
coefficient (ICC).
Results or Findings: Interobserver agreement for aortic annulus area were excellent and similar between DLR (ICC 0.85, 95% CI
0.82-0.88) and SR-DLR (ICC 0.87, 95% CI 0.85-0.90). Interobserver agreement for aortic valve planimetry was higher with SR-DLR (ICC
0.90, 95% CI 0.86-0.92) than with DLR (ICC 0.83, 95% CI 0.80-0.85). This difference was more pronounced in the subgroup of patients
with a heavily calcified aortic valve (calcium score >2000, n=24).
Intra-observer agreement for both measurements were slightly higher with SR-DLR.
Conclusion: SR-DLR with 1024 matrix could increase the reproducibility of aortic valve area planimetry, especially in heavily calcified
aortic valve.
Limitations: Potential clinical implications of SR-DLR on device selection were not analyzed in this study.
Ethics committee - additional information: This study was ethically approved by the IRB from Strasbourg University Hospital.
Enhancing preoperative risk assessment in noncardiac surgery (NCS): comparative evaluation of coronary CT
angiography (CCTA), CT perfusion (CTP), and CT-derived fractional flow reserve (CT-FFR) (7 min)
Federica Brilli; Rome / Italy
859
Author Block: F. Brilli, F. Catapano, C. Lisi, A. Caracciolo, M. Francone; Milan/IT
Purpose: Perioperative cardiovascular complications occur in approx. Three% of NCS hospitalisations. ESC guidelines recommended
use of CCTA in low-to-intermediate likelihood of CAD, or in patients unsuitable for non-invasive functional testing undergoing non-
urgent, intermediate/high-risk NCS. Our study sought to evaluate added value of CCTA-derived functional testing to predict
revascularisation prior to NCS as compared to an anatomical-based strategy.
Methods or Background: Single-cohort prospective observational study including 55 symptomatic patients with stable angina who
underwent CCTA prior to NCS; besides general CCTA and CTP contraindications, the presence of a planned invasive coronary
angiography (ICA) for preoperative investigation before surgery was a major exclusion criterion. CT-FFR was performed using a ML-
based algorithm for FFR simulation in all moderate to severe lesions. A ROC curve analysis was used to assess diagnostic
performances of CCTA vs CTP vs CT-FFR in patients undergoing ICA after non-invasive testing.
Results or Findings: Significant stenoses were found in 20 participants and confirmed with ICA and FFR-ICA in moderate lesions. At
ROC analysis, CTP had the largest AUC on a per-patient level (AUC = 0,84) compared with CT-FFR (0,41). The diagnostic accuracy of
CTP and CT-FFR at patient-based analysis were 91% and 79%, respectively. The patient-based sensitivity, specificity, PPV, and NPV of
CTP were 100%, 80%, 87% and 100%, whereas these values for CT-FFR (when using ≤0.80 as cutoff value) were 60%, 72%, 60%, and
88%. CCTA underperformed CTP for the diagnosis of flow-limiting coronary stenosis (accuracy at patient-based analysis: 77% vs 91%).
Conclusion: CTP offers a one-stop solution for assessing ischemic heart disease in NCS patients.
Limitations: Small sample size (55 patients) with stable angina, single-cohort observational design introducing potential selection
bias and lacked long-term follow-up data. Larger cohort studies are needed to confirm CTP role in these patients.
Artificial intelligence for automated classification of coronary lesions from computed tomography coronary
angiography scans (ALERT) (7 min)
Victor Verpalen; Amsterdam / Netherlands
Author Block: V. Verpalen1, C. Coerkamp1, J. J. H. Henriques1, J-F. Paul2, N. R. Planken1; 1Amsterdam/NL, 2Paris/FR
Purpose: The aim of this study was to evaluate the diagnostic performance of a deep-learning model (DLM) for quantifying coronary
stenosis on computed tomography coronary angiography (CTCA) using the Coronary Artery Disease-Reporting and Data System (CAD-
RADS).
Methods or Background: This single centre retrospective study included 50 patients suspected of coronary artery disease (CAD).
All CTCA examinations were obtained in routine clinical practice. Two expert readers and the DLM independently reassessed the CAD-
RADS score per patient (n=50) and per vessel (n=150). Binary classification (CAD-RADS 0-2 or 3-5) and six group classification (CAD-
RADS 0-5) were used for comparison among the human readers and between the readers and the DLM.
Results or Findings: Interhuman sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy,
and Cohen’s kappa for detecting ≥50% stenosis (binary classification) were 86.4, 85.2, 82.6, 88.5, 85.7%, and 0.71 at patient level.
Sensitivity, specificity, PPV, NPV, accuracy, and Cohen’s kappa of the DLM for detecting ≥50% stenosis were 100, 69.6, 75.0, 100,
84.1%, and 0.69 at the patient level for reader 1 and 100, 66.7, 71.4, 100, 81.8%, and 0.65 for reader 2 as reference, respectively.
For the six group classification at patient level, interhuman agreement was 65.3% and weighted kappa 0.78. For the DLM vs reader 1
and reader 2 this agreement was 54.5 and 56.8%, the weighted kappa was 0.70 and 0.61, respectively.
Conclusion: Ruling out obstructive CAD (≥50% stenosis) by the DLM is safe, considering the 100% sensitivity. The DLM yielded
promising results in CAD-RADS classification (0-5). This DLM has potential to support and alert CTCA-readers in clinical practice.
Limitations: The main limitation of the study is that the CAD-RADS distribution present in the study population does not necessarily
reflect local clinical practice, which might influence the local performance of the DLM.
Ethics committee - additional information: The study was approved by Ethics committee Amsterdam UMC: 2023.0484,
Super-resolution deep learning reconstruction for improved image quality of myocardial late enhancement CT (7 min)
Masafumi Takafuji; Tsu Mie / Japan
860
Author Block: M. Takafuji1, K. Kitagawa1, S. Mizutani2, A. Hamaguchi2, R. Kiso2, K. Sasaki2, Y. Funaki2, H. Sakuma1; 1Tsu Mie/JP,
2
Matsusaka/JP
Purpose: Myocardial late enhancement CT (LE-CT) allows assessment of myocardial scar. Super-resolution deep learning image
reconstruction (SR-DLR), which is trained on data acquired from ultra-high-resolution CT may improve image quality of LE-CT. The
purpose of this study was to investigate image noise and image quality with SR-DLR compared with conventional DLR (C-DLR) and
hybrid iterative reconstruction (hybrid-IR).
Methods or Background: We retrospectively analyzed 30 consecutive patients who underwent LE-CT using 320-row CT. The CT
protocol consisted of stress dynamic perfusion CT, coronary CT angiography and LE-CT. All images were reconstructed using three
different algorithms: SR-DLR, C-DLR, and hybrid-IR. Image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and overall
image quality were compared. Overall image quality was assessed by five independent observers. Each observer had 30 points in
each case and the points were allocated to LE-CT images reconstructed with the three different algorisms according to image quality.
The scores were averaged across all observers.
Results or Findings: SR-DLR significantly decreased image noise by 33% compared to C-DLR (6.5±1.4 HU vs 9.7±1.7 HU,
P<0.0001) and by 37% compared to hybrid IR (vs 10.4±2.8 HU, P<0.0001). SNR and CNR of LE-CT reconstructed using SR-DLR (SNR,
17.5±4.4; CNR, 4.6±0.8) were significantly higher than C-DLR (SNR, 11.4±2.8 p<0.0001; CNR, 3.1±0.6, p<0.0001) and hybrid-IR
(SNR, 11.0±3.2, p<0.0001; CNR, 3.3±0.6, p<0.0001). SR-DLR significantly improved overall image quality of LE-CT compared to C-
DLR (13.6±1.3 vs 8.6±0.7, p<0.0001) and hybrid-IR(vs 7.8±0.6, p<0.0001).
Conclusion: SR-DLR improved image noise, and image quality of myocardial LE-CT compared with C-DLR and hybrid-IR techniques.
The SR-DLR approach has the potential to improve the assessment of myocardial scar by LE-CT and to lower the tube voltage and/or
current of LE-CT, thus reducing the radiation dose of LE-CT.
Ethics committee - additional information: This study was approved by the institutional review board in our institution, and
written informed consent was obtained from each individual before enrolling in the study (reference number: 210604-5-2).
861
RPS 2002 - Advances in breast intervention
Categories: Breast, Interventional Radiology, Oncologic Imaging

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Tamar Sella; Jerusalem / Israel
Clinical pilot study using polygenetic risk score for personalised risk-based breast cancer screening (7 min)
Tone Hovda; Drammen / Norway
Author Block: T. Hovda1, K. K. Sahlberg1, S. Sõber2, P. Padrik2; 1Drammen/NO, 2Tartu/EE

Purpose: This study aimed to explore the potential for polygenetic risk score (PRS) as a tool for risk-based mammographic screening.
Methods or Background: Population-based screening for breast cancer is mainly age-based. This pilot study included 30 women
aged 40-50 without prior history of breast cancer, premalignant breast disease or prior genetic testing/counselling. We used a CE-
marked clinical test (AnteBC, Antegenes, Estonia) to assess participants’ PRS, and calculate absolute and relative 10-year breast
cancer risk. Women were recommended mammographic screening based on the results; standard biennial screening age 50-69 if
below average or medium level risk, or earlier start of biennial and/or annual screening if moderate (up to 2x increased) or high (≥2x
increased) risk. We also collected information on breast density and family cancer history, and referred the women for further genetic
testing for monogenetic pathogenic variants if indicated by family history.
Results or Findings: Mean age was 44.9 years (SD 3.0). Mean absolute 10-year-risk for breast cancer was 1.65% (95% confidence
interval [CI] 1.35, 1.91), compared to 1.74% (95% CI 1.63, 1.86) for an average population with the same age distribution. Relative
risk ≤1 was observed in 70% (21/30), relative risk 1-2 was observed in 27% (8/30) and relative risk ≥2 was observed in 3% (1/30). We
observed no statistically significant difference in relative risk by breast density. Five participants (17%) were recommended further
genetic testing due to positive family history.
Conclusion: Polygenetic risk score may be a useful tool for risk-based mammographic screening. Further studies with larger study
populations, exploring cost-effectiveness and women's perspective are needed.
Limitations: The small sample size is a limitation, as well as exclusion of women with prior malignant/premalignant breast disease
and prior genetic testing, as the results might be biased towards less risk.
Funding for this study: The AnteNOR consortium is sponsoring the study through a grant from the Norwegian Grants Green ICT
program.
Ethics committee - additional information: The study was approved by the regional ethical committee with approval number:
REK 494936.
Contrast-enhanced spectral mammography (CESM) vs MRI-guided vacuum-assisted breast biopsy: towards a reduced
number of unnecessary biopsies (7 min)
Mirjan M. Nadrljanski; Belgrade / Serbia
862
Author Block: M. M. Nadrljanski, I. B. Krušac, D. Dimitrijevic, L. Raspopovic, A. Djajic, M. Mihajlovic; Belgrade/RS
Purpose: The study aimed to assess contrast-enhanced spectral mammography (CESM) guided breast vacuum-assisted biopsy
(VAB), compared to MRI-guided VAB in benign lesion biopsy reduction.
Methods or Background: Consecutive 23 patients with suspicious findings on mammography or digital breast tomosynthesis
(n=23; MX-BI-RADS 4), were included prospectively (October 2022 – October 2023), for CESM-guided or MRI-guided VAB. All patients
were assessed with CESM. All enhancing lesions were subsequently biopsied with CESM-guided VAB. Non-enhancing lesions were
assessed with DCE-MRI (1.5 T/3 T) and the non-mass enhancement (NME) lesions were biopsied with MR-VAB. Non-enhancing lesions
on CESM and MRI were downgraded to MRI-BI-RADS 3. Histopathologic correlation was performed for all lesions.
Results or Findings: There were 11 CESM-enhancing lesions (n1=11, 47.83%) and 12 CESM-non-enhancing lesions assessed with
DCE-MRI (n2=12, 52.17%). In n1, there were eight malignant lesions (B5, 72.73%: 3 IDC, 1 ILC, 4 DCIS) and three benign lesions (B2,
27.27%). In n2, there were nine NME and three non-enhancing lesions. All nine lesions in n2 were benign (B2, 44.44%: 4 FCC and B3,
55.56%: 2 ADH, 3 FEA). CESM-guided VAB achieved Se=100% (63.1-100.0%); Sp=80% (51.9-95.7%) with NPV 100% and accuracy of
86.9%. MRI-guided VAB in CESM-non-enhancing MRI-NME lesions achieved Se=100% (47.8-100%); Sp=42.9% (9.9-81.6%) and
accuracy of 66.7%.
Conclusion: CESM-guided VAB identified all lesions subsequently histopathologically confirmed as malignant. All non-enhancing
lesions on CESM (and MRI) were benign. All CESM non-enhancing lesions detectable as NME lesions on MRI, were benign. The absence
of enhancement on CESM favours benign lesions. Should larger trials confirm the preliminary findings, CESM-guided VAB may
considerably reduce the number of unnecessary biopsies.
Limitations: The limited number of patients (due to the recently introduced method) was an identified limitation.
Ethics committee - additional information: The study was approved with approval code: 500-01-1335 CESM 03.
Based on the above stated decision, the authorisation was granted for the realisation of the procedure(s).
Diagnostic performance of Kaiser score in BI-RADS 3 lesions: a promising tool to reduce unnecessary biopsies (7 min)
Pietro Costantini; Novara / Italy
Author Block: E. Ostillio, P. Costantini, D. Razzini, L. Groenhoff, A. Tambasco, M. Brambilla, A. Gambaro, A. Carriero; Novara/IT
Purpose: To determine whether the Kaiser score can be a valuable tool for radiologists in assessing whether to recommend a biopsy
for a BI-RADS 3 lesion.
Methods or Background: Notably, BI-RADS 3 denotes considerable uncertainty in breast nodule diagnoses, situated between
benign (BI-RADS 2) and malignant (BI-RADS 4) classifications. The Kaiser score, a magnetic resonance imaging (MRI)-based algorithm
developed by Bazar et al., has proven effective for BI-RADS 4 and 5, but its suitability for BI-RADS 3 remains unexplored.
Between January 2016 and August 2023, 100 BI-RADS 3 lesions were studied, excluding 21 patients who did not meet the criteria.
These 79 remaining lesions, observed in adult patients with no prior biopsies or treatments, underwent MRI-guided biopsies. Two
expert radiologists (with 30 and 25 years of experience) applied the Kaiser score, and the results were compared with
histopathological reports.
Results or Findings: Among the biopsied lesions (n=79), 14 were positive, and 65 were negative. Comparing Kaiser score results
(Kaiser score: 1-4 = negative; 5-11 = positive) with biopsy findings, the Kaiser score exhibited a sensitivity of 100% (95% CI:
73%-100%) and specificity of 100% (95% CI: 93%-100%).
Conclusion: The strong concordance between the Kaiser score and biopsy outcomes for BI-RADS 3 lesions suggests its potential
value in clinical practice for guiding patients towards follow-up or biopsy decisions.
Limitations: This was a single-centre study and, even if it is of a modest number, the population sample could be bigger. Moreover,
two expert radiologists analysed the breast lesions through the KS together, after a common agreement, thus impeding us to
estimate the KS interobserver agreement.
Funding for this study: Funding was received from the Fondo di Ateneo per la Ricerca - University of Eastern Piedmont.
Ethics committee - additional information: As it is a retrospective study, no ethics committee approval was required at our
institution.
Optimising assessment and reducing benign biopsy rate in screen-recalled women (7 min)
Ernest Usang Ekpo; Sydney / Australia
863
Author Block: E. U. Ekpo1, I. Hadadi2, M. F. F. McEntee3; 1Sydney/AU, 2Abha/SA, 3Cork/IE
Purpose: The study aimed to optimise the assessment of women recalled at screening and reduce benign biopsy rate.
Methods or Background: We examined the assessment data of 538 women recalled at screening whose mammograms showed
features suggestive of breast cancer. Using histopathology results as a reference standard, we examined the diagnostic performance
of digital breast tomosynthesis (DBT) and ultrasound as assessment tools. Relative risk analysis and McNemar tests were used to
assess and compare the potential for these assessment tools to reduce benign biopsy rates.
Results or Findings: The recall rate was higher in dense breasts. DBT demonstrated higher sensitivity than ultrasound in dense
breasts (98.2% versus 80%; p<0.001), but lower specificity (15.4% versus 55%; p<0.001), positive predictive value (PPV) (61.3%
versus 71%; p=0.04), and area under the receiver operating characteristic curve (AUC) (0.57 versus 0.67; p=0.001). DBT showed
higher sensitivity than ultrasound (99.2% versus 84%; p<0.001) in non-dense breasts, but no differences were observed in specificity,
PPV, and AUC. Calcifications were easily detected but overestimated on DBT relative to ultrasound. Ultrasound reduced the benign
biopsy rate compared to DBT: BI-RADS A (21% versus 5%; p=0.04); BI-RADS B (23% versus 10%; p=0.003); BI-RADS C (34% versus
7%; p<0.001) and BI-RADS D (39% versus 9%; p<0.001). The number needed to assess to prevent one benign biopsy was
significantly lower with ultrasound than DBT in dense breasts: BI-RADS C (1.8 versus 7; p<0.001) and BI-RADS D (1.9 versus 5.1;
p=0.03).
Conclusion: In women with dense breasts, DBT has higher sensitivity, but lower specificity and PPV than ultrasound. Compared to
DBT, ultrasound reduces the benign biopsy rate for all women.
Limitations: Data were from one facility and 60% of the sample was dense breasts. Also, DBT and ultrasound images were
interpreted with knowledge of mammography findings.
Funding for this study: This study was supported by Tour De Cure mid-career funding; University of Sydney, Faculty of Medicine
and Health EMCR Emerging Star funding.
Ethics committee - additional information: This study was approved by: Sydney Local Health District; Royal Prince Alfred Hospital
Human Research Ethics Committee.
The ICE Study: to detect cryoimmunologic response induced by ultrasound-guided cryoablation on early breast cancer:
preliminary results (7 min)
Francesca Galati; Rome / Italy
Author Block: F. Galati, M. Pasculli, V. Rizzo, G. Moffa, C. Napoletano, B. Cerbelli, C. Catalano, F. Pediconi; Rome/IT
Purpose: The ICE Study is a pilot, prospective, case-control study with the primary aim of characterising the inflammatory response
induced by ultrasound-guided tumour cryoablation, in blood samples from breast cancer (BC) patients.
Secondary endpoints are the evaluation of cryoablation efficacy and safety, and imaging prediction of cryoablation effectiveness.
Methods or Background: We enrolled patients with an early-stage BC, scheduled for breast surgery, not eligible for neo-adjuvant
therapy and with a cryo-feasible cancer location. Patients enrolled in the cryo-group and in the control-group followed the same
therapeutic pathway for the treatment of BC, in terms of blood sampling and surgery. However, the control-group did not undergo
cryoablation.
Results or Findings: From July 2022 we enrolled 10 women in the cryo-group and 10 women in the control-group. On the surgical
samples of the cryo-group, ultrasound-guided cryoablation caused a steatonecrotic area in all the patients and the treatment was
complete in 9 out of 10 patients.
Cryoablation success was evaluated with magnetic resonance imaging (MRI) in 5 patients, with contrast enhanced mammography
(CEM) in 4 patients and with breast ultrasound in 1 patient who refused contrast-enhanced imaging. MRI and CEM predicted
cryoablation efficacy in 9 out of 9 cases. Regarding circulating markers of cryo-immunological response, the analysis of blood samples
revealed a release of HGBM1, which acts as a key mediator of the immune system signalling cellular stress, in both control and cryo-
groups. Furthermore, in the cryo-group increased levels of HGBM1 were associated with a significant rise in CD3+ T cells.
Conclusion: Cryoablation is safe, effective and has a role in immune system modulation.
Enhanced imaging (MRI and CEM) can predict procedure success.
Limitations: The limited patient population was an identified limitation.
Funding for this study: "The ICE Study" received funding from the Seed Grant funding programme of the European Society of
Radiology (ESR) in collaboration with the European Institute for Biomedical Imaging Research (EIBIR), kindly supported by an
unrestricted, non-exclusive grant from GE Healthcare.
Ethics committee - additional information: The study obtained the approval of the Institutional Review Board of Sapienza
University of Rome (reference: 6528, approved 24.11.2021).
Role of vacuum assisted excision in treatment of ductal carcinoma in situ: preliminary results of a monocentric
prospective pilot study (7 min)
Serena Carriero; Milan / Italy
864
Author Block: S. Carriero, L. Nicosia, A. C. Bozzini, F. Pesapane, A. Latronico, M. Pizzamiglio, E. Cassano; Milan/IT
Purpose: The study aimed to evaluate effectiveness, safety and underestimation of vacuum assisted excision (VAE) compared to
surgery in patients with low- and intermediate-grade ductal carcinoma in situ(DCIS).
Methods or Background: From April 2023 to October 2023, all patients with a single cluster of microcalcifications (BI-RADS >3) <15
mm were enrolled. The goal in these patients was to perform VAE of the cluster. The procedure was done using an 8-gauge needle in
two stages: first, eight samples were taken with macroscopic excision of the focus, then 4 "cleaning" samples were taken. The
samples were categorised into three groups: micro-positive sample (MPS), micro-negative sample (MNS), and cleaning sample (CS). In
all cases, the presence of any residual disease within the cleaning area and the rate of post-procedural complications were assessed.
A reference clip was always placed at the end of the procedure. All lesions underwent surgery. After surgery, the presence of any
residual disease in the surgical specimen was evaluated.
Results or Findings: A total of 98 patients with microcalcifications were prospectively enrolled, resulting in 49 DCI. Seven patients
underwent VAE. In 100% of cases (7/7), the underestimation rate was 0%. In 57% (4/7) of cases, no residual disease was found within
the cleaning area, and no in-situ pathology was detected in the subsequent surgery. In 43% (3/7), disease was found within the
cleaning area, and foci of DCIS were there at surgery. No major post-procedural complications were observed.
Conclusion: Preliminary data have shown that VAE is a safe procedure with no diagnostic underestimation, which could be used in
the excision of DCIS as an alternative to surgery, especially in cases where the "cleaning" area is free of DCIS.
Limitations: The major limitation of our study is the limited sample size of the patients
Ethics committee - additional information: The approval for this monocentric prospective study was obtained by the Ethics
Committee of IRCCS IEO, Milano.
Cancer detection in relation to type and stage in the randomised Mammography Screening with Artificial Intelligence
trial (MASAI) (7 min)
Kristina Lang; Malmö / Sweden
Author Block: V. Hernström, V. Josefsson, H. Sartor, D. Schmidt, A-M. Larsson, I. Andersson, A. Rosso, O. Hagberg, K. Lang; Malmö/SE
Purpose: To analyse cancer-detection rates and types of detected cancers in the full MASAI-trial study population.
Methods or Background: 105,934 screening participants were randomised (1:1) to AI-supported screening or double-reading
without AI (Standard of Care, SoC). In the intervention arm, screening examinations were triaged to low- and high-risk groups based
on AI-derived risk scores (ScreenPoint, Transpara v 1.7 score 1–9 and 10, respectively). Low-risk examinations were single read and
high-risk examinations double read. Computer-aided detection marks were available for examinations with risk score 8–10. A per
protocol analysis on cancer detection in relation to subtype and stage was performed and cancer-detection rates were compared
using the Fisher’s exact test with 95% confidence intervals (CI). ClinicalTrials.gov number NCT04838756.
Results or Findings: 53,043 women were allocated to AI-supported screening and 52,872 to SoC. Mean age was 55.1 (SD 10.2) in
both arms. 338 cancers were detected with AI-support and 263 with SoC, a cancer-detection rate of 6.4 (95% CI 5.7–7.1) vs. 5.0
(4.4–5.6) per 1000, a ratio of 1.28 (1.09–1.50, p=0.002), an increased detection of 51 invasive and 24 in situ cancers. AI-supported
screening detected 204 cancers with non-specific histologic type and 122 cancers with T1c-stage, compared to 155 and 79 with SoC.
Notably, 20 more non-luminal A invasive cancers, and 12 more DCIS grade 3 were detected with AI-support.
Conclusion: An AI-supported screen reading protocol resulted in a 28% increase in cancer detection compared to double reading
without AI. Subtype and stage of detected cancers suggest that AI can aid in early detection of clinically relevant cancers.
Limitations: It is a single-institution trial.
Funding for this study: Funding was received from: The Swedish Cancer Society; Regional Cancer Centres; Lund University ALF-
funds.
Ethics committee - additional information: The study was approved by the Swedish Ethical Review Authority (2020-04936).
Performance of vacuum-assisted excision (VAE) with "cavity margins shaving" technique in small clusters of suspicious
microcalcifications (7 min)
Francesca Morciano; Rome / Italy
865
Author Block: F. Morciano1, R. Rella1, M. Conti1, P. Belli1, O. Tommasini1, F. Fornasa2, E. Gori1, G. Romanucci2; 1Rome/IT, 2Verona/IT
Purpose: The study aimed to assess the diagnostic performance of vacuum-assisted excision (VAE) under stereotactic guidance with
"cavity margins shaving" to ensure total lesion removal in patients with ≤1 cm clusters of microcalcifications.
Methods or Background: Patients with ≤1 cm clusters of microcalcifications were eligible. All VAE procedures were performed
under stereotactic guidance with a 9G-needle. The first 12 specimens were retrieved into the first sample container (SC); then SC was
changed without removing the biopsy needle and cavity margins were shaved with 12 additional specimens into a second SC.
Inclusion criteria: complete removal of microcalcifications; B5 or B3 lesion; surgical histopathological examination or imaging follow-
up (FUP) ≥24 months. Histopathologic assessment of specimens of the first and second containers was performed separately.
Presence of residual lesion in the second SC was compared with surgical excision histology or imaging-FUP evolution, evaluating
negative predictive value (NPV) and false-negative rate (FNR).
Results or Findings: A total of 60 lesions were included: 15 B5 and 45 B3. Surgery was performed on 33/60 lesions (all B5 and 18
B3) while the remaining 27/60 B3 lesions were sent to FUP, with one interval change with subsequent surgical excision revealing a
ductal carcinoma in situ (DCIS). NPV of the absence of residual lesion in the second container was 81.5% (95% CI: 61.2%-93.0%). FNR
was 18.5% (95% CI: 7.0%-38.7%). None of the atypical ductal hyperplasia (ADH) (n=14) or DCIS (n=11) cases were false-negative.
Conclusion: The absence of residual lesion in the "cavity margins shaving" of VAE (using a second SC) can predict complete lesion
removal in small clusters of microcalcifications. VAE with "cavity margins shaving" technique is effective in confirming complete lesion
removal in small clusters of microcalcifications and seems to be a promising decision-support tool to reduce subsequent surgical
excision in selected lesions, such as ADH or DCIS.
Influence of complete lesion removal at VAB on the upgrade of B3 lesions presenting as microcalcifications: five years
of experience from a referral centre (7 min)
Giovanni Irmici; Milan / Italy
Author Block: G. Irmici1, C. Depretto1, L. Rabiolo2, G. Della Pepa1, E. D’Ascoli1, C. De Berardinis1, S. Schiaffino3, A. Cozzi3, G. P.
1 1 2 3
Scaperrotta ; Milan/IT, Palermo/IT, Lugano/CH
Purpose: B3 breast lesions have uncertain malignant potential, with different risks of upgrade to malignancy at surgery and/or
follow-up. This study on B3 lesions presenting as microcalcifications, for which vacuum-assisted biopsy (VAB) represents the standard
approach, aimed to investigate the influence of complete or partial lesion removal at VAB on the subsequent upgrade rate (UR).
Methods or Background: For this retrospective monocentric study conducted in a referral centre, we retrieved 165 lesions
presenting solely as microcalcifications and being categorised as BI-RADS 4/5 at mammography, then subsequently diagnosed as B3
at VAB (40 ADH, 53 FEA, 40 ALH, 18 PL, and 14 RS) between January 2016 and December 2020. Surgical pathology or at least 3-years
follow-up were obtained to determine eventual lesion upgrade to malignancy. The χ², Fisher’s, and Mantel-Haenszel tests were
performed to assess if complete lesion removal influenced URs, overall and among different B3 subtypes.
Results or Findings: Complete lesion removal was achieved in 99/165 (60.0%) cases. The rate of complete removal did not
significantly differ among B3 subtypes (p=0.092, Bonferroni-adjusted multiple comparisons p≥0.101), ranging from 33.3% of PL
(6/18) to 78.6% (11/14) of RS.
The overall UR was 8.5% (95% CI 5.1–13.7%, 14/165), not significantly differing among B3 subtypes (p=0.562).
Conversely, the UR of completely removed lesions (4.0%, 95% CI 1.6–9.9%) was significantly lower than that of partially removed
lesions (15.2%, 95% CI 8.4–25.7%, p=0.020).
At stratified analysis according to B3 subtypes, the risk ratio of upgrade among completely and partially removed FEA (0.15, 95% CI
0.01–1.26) was significantly lower (Mantel-Haenszel test p=0.034) than those of ADH (0.36, 95% CI 0.07–1.90) and of ALH (0.74, 95%
CI 0.05–10.99).
Conclusion: The UR of B3 lesions is significantly influenced by complete lesion removal, both overall and among different B3
subtypes.
Limitations: The relatively small sample size was an identified limitation.
Ethics committee - additional information: No additional information provided by the submitter.
High-risk breast lesions treated with vacuum-assisted excision (VAE): a systematic review and meta-analysis of 3,975
cases (7 min)
Andrea Cozzi; Lugano / Switzerland
866
Author Block: A. Cozzi1, M. Cao2, F. Del Grande1, N. Sharma3, S. Schiaffino1; 1Lugano/CH, 2Milan/IT, 3Leeds/UK
Purpose: The objective of this study was to perform a systematic review and meta-analysis of the upgrade rates to malignancy and
the surgical excision rates of B3 lesions treated with vacuum-assisted excision (VAE) after core-needle biopsy (CNB), while also
analysing the procedural complication rate of VAE.
Methods or Background: After protocol registration on PROSPERO (CRD42023396663), PubMed and EMBASE were searched for
articles published up to 15/04/2023 reporting the use of VAE on B3 lesions diagnosed at CNB, taking surgical pathology and/or follow-
up as reference standard. Three readers independently performed article selection and extracted data for the following endpoints:
immediate upgrade rate, rate of complications, rates of immediate and long-term surgical excision, upgrade rates after immediate
surgical excision and during follow-up. Random-effects meta-analyses of single proportions were performed for each endpoint.
Results or Findings: Sixteen studies (3,941 patients, 3,975 VAE procedures) from six countries, published between 2008 and 2023,
were included in quantitative synthesis. The summary immediate upgrade rate at VAE was 4.0% (95% CI 1.4-7.6%). A total of 81
complications were found among 854 patients, the overall summary complication rate being 6.6% (95% CI 1.3-14.6%): 79 (97.5%)
were minor complications whereas only two (2.5%) were major complications. Immediate surgical excision was performed in 214
cases, with a 1.5% immediate surgical excision summary rate (95% CI 0.1%-3.6%) and a corresponding 25.4% summary upgrade rate
(95% CI 11.0-42.4%). Among 1,375 patients in follow-up, surgical excision was performed in 39 cases, with a 1.4% summary rate of
surgical excision during follow-up (95% CI 0.1%-3.4%) and a summary upgrade rate during follow-up of 0.01% (95% CI 0.0-0.6%).
Conclusion: VAE of B3 lesions has low rates of procedural complications, immediate upgrade to malignancy, and subsequent surgical
excision, highlighting its promising role as a first-line treatment for high-risk lesions.
Limitations: Substantial-to-high heterogeneity (I²>74.1%) was an identified limitation.
Ethics committee - additional information: This systematic review and meta-analysis did not need any specific ethics committee
approval.
Intertumoral heterogeneity in primary breast tumours and synchronous axillary lymph node metastases: comparing
immunohistochemical expression for concurrent core needle biopsy consideration (7 min)
Muhammed Şamil Aydın; Kayseri / Turkey
Author Block: M. Ş. Aydın, S. Doğan, H. Akgün; Kayseri/TR

Purpose: Breast cancer is a heterogeneous disease. Receptor expression can change throughout the progression of the disease.
Clinically significant discordance has been detected between the primary breast tumour (PT) and metastases other than metastatic
synchronous axillary lymph nodes (ALN). Therefore, routine biopsy is recommended from metastatic lesion. We aimed to evaluate the
discordance between PT and ALN and address the need to concurrently biopsy ALN.
Methods or Background: We prospectively performed core needle biopsies on 95 patients from September 2018 to September
2023 and performed immunohistochemistry (IHC) staining for oestrogen receptor (ER), progesterone receptor (PR), human epidermal
growth factor receptor-2 (HER2), and Ki67 in both PTs and ALNs. Cutoff values were >1% for ER/PR and %15 for ki67. HER2 status was
determined following ASCO/CAP guidelines. We categorized PTs and ALNs into intrinsic subtypes based on the St Gallen classification.
We assessed the consistency using the Cohen's Kappa test.
Results or Findings: Discordance rates between PT and ALN were %6 for ER, %15 for PR, %7 for HER2 and %8 for Ki67. Very good
agreement was observed for ER and HER2 (k-values 0.845, 0.812 respectively) and good agreement was observed for PR and Ki67 (k-
value 0.666, 0.619). Discordance rates for intrinsic subtypes was found to be %21, and good agreement (k-value 0.724) was
observed. There were 10 patients in whom PT was hormone receptor and/or HER2 negative but were positive for these markers in
ALN.
Conclusion: There was good statistical consistency between PT and ALN receptor expression. But there were clinically significant
discordances that may affect the treatment selection in some patients. Given the lack of complete concordance between PT and ALN,
concurrent ALN core needle biopsy may still be considered. Larger studies will be necessary to further define the need of concurrent
biopsy.
Limitations: The study included a small sample size.
Ethics committee - additional information: This study was approved by the Ethics Committee of Erciyes University Faculty of
Medicine under reference number 2018/588.
867
RPS 2005 - Recent AI advancements in MRI for precision imaging
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Multidisciplinary

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Marwan El-Koussy; Berne / Switzerland
T2-weighted fat-saturated and PD-weighted contrast synthesis from knee MRF maps via deep-learning: clinical
feasibility study (7 min)
Mika Tapani Nevalainen; Oulu / Finland
Author Block: M. T. Nevalainen1, O. Nykänen2, J. Järvinen1, V. Casula1, L. Räsänen1, M. Nissi2, M. T. Nieminen1; 1Oulu/FI, 2Kuopio/FI
Purpose: To compare the diagnostic performance of the magnetic resonance fingerprinting (MRF)-derived synthetic MR images of
the knee against conventional MR images.
Methods or Background: MRF is an emerging technique to rapidly produce quantitative MRI maps, but its clinical feasibility remains
low. However, through deep-learning (DL) synthetic conventional contrasts can be now derived. In this single-center prospective study
78 subjects with knee osteoarthritis were scanned on 3T scanners with isotropic T2-weighted fat-saturated (T2w fs) and PD-weighted
(PDw) sequences, and with sagittal MRF sequence. U-net’s were trained to synthesize contrasts from the MRF raw data. Thirty nine
cases (1014 images) were used for DL training/validation and 39 cases (1008 images) for diagnostic comparison. Two experienced
fellowship-trained musculoskeletal radiologists performed the MOAKS grading and Likert scale assessment of image quality. The inter-
rater and inter-method reliability were evaluated using Cohen's kappa and percentages of exact matches.
Results or Findings: The inter-rater reliability was 0.720 (CI 0.625-0.814) for synthetic images and 0.755 (CI 0.678-0.832) for
conventional images. Varying inter-method reliability was observed between the synthetic and conventional images: for cartilage
grades -values varied between 0.418-0.862 (mean 0.747; exact matches 85.9%), for bone marrow lesions between 0.196-0.945
(mean 0.637; exact matches 91.0%), for osteophytes between 0.701-0.885 (mean 0.729; exact matches 60.3%), for meniscus
pathology between 0.505-0.782 (mean 0.621; exact matches 78.2%). For effusion and Baker’s cyst -values were 0.868 and 0.846,
and exact matches 71.8% and 92.3%, respectively. The average Likert scores were better for the conventional than synthetic images
(4.6 vs 3.2 for T2w fs) and (4.8 vs. 3.9 for PDw).
Conclusion: The MRF-derived DL-based synthetic clinical contrasts provide good interchangeability with state-of-the-art conventional
MR sequences; however, further development is needed to enhance the image quality.
Limitations: Small number of patients, only sagittal images.
Funding for this study: The funding of this study was received from The Finnish Medical Foundation, The Finnish Cultural
Foundation, The Terttu Foundation.
Ethics committee - additional information: This study was approved by the regional medical research ethics committee of the
Wellbeing services county of North Ostrobothnia
Leveraging brain MRI and wearable sensor data for early detection of neurodegenerative diseases (7 min)
Jie Lian; Hong Kong / Hong Kong SAR China
868
Author Block: J. Lian, V. Vardhanabhuti; Hong Kong/HK
Purpose: Neurodegenerative diseases, including Parkinson's disease (PD) and Alzheimer's disease (AD), pose a significant healthcare
burden to the aging population. Timely detection of these diseases, even before symptom onset, is crucial for early intervention.
Although prior studies have shown that wearable movement-tracking data holds promise as a PD biomarker, the potential of
combining it with brain MRI for improved prediction remains unexplored. This study aimed to investigate the role of brain MRI as a
biomarker for predicting neurodegenerative diseases, alongside accelerometry data, using the UK Biobank dataset.
Methods or Background: This study comprises 19,793 participants with brain MRI scans (T1 and T2), accelerometry data, polygenic
risk scores (PRS), and questionnaire-derived lifestyle information. In total, 56 participants were classified as positive cases if
diagnosed with any of AD, other forms of dementia, or PD at least one year after the initial assessment. We preprocessed the brain
MRI data by segmenting it into 144 segments (normalized by intracranial volume). Disease incidence prediction was employed using
XGBoost models, with results reported using the Area Under the Receiver Operator Characteristic Curve (AUROC).
Results or Findings: The comprehensive model, which incorporated all modalities, achieved an AUROC of 0.760 on the test dataset.
Among the top 20 predictive features, 17 were related to MRI data, pertaining to different regions in the basal ganglia (e.g., ventral
striatum, amygdala), hippocampus, and white matter hyperintensities. In contrast, an ablation study with a model excluding MRI data
only achieved an AUROC of 0.645. Other important features include average activity acceleration between 5-7 p.m., and duration of
sleep.
Conclusion: Our findings suggest the potential of brain MRI in conjunction with activity tracking data as predictive biomarkers for
early neurodegenerative disease detection.
Limitations: Positive cases were not large enough, causing imbalanced dataset.
Ethics committee - additional information: The study was approved by the author’s institution's local ethics board (UW-20814) at
the University of Hong Kong. The population cohort in this study was from the UK Biobank 20 [Application Number 78730] which
received ethical approval from the North West Multicentre Research Ethics Committee (REC reference: 11/NW/03820). All participants
gave written informed consent before enrollment.
Synthesis of artificial T1w full-dose images using low doses of gadolinium-based contrast agents: a new deep neural
network approach of contrast signal mapping (7 min)
Robert Haase; Bonn / Germany
Author Block: R. Haase, T. Pinetz, E. Kobler, Z. Bendella, C. Gronemann, D. Paech, A. Effland, K. Deike, A. Radbruch; Bonn/DE
Purpose: The main objective of this study was to test a new DNN approach to synthesize artificial T1w-full-dose images from
corresponding non-contrast and low-dose images and compare its performance with two reimplemented state-of-the-art approaches
(referred to as setting-A and B).
Methods or Background: Two hundred and thirteen participants received an MRI brain scan with an adapted imaging protocol
including a T1w-low-dose after 20% of the standard dose of a gadolinium-based contrast agent. Fifty participants were randomly
chosen as test set. The new approach is referred to as setting-C. Synthesized artificial T1w-full-dose images were assessed using a
reader-based study. Two readers scored the overall image quality, the interchangeability with the true T1w-full-dose in regard to the
clinical conclusion, the contrast enhancement of lesions and their conformity to the respective true reference lesions.
Results or Findings: The overall image quality was rated lower in setting-A than in the two remaining settings, whose image quality
did not differ from each other and the true full-dose images. The average counts of false positives per case were 0.33±0.93,
0.07±0.33, and 0.05±0.22 for the settings A-C, respectively. The proportion of scans scored as fully or mostly interchangeable was
significantly higher in setting-C (70%) than in settings A (40%) and B (57%). The contrast enhancement was significantly reduced in
all settings compared to the original T1w-full-dose. Using a five-point Likert-scale, there was no significant difference between the
contrast signal reduction of setting-A (-1.10±0.98) and setting-B (-0.91±0.67), but between both settings and setting-C (-0.50±0.55).
The average scores of conformity were 1.75±1.07, 2.19±1.04, and 2.48±0.91 for settings A-C, respectively.
Conclusion: The new approach showed a significantly better qualitative performance than published alternatives. Nevertheless, a
relevant proportion of cases with inadequate synthesis of the contrast signal remains using a low-dose of 20% of the standard dose.
Limitations: Monocentric study.
Funding for this study: R.H. is funded by a research grant (BONFOR; O-194.0002.1) of the Medical Faculty of the University of
Bonn. A.E. and T.P. are funded by the German Research Foundation under Germany's Excellence Strategy (EXC-2047/1, 390685813
and EXC-2151, 390873048).
Ethics committee - additional information: This study was approved by the Ethics Committee for Clinical Trials on Humans and
Epidemiological Research with Personal Data of the Faculty of Medicine of the Rheinische Friedrich-Wilhelms University Bonn:
reference no. 450/20.
Brain age fingerprinting from MR image using multi-level information fusion networks and its application in cognitive
impairment patient screening (7 min)
Feng Shi; Shanghai / China
869
Author Block: N. Zhao, Y. Pan, Z. Xue, F. Gao, F. Shi, D. Shen; Shanghai/CN
Purpose: The aim of this study was to develop a model for estimating brain age from MR image on a large-scale normal aging
population covering entire lifespan, and to assess its potential for early screening of cognitive impairment patients.
Methods or Background: We proposed a novel approach to build a brain age prediction model in lifespan datasets using T1-
weighted MR images. This approach consists of extracting three-level hierarchical information through neural networks and fusing
them with the cross-attention mechanism, to capture inherent brain age fingerprinting in MR images. Specifically, the hierarchical
information included: (1) brain volumes and ratios of 106 parcellations derived from a pre-trained segmentation model, (2) 2D image
slices selected from specific brain regions, which potentially contain brain lesion information such as white matter hypointensities,
lacunes, and perivascular spaces, (3) 3D CNN features with input of MR image.
Results or Findings: This study included 3,711 subjects aged 6-96 years from in-house datasets, with 3,372 cognitively normal
(CN), 207 late MCI (LMCI), and 132 AD. Based on the proposed model, CN subjects achieved a mean absolute error of 2.72 years.
Furthermore, when applying this model to cognitively impaired subjects, AD group had higher brain age gap (BAG) compared to both
LMCI and CN groups (4.43 vs. 2.47 vs. -0.5 years; P < .001). Finally, combing BAG with learned age-related features as inputs of multi-
layer perceptron for differentiating between CN, LMCI, and AD yielded predictive accuracy of 91% for CN vs. LMCI, 91% for CN vs. AD,
and 96% for LMCI vs. AD.
Conclusion: The BAG from prediction model appears to be highly correlated with cognitive impairment and could be used for
screening of cognitive impairment patients.
Limitations: The model utilises 3D high-resolution images while the extension to clinically low-resolution MRI scans should be
studied.
Funding for this study: This work was supported in part by National Natural Science Foundation of China (62131015), Science and
Technology Commission of Shanghai Municipality (STCSM) (21010502600), Key R&D Program of Guangdong Province, China
(2021B0101420006), STI2030-Major Projects (2022ZD021 3100), The China Postdoctoral Science Foundation (Nos. BX2021333,
2021M703340), and National Key Research and Development Program of China (2022YFE02 05700). Data collection and sharing for
this project was funded by Shanghai Zhangjiang National Innovation Demonstration Zone Special Funds for Major Projects (ZJ2018-
ZD-012), Shanghai Pilot Program for Basic Research (JCYJ- SHFY-2022-014), and Shanghai Pujiang Program (21PJ1421400).
Ethics committee - additional information: This study was approved by Autism Brain Imaging Data Exchange (ABIDE), Attention
Deficit Hyperactivity Disorder (ADHD-200), Alzheimer’s Disease Neuroimaging Initiative, Open Access Series of Imaging Studies
(OASIS), Consortium for reliability and reproducibility (CoRR), Consortium of Chinese Brain Molecular and Functional Mapping,
HUASHAN Hospital, and RENJI Hospital, China, approved this study.
A fully automated approach for contrast-agent-free myocardial tissue characterisation using T1-rho mapping (7 min)
Victor De Villedon De Naide; Pessac / France
Author Block: V. De Villedon De Naide1, K. Narceau1, N. Brillet1, V. Nogues1, J. H. Zhang2, M. Villegas-Martinez1, M. Stuber2, H.
Cochet1, A. Bustin1; 1Pessac/FR, 2Lausanne/CH
Purpose: Myocardial T1-rho mapping is a promising biomarker, allowing for contrast-agent-free myocardial injury detection and
quantification. However, its operator-dependant processing induces operator variability and radiologist workload rise. The aim of this
study was to explore the feasibility of artificial intelligence-driven analysis for efficient and automated myocardial T1-rho mapping.
Methods or Background: A cohort of 573 patients presenting various cardiomyopathies was divided into a training set (n=500) and
a distinct test set (n=73). CMR imaging was conducted using a 1.5T Siemens Aera scanner. For each patient, pre-contrast breath-held
3-slice T1-rho mapping was performed, and contrast-enhanced LGE images were acquired in short-axis 12min post-injection of
gadolinium. For each patient, a transformer-based model automatically segmented the left ventricular wall on T1-rho images. Then,
the right-ventricle insertion points were detected using a U-Net. A 16-segment AHA model was then created for segmental T1-rho
values analysis. Segmentation quality was assessed. T1-rho values were quantitatively retrieved for both manual and automated
processing across patient, slice and segment levels. The concordance between methods was gauged. Processing times were
measured.
Results or Findings: Automated processing of the T1-rho slices revealed a reduced processing time (~3 s vs. 1 min 51 s±22 s) in
comparison to manual processing. Automated segmentation quality yielded favourable results (global DICE of 81.9±9.0%). Analysis of
T1-rho values indicated no difference between manual and automated processing (54.9±4.6 ms vs. 55.4±5.1 ms, P=0.098). Strong
correlations (ICC>0.8) were found with minimal biases at patient and slice levels, while agreement was lower at the segment level. All
AHA segments did not differ significantly between manual and automated T1-rho measurements.
Conclusion: Automated processing of myocardial T1-rho maps shows strong agreement with manual processing and comparable
segmentation quality with enhanced time efficiency, highlighting its promising clinical application.
Limitations: Further clinical studies in various cardiomyopathies are warranted.
Funding for this study: This project was supported by funding from the French National Research Agency under grant agreements
Equipex MUSIC ANR-11-EQPX-0030, ANR-22-CPJ2-0009-01, ANR-21-CE17-0034-01, Programme d’Investissements d’Avenir ANR-10-
IAHU04-LIRYC. This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020
research and innovation programme (grant agreement No101076351).
870
Predicting Alzheimer’s progression from mild cognitive impairment: a generalised approach integrating automated
MRI segmentation and longitudinal atrophy analysis (7 min)
Tobias Lindig; Tübingen / Germany
Author Block: J. Steiglechner, B. Bender, G. Lohmann, K. Scheffler, U. Ernemann, T. Lindig; Tübingen/DE

Purpose: This study sought to formulate a generalised classifier for differentiating progressive mild cognitive impairment (pMCI) from
stable MCI (sMCI) by integrating brain MRI segmentation and longitudinal atrophy analysis, essential for early Alzheimer's disease
(AD) intervention.
Methods or Background: Utilising a segmentation model [AIRAmed] on 3D T1w high-field MRI, which labels 30 anatomical regions,
we generated z-statistics through age- and sex-adjusted comparisons with a reference cohort concerning total intracranial volume
(TIV) measures. Atrophy rates were quantified using a time-scaled approach. Z-statistics and atrophy rates were inputted into a
classifier trained via an 80:20 train-test partition of the Alzheimer’s disease national initiative (ADNI) using logistic regression to
distinguish pMCI from sMCI.
Results or Findings: Model generalisation was evaluated on four cohorts (retained test set of ADNI (53 pMCI/93 sMCI), Australian
Imaging Biomarkers and Lifestyle Flagship Study of Ageing [AIBL] (15/42), Minimal Interval Resonance Imaging in Alzheimer’s Disease
[MIRIAD] (4/8), and Open Access Series of Imaging Studies [OASIS] (27/80)) achieving ROC AUCs of 0.86, 0.78, 0.90, and 0.85
respectively. Including atrophy rates improved overall performance from AUC 0.826 to 0.847, enhancing specificity from 70.4% to
75.8%.
Conclusion: This research underscores the efficacy of leveraging both static (volumetric measures) and dynamic (atrophy rate)
features to predict MCI to AD progression. The novel biomarker introduced, combining MRI segmentation and longitudinal atrophy
evaluation, not only presents high sensitivity and specificity, as validated by AUC ROC across various datasets but also emerges as a
robust tool applicable in future research and diagnostic assistance.
Limitations: Limitations are not applicable for this study.
Ethics committee - additional information: This study used anonymised data.
Towards precision diagnosis: machine learning in identifying malignant orbital tumours with multiparametric 3 Tesla
MRI (7 min)
Emma O'Shaughnessy; Paris / France
Author Block: E. O'Shaughnessy; Paris/FR

Purpose: Orbital tumours pose a diagnostic challenge due to their diverse locations and histopathological variations. In recent years,
advancements in imaging have enhanced diagnosis, but classification remains challenging. The application of artificial intelligence in
radiology and ophthalmology has shown promising results. The primary aim of this study was to develop and assess the performance
of machine learning in accurately identifying malignant orbital tumors based on multiparametric 3 Tesla MRI data.
Methods or Background: This single-center prospective study enrolled patients with orbital masses who underwent pre-surgery 3
Tesla MRI scans between December 2015 and May 2021. We employed a Random Forest model with nested stratified cross-
validation, considering various combinations of explanatory features. SHAP (SHapley Additive exPlanations) values have been used to
evaluate feature contributions, and multiple metrics assessed model performance.
Results or Findings: We analysed 113 patients (50.4% female, 49.6% male) with an average age of 51.5 [19-88]. Among the eight
machine learning models evaluated, the one that incorporated all 46 explanatory features (morphology, DWI, DCE, and IVIM) achieved
an AUC of 0.9 [0.73-0.99], while the "10-features signature" model attained an AUC of 0.88 [0.71-0.99]. The ten most influential
features for the Random Forest model included three quantitative IVIM features, four quantitative DCE features, one quantitative DWI
feature, one qualitative DWI feature, and age.
Conclusion: This study suggests that machine learning, when combining multiparametric MRI data, including DCE, DWI, and IVIM,
can yield highly effective models for classifying orbital tumors. The "10-features signature" model may be preferred due to its strong
performance, simplicity, and adherence to the parsimony principle.
Limitations: The study's limitations encompass its single-center design and the relatively small number of subjects.
Funding for this study: This study was publicly funded.
Ethics committee - additional information: This study was approved by an Institutional Research Ethics Board and adhering to
Declaration of Helsinki.
Improving whole body MRI in multiple myeloma: reduced acquisition time with a deep learning reconstruction for
diffusion-weighted imaging at 3 Tesla-preliminary results (7 min)
871
Author Block: J. Herrmann, S. Gassenmaier, S. Ursprung, H. Almansour, S. Werner, S. Afat; Tübingen/DE
Purpose: The evaluation of bone disease in multiple myeloma (MM) is an important topic in oncologic imaging. The objective was to
determine the impact of a Deep Learning (DL)-reconstruction for whole body (WB) diffusion-weighted-imaging (DWI) for staging MM
patients at 3 Tesla compared to standard DWI on reducing acquisition time and improving image quality.
Methods or Background: Thirty patients (mean age, 61±11 years; range, 35–82; 16 men, 14 women) were consecutively included
in this retrospective, monocentric study between February and August 2023. Inclusion criteria were standard DWI (DWI_S) in clinically
indicated MRI at 3 Tesla, and DL-reconstructed WB-DWI (DWI_DL). All patients were examined using the institution's standard MRI
protocol for staging MM including DWI with two different b-values (0 s/mm² and 800 s/mm²) and calculation of apparent diffusion
coefficient (ADC) maps. Image quality was qualitatively assessed by two radiologists using a visual 5-point Likert scale (5=best).
Results or Findings: The overall image quality was evaluated to be significantly superior in DWI_DL as compared to DWI_S for b=0
s/mm², b=800 s/mm², and ADC maps by all readers (p<0.05). The extent of noise was evaluated to be significantly less in DWI_DL as
compared to DWI_S for b=0 s/mm², b=800 s/mm², and ADC maps by all readers (p<0.001). No significant differences were found
regarding artifacts, lesion detectability, sharpness of organs, and diagnostic confidence (p>0.05). Acquisition time for DWI_S was 7:45
min and simulated acquisition time for DWI_DL was 5:03 min.
Conclusion: DWI_DL enhances image quality for WB-MRI in staging of multiple myeloma patients at 3 Tesla. Simulation results
suggest a potential reduction in acquisition time of 35 %, highlighting the promise of DL in advancing clinical efficiency.
Limitations: Priliminary results of this study with a small sample size and retrospective design.
Ethics committee - additional information: This study was approved by the University of Tuebingen
Using deep learning muscle segmentation to optimise MRI outcome measures in Charcot-Marie-Tooth disease (7 min)
Hamza A Salhab; London / United Kingdom
Author Block: H. A. Salhab1, C. Doherty1, R. Zuccarino2, A. McDowell1, T. A. Yousry1, J. Thornton1, J. M. Morrow1, M. M. Shy2, M. Reilly1;
1
London/UK, 2Iowa, IA/US
Purpose: Several studies have demonstrated the use of magnetic resonance imaging (MRI) to measure quantitative outcomes in
neuromuscular disease, however, multiple approaches can be employed for this purpose. This study aims to evaluate and compare
the responsiveness of different segmentation and data analysis approaches in detecting cross-sectional and one year longitudinal
changes of muscle fat fraction (FF) and cross-sectional area (CSA) in charcot-marie-tooth patients and healthy controls.
Methods or Background: A total of 60 Charcot-Marie-Tooth patients (CMT1X,2A,1B)and 30 matched controls were included in this
study. We compared the effect of (1) manual segmentation, (2) neural network-generated segmentation, and (3) manual correction of
automated segmentation on the standardised response means (SRMs) and responsiveness of measuring outcome biomarkers. We
also compared values obtained from 2D and 3D Dixon sequences. Finally, we analysed the value of segmenting multiple consecutive
slices versus a single slice as well as muscle compartments versus the entire calf.
Results or Findings: Our findings suggest that automated segmentation provides SRMs that are comparable to manual
segmentation with or without correction. The sequence acquisition does not significantly affect the SRMs. Analysing multiple
consecutive slices does not significantly improve responsiveness compared to a single slice. Measuring longitudinal biomarker
changes in individual compartments is less responsive than the entire calf.
Conclusion: Some segmentation approaches may unnecessarily prolong and complicate the ROI drawing process without
significantly enhancing the responsiveness of outcome measures
Limitations: Our study includes CMT diseased patients and healthy controls only; Keeping in mind that neuromuscular diseases can
cause different muscle fatty infiltration and atrophy patterns, these results may not be generalizable to other conditions. Moreover, it
is important to note that our analysis is based on Musclesense outputs, consequently, alternative segmentation algorithms are
expected to yield different outcomes
Funding for this study: This study was funded by the Muscular Dystrophy Association (USA).
Ethics committee - additional information: This study was ethically approved by Regional Ethic committee, reference
09/H0716/61.
Prediction of therapy response of breast cancer patients with machine-learning based on clinical data and imaging
data derived from breast [18F]FDG-PET/MRI (7 min)
Kai Jannusch; Düsseldorf / Germany
872
Author Block: K. Jannusch1, H. A. Peters1, N-M. Bruckmann1, J. Morawitz1, F. Dietzel1, L. Umutlu2, G. Antoch1, J. Kirchner1, J. Caspers1;
1
Düsseldorf/DE, 2Essen/DE
Purpose: The objective of this study was to evaluate if a machine-learning prediction model based on clinical- and imaging features
derived from baseline breast [18F]FDG-PET/MRI staging can predict pathologic complete response (pCR) in patients with newly
diagnosed breast cancer prior to neoadjuvant systemic therapy (NAST).
Methods or Background: This study retrospectively enrolled 143 women with newly diagnosed breast cancer. All women
underwent a breast [18F]FDG-PET/MRI, histopathological workup of their breast cancer lesions, and evaluation of clinical data. Fifty-six
features derived from PET, MRI, sociodemographic/anthropometric, histopathologic, and clinical data were generated and used for
input into an extreme-Gradient-Boosting model (XGBoost) to predict pCR. The model was evaluated in a five-fold nested-cross-
validation and reduced the risk of overoptimistic estimations. Diagnostic model-performance was assessed by determining ROC-AUC,
sensitivity, specificity, PPV, and NPV. Feature importances of XGboost were evaluated to assess which features contributed most to
distinguish between pCR and non-pCR.
Results or Findings: Nested-cross-validation yielded a mean ROC-AUC of 80.4±6.0% for prediction of pCR. Mean sensitivity, -
specificity, -PPV and -NPV of 54.5±21.3%, 83.6±4.2%, 63.6±8.5% and 77.6±8.1% could be achieved. Histopathological data were the
most important features for classification of the XGBoost-model followed by PET-, MRI-, and sociodemographic/anthropometric
features.
Conclusion: The evaluated multi-source XGBoost model shows promising results for reliably defining pCR in breast cancer patients
prior to NAST. However, the yielded performance is yet insufficient for the algorithm to be implemented in the clinical decision-making
process.
Limitations: Inhomogeneity, especially with regard to the histopathological characteristics or NAST therapy regimes might be one
limitation. Nonetheless, the risk of inhomogeneity is consistent with the clinical reality of breast cancer patients and underlines the
need for a large number of patients to be included in a machine-learning based approach. Thus, as many combinations as possible
can be learned by the model.
Funding for this study: The study has been funded by Deutsche Forschungsgemeinschaft (DFG), the German Research Foundation
(DFG (BU3075/2‑1 and KI2434/1-2).
Ethics committee - additional information: All procedures performed were in accordance with the ethical standards of the
institutional research committee and with the principles of the 1964 Declaration of Helsinki and its later amendments. University
Duisburg-Essen (study number 17-7396-B0) and University Düsseldorf (study number 6040R).
Predicting Ki-67 proliferation index of meningiomas on MRI based on multi-modal information: a deep learning method
to facilitate the therapeutic decisions (7 min)
Chaoyue Chen; Chengdu / China
Author Block: C. Chen; Chengdu/CN

Purpose: A high Ki-67 index usually suggests accelerated cell proliferation of meningioma related to significant tumor growth as well
as increased recurrent risk. This study aimed to explore the feasibility of deep learning method in predicting Ki-67 index of
meningiomas with multi-modal information.
Methods or Background: Pre-treatment magnetic resonance images were retrospectively curated from 521 patients with surgically
resected, pathologically confirmed meningiomas from three institutions. The cases were classified into low-expressed or high-
expressed groups with a threshold of 5% of Ki-67 index. Predictive models were developed with multi-modal deep learning network by
mixing traditional radiological findings, radiomics features, and images. The performance of the models was evaluated with area
under curve (AUC), accuracy (ACC), sensitivity, and specificity. In addition, 127 cases with incidental small meningioma were recruited
and followed up in 2 years, to investigate if the model could be used for predicting the tumour growth to assist in patient
management.
Results or Findings: Overall, 371 patients were enrolled for model development and primary analysis. The mixing predictive model
showed superior performance to the single modality models, with AUC of 0.798, ACC of 0.710, sensitivity of 0.613, and specificity of
0.806 in the internal test. It also achieved robustness in the external test cohort consisted of 150 cases, with AUC of 0.758, ACC of
0.661, sensitivity of 0.677, and specificity of 0.645. Moreover, model-predicted high Ki-67 tumor was associated with significant tumor
volume growth happened in two years.
Conclusion: The predictive model can efficiently predict the Ki-67 index in meningioma patients, and showed good potential in
facilitating the therapeutic decisions.
Limitations: The limitations of the study are inherent selection bias. Moreover, only contrast-enhanced images were used, while
other sequences were excluded in modeling.
Funding for this study: This work was funded by the China Postdoctoral Science Foundation, Postdoctoral Research Foundation of
China (2022M712245).
Ethics committee - additional information: The Institutional Review Board of West China Hospital, Sichuan University approved
this multi-center, retrospective study, and waived the requirement of informed consent (ID: 2021_S_851).
Trading speed for certainty: artefacts and non-inferiority in AI-accelerated FLAIR imaging of the brain (7 min)
873
Matthias Anthony Mutke; Basle / Switzerland
Author Block: M. A. Mutke, T. Rusche, A. Lonak, K. Blackham, M-N. Psychogios, J. M. Lieb; Basle/CH
Purpose: AI-enhanced MRI sequences promise improved image resolution at reduced acquisition times but are often proprietary
products, with limited evaluation. The goal of this study was to investigate whether an AI-enhanced FLAIR sequence of the brain is
non-inferior to a standard sequence, as assessed qualitatively by experienced human radiologists at both general and individual
levels.
Methods or Background: Patients underwent both standard FLAIR-sequence and an AI-enhanced FLAIR product sequence with
reduced acquisition time of 40% (deep-resolve-boost, Siemens Healthineers), post contrast injection on 1.5T and 3T MRIs. Two
experienced neuroradiologists conducted a side-by-side comparison. Using a five-point Likert scale (ranging from non-diagnostic to
excellent), images were assessed for signal-to-noise ratio, anatomic clarity, overall quality, imaging artefacts, and diagnostic
confidence. Potential lesions (pseudolesions) unique to one sequence and not readily identifiable as artifacts were noted and
compared (McNemar's test). A mixed-model analysis determined the sequence type's effect on ratings, adjusting for rater and patient
variances. Non-inferiority was tested (lower bound of the confidence interval not exceeding -0.5 points).
Results or Findings: Thirty patients were assessed for cerebral metastases (n=11), glioma (n=9), meningioma (n=5) and other
pathologies (n=5). In the mixed model, AI sequences significantly outperformed by 0.4 points (Likert scale) in anatomic clarity but
underperformed by -0.5 points for artifacts, failing to meet non-inferiority. In all other categories, the AI sequence matched the
standard. Both readers identified pseudolesions on a total of 5/30 patients.
Conclusion: The faster, AI-enhanced FLAIR sequence is non-inferior to the standard except for an increase in imaging artifacts. These
artefacts can result in pseudolesions, potentially causing unnecessary follow-up imaging. We emphasize the need for critical and
rigorous evaluation of new sequences and suggest a training phase for radiologists.
Limitations: Small sample size, single-centre with qualitative, subjective evaluation limit this study.
Ethics committee - additional information: The imaging was part of a quality control program, informed consent was waived by
the local ethics committee.
874
EFRS - Radiography: meet the editors

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderators:

Journal performance and why engage (12 min)

1. To outline the benefits of engaging with academic journals.

2. To highlight the performance of the Radiography Journal.
3. To discuss methods of engagement and strategies for improving journal performance.
Addressing equality, diversity, and inclusivity (12 min)

Amy Hancock; Exeter / United Kingdom
1. To outline equality, diversity, and inclusivity issues within research publishing.

2. To discuss methods for improving EDI within publishing.
3. To highlight best equality, diversity, and inclusivity practices regarding research publishing.
Fostering clinical engagement (12 min)

1. To discuss reasons for fostering clinical engagement in research.

2. To emphasise successful strategies for fostering clinical engagement.
3. To discuss the interface between academic and clinical practice in research publications.
Getting published: my ten top tips (12 min)

Robert Meertens; Exeter / United Kingdom
1. To discuss reasons why a manuscript may be rejected.

2. To emphasise how common manuscript errors can be avoided.
3. To highlight the author's responsibilities in promoting their article.
Promoting your research through social media: my ten top tips (12 min)
Nejc Mekis; Ljubljana / Slovenia
1. To discuss opportunities for promoting research.

2. To highlight the potential of social media in journal article promotion.
3. To know how best to promote peer-reviewed journal articles on social media.
875
Becoming a reviewer: my ten top tips (12 min)

Theophilus N Akudjedu; Bournemouth / United Kingdom
1. To identify the purpose of peer review and to explain the steps within the peer review process.
2. To understand the qualities of a good review.
3. To know how best to provide reviewer feedback.
Panel discussion: Ask the editors (16 min)
876
E³ 2021 - Pitfall and mimickers in gynaecological cancer imaging
Categories: Abdominal Viscera, Genitourinary, Imaging Methods, Oncologic Imaging

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Endometrial and cervical cancer (45 min)

1. To describe the imaging findings of endometrial and cervical cancer.

2. To learn about FIGO classification and its implications.
3. To recognise imaging pitfalls and mimickers in endometrial and cervical cancers.
Ovarian cancer (45 min)

Teresa Margarida Cunha; Lisboa / Portugal
1. To become familiar with the histological subtype of ovarian cancer.

2. To become familiar with mimickers of ovarian cancer.
3. To learn about the role of imaging in recognising "difficult to resect" disease.
877
PA 20 - Next-generation patient care
Categories: Artificial Intelligence & Machine Learning, Education, Interventional Radiology, Multidisciplinary, Oncologic Imaging
Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderators:

Responsibility and opportunity (10 min)

1. To learn why high-speed radiology benefits only when the patient can follow and is not left behind.
2. To appreciate tools to optimise for good and to cultivate the madness of moment of diagnoses within a limited time.
3. To understand why patient education and relation are important for AI developments for patients and radiologists (anxieties, fast
track one-stop, impersonal, misinterpretation, address all groups of patients).
How to get imaging information and basic knowledge for shared decision-making to the patient (10 min)
1. To learn what imaging technologies are used for target groups.

2. To appreciate that a single source offering information in the European context is inefficient.
3. To understand the role of ESR and the national societies in making it happen.
When the radiologist-patient dialogue is established (10 min)

Dominique-Gérard Carrie; Balma / France
1. The radiologist's view on how patients should prepare for consultation.

2. To appreciate diversity in patients and radiologists.
3. To understand what patients expect from radiologists.
The challenge of interventional radiology (10 min)

1. To learn about the expectations of patients and radiologists in interventional radiology.

2. To appreciate the dialogue before, during, and after interventional radiology.
3. To understand how to transmit best and communicate interventional radiology results.
Better ways to communicate: the views of patients and radiographers (20 min)
878
1. To learn about the expectations of patients and radiographers.
2. To appreciate communication tools and recommendations.
3. To understand the important role of radiographers in the radiological process.
Presentation of the posters (5 min)
Panel discussion: How to implement improvements that work for all stakeholders? All speakers and (20 min)
879
E³ 24D - Heart and brain

ETC Level: LEVEL II
Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

PET MR in cardiac imaging (15 min)

Patrick Krumm; Tübingen / Germany
1. To get acquainted with the key indications of PET/MRI in myocardial disease.

2. To understand the appropriate examination protocols.
3. To discuss representative clinical cases and appreciate where synergistic PET/MRI protocols contribute.
Propelling new dementia therapies: PET/MR leading the way? (15 min)
Alexander Hammers; London / United Kingdom
1. To know about the clinical scenario of dementia.

2. To carve out the central indications where hybrid imaging contributes to optimised therapy guidance.
3. To discuss novel radiotracers with relevance to dementia.
Brain tumour management with hybrid imaging (15 min)

Nathalie L. Albert; München / Germany
1. To understand the patient journey in patients with brain tumours.

2. To discuss the indications in which hybrid imaging contributes decisive diagnostic information.
3. To evaluate the potential of PET/MRI in brain tumour patients.
Panel discussion: Finally there? Key indications of PET/MRI in heart and brain (10 min)
880
EU 20 - Radiological exposure in newborns and pregnancy
Categories: EuroSafe Imaging/Radiation Protection, Paediatric

ETC Level: LEVEL II
Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Claudio Granata; Trieste / Italy

Good practice in radiation protection of newborns (20 min)

Sergio Salerno; Palermo / Italy
1. To appreciate that newborns may require radiological examinations, especially if they are premature, and that due to risk for short-
and long-term morbidities, neonatal intensive care units (NICU) typically carry out a large number of different radiological
examinations.
2. To understand that different clinical situation may be resolved by ultrasound, reducing the number of routine chest and abdominal
x-rays in newborns.
3. To learn about the clinical settings in which x-ray examinations can be substituted with ultrasound or in which ultrasound may
integrate clinical information and reduce the number of x-ray examinations, reducing the cumulative dose in these patients.
Imaging in pregnancy: the best investigation, at the right time (20 min)
Joanna Kasznia-Brown; Bristol / United Kingdom
1. To review current scientific evidence and radiation effects on pregnant patients and foetuses.
2. To understand the most common clinical indications for imaging in pregnancy and justification challenges.
3. To learn how to choose and optimise radiological investigations to provide the best and most appropriate imaging for pregnant
patients.
Radiological procedures and dose exposure to the newborn and conceptus (20 min)
Timo De Bondt; Sint-Niklaas / Belgium
1. To appreciate the role of the MPE in estimating dose and risks.

2. To learn how effective doses and risks are calculated.
3. To understand the order of magnitude of risks associated with typical radiation doses.
Panel discussion: How to deal with a woman unaware of pregnancy and undergoing a radiological procedure? (25 min)
881
RC 2010 - Technological advances in musculoskeletal imaging: where are we in 2024

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:
Vasco V Mascarenhas; Lisbon / Portugal

Vasco V Mascarenhas; Lisbon / Portugal
Photon-counting CT: what can we see? (15 min)

Ronald Booij; Rotterdam / Netherlands
1. To name and define the technical differences between an energy-integrating detector (EID) CT and a photon-counting detector
(PCD) CT.
2. To describe and reflect on how PCD-CT provide more imaging and visualisation capabilities for musculoskeletal imaging than EID-
CT.
3. To identify and describe prospects of PCD-CT in musculoskeletal imaging.
AI-based acceleration: what is the limit? (15 min)

Filippo Del Grande; Lugano / Switzerland
1. To review the different technical adjustments to speed MR protocols.

2. To learn basic concepts of the different options of AI-based acceleration.
3. To understand the advantages and challenges of AI-based speed MR protocols.
Going low-field in MSK, is it a good idea? (15 min)

Ricardo Donners; Basel / Switzerland
1. To identify and to list benefits of new-generation low-field MRI in MSK imaging.

2. To summarise and reflect on the current body of evidence for new-generation low-field MRI in MSK imaging.
3. To reflect on the future role of new-generation low-field MRI in MSK radiology.
Panel discussion: Do we still need x-rays? (10 min)
882
Meets 20 - Brave new world: how Singapore is forging forward in a post-pandemic future
Categories: Education, Interventional Radiology, Management/Leadership, Molecular Imaging, Neuro, Nuclear Medicine
Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderators:
Charles Xian-yang Goh; Singapore / Singapore
Molecular imaging and targeted therapies in the era of precision medicine (12 min)
1. To describe how Singapore is integrating molecular imaging studies into clinical practice.
2. To discuss the evolving role of targeted radionuclide therapies in oncology.
3. To highlight the importance of interprofessional collaboration in changing models of care.
Evolving trends in interventional radiology (IR) and challenges in meeting them (12 min)
Lawrence Han Hwee Quek; Singapore / Singapore
1. To highlight evolving global trends in interventional radiology.

2. To discuss how IR services in Singapore map into these global trends.
3. To discuss territorial challenges eroding IR's diversity of services.
CNS features of infection: past outbreaks, future pandemics (12 min)

Tchoyoson Lim; Singapore / Singapore
1. To review past outbreaks of infectious (e.g. Nipah, GBS, SARS) and non-infectious (e.g. buprenorphine embolisation) CNS disease in
Singapore.
2. To understand how basic principles of zoonotic CNS infection contribute to coronavirus disease 2019 pandemic preparation in
radiology departments.
3. To combine past knowledge with future technical and multidisciplinary infection control tools in the flexible response to predicted
future pandemics.
Evolving residency as radiology evolves: from apprentices to entrustable professional activities in Singapore (12 min)
Lim Chee Yeong; Singapore / Singapore
1. To discuss challenges in training residents amidst rapidly changing radiology practice.

2. To describe how a structured competency-based medical education focuses on the development of holistic attributes in residents.
3. To compare synergistic components within a programme of assessment, from certifying summative examinations to work-based
models such as EPAs (entrustable professional activities).
883
RC 2001 - Cholangitis or cholangiocarcinoma: how to distinguish inflammation from a tumour?
Categories: Abdominal Viscera

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:

Primary sclerosing cholangitis (15 min)

1. To be able to list the features of primary sclerosing cholangitis (PSC) on MRI and MRCP.
2. To know the various stricture definitions and their clinical relevance.
3. To describe MRI protocol for surveillance
IgG4 cholangitis (15 min)

Martina Gilda Pezzullo; Brussels / Belgium
1. To be able to list the features of IgG4-SC on MRI and MRCP.

2. To learn about the differential diagnosis of IgG4-SC.
3. To demonstrate how to differentiate between IgG4-SC and Cholangiocarcinoma.
Cholangiocarcinoma (15 min)

1. To become familiar with the spectrum of imaging features of cholangiocarcinoma.

2. To learn about the new WHO classification of cholangiocarcinoma.
3. To describe challenging issues in preoperative staging workup.
Panel discussion: Mistakes I've made and what I learnt from them (10 min)
884
SA 20 - Radiomics application in clinical practice
Categories: Artificial Intelligence & Machine Learning, Education, Imaging Informatics, Oncologic Imaging, Research
Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:

Introductions to radiomics (18 min)

1. To define the concept of radiomics.

2. To understand the basics of radiomics workflow.
3. To discuss the potential of radiomics as well as the challenges to be addressed.
Radiomics for treatment response and outcome prediction (18 min)

Lorenzo Ugga; Naples / Italy
1. To describe radiomics methods pertinent to treatment response and outcome prediction.

2. To summarise current evidence about the role of radiomics for this task.
3. To present clinical implementations and barriers to their adoption.
Nomogram development: integrating radiomics and clinical features for risk stratification (18 min)
Katja Pinker-Domenig; New York / United States
1. To understand building, interpreting, and using nomograms that integrate radiomics and clinical features.
2. To be aware of current use cases.
3. To identify the challenges and limitations for clinical implementation.
Panel discussion: How to accelerate radiomics translation to the clinics? (30 min)
885
RC 2008 - Orbital imaging
Categories: Head and Neck, Oncologic Imaging, Paediatric

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:
Soraya Robinson; Vienna / Austria

Paediatric masses with emphasis on retinoblastoma (15 min)

Pim De Graaf; Amsterdam / Netherlands
1. To review the most common paediatric orbital masses.

2. To present up-to-date imaging for retinoblastoma.
3. To suggest tips and tricks to differentiate retinoblastoma from mimics.
Adult masses with emphasis on melanoma (15 min)

Teresa A G G Guerreiro Gonçalves Ferreira; Leiden / Netherlands
1. To give an overview of adult orbital masses and their imaging appearance.

2. To present advances in melanoma imaging.
3. To review the novel techniques in imaging the orbital masses.
Inflammatory lesions mimicking masses (15 min)

Katharina Erb-Eigner; Berlin / Germany
1. To review the new terminology used for inflammatory and granulomatous conditions of the orbit.
2. To review the pertinent imaging findings of common orbital inflammatory and granulomatous conditions.
3. To discuss how to distinguish mass-like inflammatory conditions from malignant lesions.
Panel discussion: Current state of imaging in orbital lesions (10 min)
886
RC 2011 - Pitfalls in neuroradiology
Categories: Neuro, Vascular

Date: March 2, 2024 | 14:00 - 15:00 CET
CME Credits: 1
Moderator:
Sonja Jankovic; Nis / Serbia

Sonja Jankovic; Nis / Serbia
Perfusion imaging and rainbow scales, pitfalls and alternatives (15 min)
Anouk Van Der Hoorn; Groningen / Netherlands
1. To name and understand common pitfalls of frequently used colour scales

2. To judge colour scales on their appropriateness.
3. To define which (colour) scales are good alternatives.
Postictal imaging: a great mimicker (15 min)

Eftychia Kapsalaki; Larissa / Greece
1. To understand the findings that should be expected post ictally.

2. To understand the hemodynamics that contributes to postictal changes.
3. To use the appropriate sequences to identify and differentiate postictal changes and their mimickers.
Incidental findings: no-touch or clinically relevant? (15 min)

1. To differentiate between no-touch and clinically relevant incidental findings.

2. To explore the prevalence and impact of incidental findings in neuroradiology.
3. To discuss strategies for appropriate management of incidental findings to optimise patient care.
Panel discussion: Tips and strategies to avoid pitfalls in neuroradiology (10 min)
887
MS 20 - Interstitial lung diseases
Categories: Chest, Multidisciplinary

Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:

Diagnosis and management of fibrosing non-IPF ILDs (30 min)

Sujal R. Desai; London / United Kingdom
Peter M George; LONDON / United Kingdom
1. To get familiar with the signs of progressive fibrosis.

2. To learn which non-IPF diseases are prone to progressive fibrosis.
3. To know more about which conditions trigger type and upstage of treatment (anti-inflammatory vs antifibrotic)
The many faces of sarcoidosis: diagnosis and management (30 min)

Marco Idzko; Vienna / Austria
1. To get familiar with typical and atypical imaging patterns of sarcoidosis.

2. To learn about pulmonary findings with prognostic implications in sarcoidosis.
3. To know more about the correlation of imaging patterns, clinical and laboratory findings and management consequences.
Case discussion (25 min)
888
SF 20 - Photon-counting CT
Categories: Contrast Media, EuroSafe Imaging/Radiation Protection, General Radiology, Imaging Methods, Physics in Medical Imaging
Date: March 2, 2024 | 14:00 - 15:30 CET
CME Credits: 1.5
Moderator:
Anders Persson; Linköping / Sweden

Anders Persson; Linköping / Sweden
Photon-counting CT: image quality and radiation dose (20 min)

Erik Tesselaar; Linköping / Sweden
1. To describe the principles and technological advancements of photon-counting CT that distinguish it from conventional CT.
2. To describe the advantages and limitations of photon-counting CT in terms of image quality, including the potential for improved
spatial and contrast resolution, reduced image noise, and enhanced tissue characterisation.
3. To evaluate the impact of photon-counting CT on radiation dose and patient safety, including comparisons with conventional CT
and other imaging modalities, as well as the potential for dose reduction and optimisation.
Optimisation of contrast media (20 min)

Joachim E. Wildberger; Maastricht / Netherlands
1. To understand how iodinated contrast media interacts differently with photon-counting CT compared to conventional CT and to
understand the impact of spectral sensitivity, higher resolution, and reduced noise of photon-counting CT on image quality and
contrast resolution.
2. To optimise iodinated contrast media dosage and how the unique properties of photon-counting CT can allow for the reduction of
contrast media dose without compromising the image quality.
3. To identify and understand how to mitigate enhancement artifacts unique to photon-counting CT to improve homogeneity of
contrast enhancement, leveraging the advantages over conventional CT.
Clinical application (20 min)

1. To understand the clinical benefits of using photon-counting CT over conventional CT in various clinical scenarios; this includes
situations where the superior contrast resolution, reduced radiation dose, and ability to perform spectral imaging of photon-counting
CT are particularly beneficial.
2. To learn how the superior image quality and spectral capabilities can enhance disease detection, characterisation, and assessment
of disease progression.
3. To learn from photon-counting CT examinations where iodinated contrast media were used, such as vascular diseases and cancer.
Panel discussion: ECR 2024: Exploring the advancements and challenges of photon-counting CT technology (25 min)
889
ST 19 - Unveiling the Future: The Impact of EUCAIM's First Platform Release on AI in Cancer Research
Date: March 2, 2024 | 15:00 - 15:30 CET
Join us for an insightful interview as we delve into the transformative impact of EUCAIM's first platform release on the landscape of
cancer research. Our discussion will unravel the significance of this milestone, examining how the platform is set to revolutionize
cancer research.
We will also explore the crucial topic of safeguarding the privacy and security of sensitive imaging data, shedding light on the
measures implemented by EUCAIM to ensure ethical and secure practices.
Don't miss this opportunity to gain exclusive insights into the future of cancer imaging, to learn how you could collaborate and
contribute to this initiative, as well as to learn how EUCAIM is shaping the way forward for researchers, clinicians, and the radiology
community at large.
Moderator:
Interview (30 min)

890
AI-SC 8 - Societal implications of AI in radiology

Date: March 2, 2024 | 15:00 - 16:00 CET
Moderator:


Pablo Valdes Solis; Marbella / Spain
Mónica Cano Cano Abadía; Graz / Austria
Melanie Goisauf; Graz / Austria
1. To discuss the impact on Radiologists' tasks.

2. To appreciate possible bias.
3. To discuss access to trustworthiness through XAI.
891
CUBE 24 - IR strategies in particular situations

Date: March 2, 2024 | 15:30 - 16:00 CET
1. To learn about the techniques available nowadays.

2. To learn about indicaations and application.
3. To get an overview on the outcomes of IR procedures.
Moderator:

Paraganglioma embolization: indications and techniques (14 min)

Pierpaolo Biondetti; Milano / Italy
Percutaneous treatment of neurogenic pain due to discal herniation (14 min)

Carolina Lanza; Milan / Italy
892
RPS 2111 - Neurodegeneration and cognition
Categories: Artificial Intelligence & Machine Learning, Hybrid Imaging, Neuro, Nuclear Medicine, Research
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Relationship between hippocampal subfield volumes and cognitive decline in healthy subjects (7 min)
Simon Doran; Dublin / Ireland
Author Block: S. Doran, R. A. Kenny, J. F. Meaney, C. De Looze; Dublin/IE

Purpose: We examined the relationship between hippocampal subfield volumes and cognitive decline over a four-year period in a
healthy older adult population with the goal of identifying subjects at risk of progressive cognitive impairment which could potentially
guide therapeutic interventions and monitoring.
Methods or Background: 482 subjects (68.1 years +/- 7.4; 52.9% female) from the Irish Longitudinal Study on Ageing underwent
magnetic resonance brain imaging and a series of cognitive tests. Using K-means longitudinal clustering, subjects were first grouped
into three separate global and domain-specific cognitive function trajectories; high-stable, mid-stable and low-declining. Linear mixed
effects models were then used to establish associations between hippocampal subfield volumes and cognitive groups.
Results or Findings: Decline in multiple hippocampal subfields was associated with global cognitive decline, specifically the
presubiculum (estimate -0.20; 95% confidence interval (CI) -0.78 – -0.02; p=.03), subiculum (-0.44; -0.82 – -0.06; p=.02), CA1 (-0.34;
-0.78 – -0.02; p=.04), CA4 (-0.55; -0.93 – -0.17; p=.005), molecular layer (-0.49; -0.87 – -0.11; p=.01), dentate gyrus (-0.57; -0.94 –
-0.19; p=.003), hippocampal tail (-0.53; -0.91 – -0.15; p=.006) and HATA (-0.41; -0.79 – -0.03; p=.04), with smaller volumes for the
Low-Declining cognition group compared to the High-Stable cognition group. In contrast to global cognitive decline, when specifically
assessing the memory domain, cornu ammonis 1 subfield was not found to be associated with low declining cognition (-0.14; -0.37 –
0.10; p=.26).
Conclusion: Previously published data shows that atrophy of specific hippocampal subfields is associated with cognitive decline but
our study confirms the same effect in subjects asymptomatic at time of enrolment. This strengthens the predictive value of
hippocampal subfield atrophy in risk of cognitive decline and may provide a biomarker for monitoring treatment efficacy.
Limitations: T1 imaging can be unreliable for hippocampal segmentation.
Funding for this study: Funding for The Irish Longitudinal Study on Ageing (TILDA) is provided by the Irish Government, the Health
Research Board (HRB), The Atlantic Philanthropies, and the Irish Life Plc.
Ethics committee - additional information: This study was approved by the Trinity College Faculty of Health Sciences Research
Ethics Committee, Dublin, Ireland. Protocols conformed with the Declaration of Helsinki. Signed informed consent was obtained from
all respondents prior to participation. Additional ethics approval was received for the magnetic resonance imaging (MRI) sub-study
from the St James’s Hospital/Adelaide and Meath Hospital, Inc. National Children’s Hospital, Tallaght (SJH/AMNCH) Research Ethic
Committee, Dublin, Ireland. Those attending for MRI also completed an additional MRI-specific consent form. (De Looze et al)
Differential atrophy along the longitudinal axis of the hippocampus in Alzheimer’s disease and suspected non-
Alzheimer’s disease pathophysiology (SNAP) (7 min)
Torcato Meira; Braga / Portugal
893
Author Block: T. Meira, R. Morais-Ribeiro, T. Jesus, M. Dias, A. Coelho, T. G. Oliveira; Braga/PT
Purpose: Cerebrospinal fluid (CSF) biomarkers have been increasingly used to support diagnosis of Alzheimer’s disease (AD).
Suspected non-AD pathophysiology (SNAP) refers to normal CSF levels of amyloid-beta (AB) with increased tau, whereas Alzheimer’s
Disease continuum (ADc) is defined by AB pathology evidence. Since hippocampus studies have highlighted differential properties
along its longitudinal axis, we aim to evaluate how its various subregional volumetric markers differ between ADc, SNAP and controls,
as well as their association with clinical presentation.
Methods or Background: We included 1242 participants from the Alzheimer’s Disease Neuroimaging Initiative. Controls (n=234)
were defined as having normal CSF AB (≥192 pg/ml), total tau (<93 pg/ml) and phosphorylated tau (<23 pg/ml). ADc individuals
(n=784) were abnormal for AB, whereas SNAP subjects (n=224) had normal AB with either abnormal total or phosphorylated tau.
Structural MRI acquisitions were analyzed with a method developed in our laboratory for segmenting the hippocampus in anterior,
intermediate and posterior parts. Controlling for age, sex and multiple comparisons, groups were compared with one-way ANOVA and
Pearson coefficients were calculated to assess correlations between volumetric variables and age, CSF biomarkers or
neuropsychological scores.
Results or Findings: ADc showed lower total and subregional hippocampal volumes than SNAP or controls (P<0.001). When
normalizing the volume of each subregion for its total ipsilateral hippocampus volume, ADc showed increased posterior atrophy
(P<0.001) and higher relative anterior volume (P<0.01) when compared to the other groups. SNAP and controls showed greater
correlations between volumetric measures and age, with attenuated differences over age. Several subregional hippocampal volumes
correlated significantly with altered levels of CSF biomarkers, cognition, neuropsychiatric features and functional impairment.
Conclusion: Our work contributes to better understanding of ADc and SNAP pathophysiology and to predicting their respective
clinical features.
Ethics committee - additional information: As per ADNI protocols, all procedures performed in studies involving human
participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964
Helsinki declaration and its later amendments or comparable ethical standards. More details can be found at adni.loni.usc.edu.
Grey matter atrophy mediates the association between tau pathology and cognition in Alzheimer's disease: a
simultaneous PET/MRI study (7 min)
Xinru Xu; Nanjing / China
Author Block: X. Xu, B. Zhang; Nanjing/CN

Purpose: This study aimed to investigate the area-specific causal pathways between regional tau pathology, grey matter atrophy,
and cognitive impairment in MCI/AD patients.
Methods or Background: Thirty-four MCI/AD patients covering the prodromal to dementia spectrum and 23 cognitive normal
controls experienced standardised neuropsychological assessments followed by 18F-fortaucipir (FTP) PET and 3D T1 magnetic
resonance imaging (MRI). We explored voxel-wise and region of interest- based inter-group differences in grey matter volume (GMV)
and regional tau standardised uptake value ratio (SUVR) in the two groups. Multimodal correlation analyses and mediation analyses
were carried out to assess the relationship between GMV and the SUVR and cognition scores in MCI/AD group.
Results or Findings: The 18F-fortaucipir retention was observed in the fusiform, lateral temporal lobe, supramarginal gyrus,
precuneus and posterior cingulate in MCI/AD groups. Mediation analyses showed the GMV of the medial parietal lobes and medial
temporal lobes mediated the effect of local and distant region tau on cognitive impairment. Besides, in the medial temporal, GMV of
the entorhinal cortex mediated the effect of local region tau on cognitive impairment.
Conclusion: Our finding help to clarify the spatial relationship of tau propagation and gray matter atrophy. Local and distant atrophy
played a important mediating role between tauand cognitive impairment in MCI/AD patients. Besides, temporoparietal region is an
important hub linking gray matter atrophy, tau pathology, and cognitive impairment for AD, and might be a potential treatment target
region for AD.
Limitations: Firstly, the 18F-fortaucipir PET tracer demonstrating a low degree of off-target binding in the choroid plexus may lead to
inaccurate quantification of tau protein in the hippocampus ,Another limitation is the small sample size, longitudinal measurements
utilising a larger sample size will be conducted in future studies.
Funding for this study: This work was supported by Clinical Research Special Funded Project of Nanjing Drum Tower Hospital (No.
2023-LCYJ-MS-11).
Ethics committee - additional information: Northern Theater General Hospital Ethics Committee: No.2021(024).
Compensatory deregulation of attention and executive networks in prodromal Lewy body dementia. A resting-state
functional magnetic resonance imaging study (7 min)
Valeria Onofrj; Rome / Italy
894
Author Block: V. Onofrj1, R. Franciotti2, A. Ferretti2, C. Padulo2, S. Sansi2, I. Rektorova3; 1Brussels/IT, 2Chieti/IT, 3Brno/CZ
Purpose: Early diagnosis of Lewy Body Dementia (MCI-LBD) is challenging due to the lack of specific diagnostic tests and imaging
findings. MCI-LBD subjects show attentional, executive, spatial perception, verbal, spatial memory, and intelligence deficits. These
occur early and imply the dysfunction of selected sensory and associative networks.
Methods or Background: We performed a global analysis of resting-state functional MRI data on 38 MCI-LBD subjects and 24
healthy controls (HC) and extracted the connectivity matrices of regions included in the Cingulo-opercular Network (CON), Fronto-
Parietal Network (FPN), Default Mode Network (DMN), Dorsal Attention Network (DAN), Somato-Motor Network (SMN), Visual Network
(VN) and Language Network (LN). We compared intra- and inter-network connectivity between the two groups of subjects and
correlated neuropsychological test scores with intra- and inter-network connectivity of MCI-LBD subjects.
Results or Findings: The data revealed increased between-networks connectivity (p<0.05) between the DAN and SMN, the CON and
FPN, and the FPN and DMN in MCI-LBD subjects vs. HC. Decreased between-network and intra-network connectivity (p<0.05) was
found between the SMN and DMN, the DMN and DAN, and between the right rostral prefrontal and right anterior insular nodes of the
CON in MCI-LBD subjects vs. HC. Significant correlations (p<0.05) were found among intra-network and between-network FC values
and attention, executive, visuo-perceptual, verbal memory, spatial memory deficit, and intelligence tests.
Conclusion: Overall, we found early demodulation of intra- and inter-network connectivity starting with the internal dissociation of
the CON. Intra- and inter-network connectivity correlates with neuropsychological tests results and may potentially serve as early
imaging biomarkers of LBD.
Limitations: Our study limitation is mainly the limited number of subjects.
Funding for this study: The authors received no financial support for the research, authorship, and publication of this article.
Ethics committee - additional information: This study was approved by the Ethics Committee of Masaryk University and was
performed according to the Declaration of Helsinki (1997) and subsequent revisions. All participants gave written informed consent.
Normative and individual, non-normative intrinsic networks and the transition to impaired cognition (7 min)
Qirui Zhang; Philadelphia / United States
Author Block: Q. Zhang1, S. Hudgins1, A. Struck2, A. A1, S. Javidi1, M. Sperling1, B. Herman2, J. Tracy1; 1Philadelphia, PA/US, 2Madison,
WI/US
Purpose: The relationship between well-defined (normative) brain network and cognition in disease has been extensively studied. In
the present study we focused not only on the relationship between typical network topological properties and cognitive impairment in
temporal lobe epilepsy (TLE) patients , but also on the role that highly individualized functional brain systems (non-normative
networks) play in cognitive impairment.
Methods or Background: This study included 88 TLE and matched 91 healthy controls. FMRI data were decomposed using
independent component analysis to obtain individualized networks. Here, we calculated the degree of match between individualized
networks and canonical networks (e.g., Yeo et.al 17 resting-state network) and divided each participant's networks into normative or
non-normative status based on the degree of match.
Results or Findings: K-means clustering produced two substantive clusters identified as having intact or impaired neurocognitive
profiles. We found that the individualized networks matched the canonical networks less well in the cognitively impaired compared to
the intact TLE patients. The cognitively impaired patients showed significant abnormalities in the profiles of both normative and non-
normative networks, whereas the intact patients showed abnormalities only in non-normative networks. Besides, we found normative
networks held a strong, positive association with the neuropsychological measures, with this association negative in non-normative
networks.
Conclusion: We concluded that explanations of cognitive dysfunction in impaired TLE patients will come from an understanding of
both their normative and individualized, non-normative intrinsic connectivity systems. We were able to provide the initial data
demonstrating that significant cognitive deficits are associated with the status of highly-individual networks, making clear that the
transition from intact to impaired cognitive status is not simply the result of disruption to normative brain networks.
Limitations: The specific functions and contributions of each on-normative network was not sufficiently explored.
Funding for this study: Work was supported by Grant: JIT, NIH/NINDS, R01 NS112816-01
Ethics committee - additional information: This study was approved by the Institutional Review Board for Research with Human
Subjects at Thomas Jefferson University. All participants provided informed consent in writing.
Can AI augmented MRI replace FDG PET CT brain in the evaluation of patients with cognitive impairment (7 min)
Henry Dillon; Dublin / Ireland
895
Author Block: H. Dillon, B. S. Kelly, G. J. Mcneill, R. O'Donohoe, A. Stone, C. Hickie, M. Colombie, J. Kinsella, R. P. Killeen; Dublin/IE
Purpose: Cerebral atrophy and hypometabolism are crucial indicators of neurodegenerative diseases such as Alzheimer's and other
cognitive disorders. Assessing cerebral atrophy on MRI is time consuming and subjective. This study aims to compare cerebral
atrophy detected on AI augmented MRI with patterns of hypometabolism on FDG PET brain, the current gold standard. This software
has the potential to reduce the need for costly FDG PET brain studies as well as increase the number of radiologists who can interpret
neurodegenerative studies.
Methods or Background: There is currently limited availability of FDG PET brain in the diagnosis of neurodegenerative conditions
worldwide. A new commercially available AI software available through Siemens known as RadCompanion utilises artificial intelligence
to detect cerebral atrophy on T1w MRI brain using a simplistic traffic light colour coded system while also providing quantitative
analyses. We compared atrophy detected on this new software with the pattern of hypometabolism on FDG PET brain. The images
were interpreted by a dual specialised neuromolecular radiologist, specialist neuroradiologist and neuroradiology fellow.
Results or Findings: Twenty-three patients were assessed. The patten of cerebral atrophy on MRI was compared to corresponding
FDG PET brain and a pattern of neurodegeneration suggested. In 78% of patients the pattern of neurodegeneration suggested
matched the suggested diagnosis on the corresponding FDG PET brain. In 100% of patients a normal MRI correlated with a normal
FDG PET brain.
Conclusion: Our study demonstrates good correlation between AI augmented MRI and FDG PET Brain. This software could provide a
useful screening test in centres with limited availability to PET CT and perhaps a suitable alternative in centres with no access. We
recommend further evaluation in future using a larger sample size.
Limitations: Limitations included small sample size, single centre and retrospective study.
Ethics committee - additional information: This study as ethically approved by through local hospital audit committee.
Increased brain iron deposition associated with cognitive and motor dysfunction in type 2 diabetes (7 min)
Chaofan Sui; Jinan / China
Author Block: C. Sui1, Y. Gao1, N. Wang1, Y. Wang2, X. Zhang1, N. Zhang1, Q. Zhang3, L. Guo1; 1Jinan/CN, 2Yantai/CN, 3New York, NY/US
Purpose: We aim to explore the relationship between brain iron metabolism and cognitive and motor function in type 2 diabetes
(T2DM) using quantitative susceptibility mapping (QSM).
Methods or Background: Thirty-four diabetic peripheral neuropathy (DPN), fifty-five people with diabetes without DPN (NDPN), and
fifty-one healthy controls (HCs) were recruited and underwent cognitive and motor assessments, and blood biochemical tests. Brain
QSM map was calculated from multi-echo GRE data using morphology-enabled dipole inversion with an automatic uniform
cerebrospinal fluid zero reference algorithm (MEDI+0). One-way ANOVA was applied to evaluate the above indicators among the
three groups, and the susceptibility of altered brain regions was correlated to the clinical features of patients. Moreover, multiple
linear regression analysis was performed to explore which factors may influence the susceptibility in T2DM.
Results or Findings: Susceptibility of the striatum, including the putamen and the caudate nucleus, in T2DM were higher than those
in HCs (p<0.05, FDR correction), and there was no significant difference between the DPN and NDPN groups. Multivariate linear
regression analysis revealed that diabetes and age affected iron deposition in the putamen and the caudate. Notably, susceptibility of
the putamen positively correlated with Timed Up and Go test’ time and negatively correlated with gait speed, Montreal Cognitive
Assessment and Symbol Digit Modalities Test scores in T2DM; its also showed a negative correlation with cognitive parameters and
had nothing to do with motor function in the HC group.
Conclusion: Brain tissue iron, measured by QSM, can track motor function in T2DM. This may be useful to detect signs of early motor
change to monitor disease progression.
Limitations: This research was a cross-sectional study of brain iron changes in T2DM. We should enlarge sample size and further
examine brain iron levels of T2DM longitudinally.
Funding for this study: This work was supported by grants from Shandong Provincial Natural Science Foundation (ZR2020MH288),
the Technology Development Plan of Jinan (201301049, 201602206, 201907052, 202134072), Medical and Health Science and
Technology Development Project of Shandong Province (2016WS0529), and Funding for Study Abroad Program by Shandong Province
(201803059).
Ethics committee - additional information: This study was ethically approved by the Ethical Committee of the Institutional
Review Board (IRB) of Shandong Institute of Medical Imaging (2019-002).
Brain iron deposition and cognitive decline in cerebral small vessel disease patients: a quantitative susceptibility
mapping study (7 min)
Yian Gao; Jinan / China
896
Author Block: Y. Gao1, Q. Zhang2, H. Zhuang2, C. Sui1, N. Zhang1, C. Liang1, L. Guo1, Y. Wang2; 1Jinan/CN, 2New York, NY/US
Purpose: Our study aims to compare brain iron deposition in gray matter (GM) nuclei between cerebral small vessel disease (CSVD)
patients and healthy controls (HCs), exploring factors that affect iron deposition and cognitive function.
Methods or Background: A total of 321 subjects were enrolled in this study. All subjects had cognitive examination including the
Stroop color word test (SCWT) and MRI including multiecho gradient echo (mGRE) sequence. The CSVD patients were divided into
mild to moderate group (CSVD-M, total CSVD score≤1) and severe group (CSVD-S, total CSVD score>1). Morphology-enabled dipole
inversion with an automated uniform cerebrospinal fluid zero reference algorithm (MEDI+0) was used to generate brain QSM maps
from mGRE data. Deep gray regional susceptibility values and cognitive function were compared among three groups (CSVD-S, CSVD-
M, and HC) using multiple linear regression analysis and mediation effect analysis.
Results or Findings: There were significant differences in the SCWT scores and mean susceptibility values of the globus pallidus
(GP), putamen (Put), and caudate nucleus (CN) among the three groups (P<0.05, FDR correction). Age had a significant positive
impact on the susceptibility values of GP (p=0.018), Put (p<0.001), and CN (p<0.001). A history of diabetes had a significant positive
influence on the susceptibility values of Put (p=0.011) and CN (p<0.001). A smoking history had a significant positive association with
the susceptibility values of CN (p=0.019). Mediation effect analysis demonstrated that iron deposition in the neostriatum partially
mediated the relationship between hypertension and cognitive function.
Conclusion: Age, diabetes, and smoking may increase iron deposition in the basal ganglia, leading to cognitive decline. This can help
with understanding CSVD and its prevention and treatment.
Limitations: QSM values were analyzed using region of interest (ROI), which is a vast reduction of imaging information.
Funding for this study: This work was supported by grants from Shandong Provincial Natural Science Foundation (ZR2020MH288),
the Technology Development Plan of Jinan (201301049, 201602206, 201907052, 202134072), Medical and Health Science and
Technology Development Project of Shandong Province (2016WS0529), and Funding for Study Abroad Program by Shandong Province
(201803059).
Ethics committee - additional information: All subjects signed informed consent prior to the study, which was approved by the
Shandong Provincial Hospital Affiliated to Shandong First Medical University Subcommittee on Human Studies Institutional Review
Board.
Brain tissue atrophy and cognitive decline in relation to serum uric acid variance (7 min)

Purpose: This study aimed to evaluate the long-term associations between serum uric acid (SUA) variance and neuroimaging indices
of brain health.
Methods or Background: This cohort study recruited 1,111 participants aged 25-83 years from the subset study of brain MRI
acquisition within the Kailuan study from 2020 onward. The SUA concentrations were measured every two years from 2006 to 2018.
We primarily assessed SUA variance as the average slope incorporating seven measurements and further defined the direction of
changes. The multivariate-adjusted associations of SUA variance with MRI markers of brain tissue volumes and microstructural
integrity and cognitive function were examined using generalised linear models.
Results or Findings: Compared with the stable group, brain white matter (WM) volume decreased irrespective of the increase or
decrease in SUA levels (beta=−0.26, 95% CI −0.40 to 0.12 and beta=−0.15, 95% CI −0.28 to −0.02). Elevated SUA levels were also
associated with smaller cerebral parenchyma volume (beta=−0.01, 95% CI −0.01 to 0). Participants with progressively increased SUA
levels exhibited lower global fractional anisotropy and higher radial diffusivity (beta=−0.27, 95% CI −0.41 to −0.13 and beta=0.23,
95% CI 0.10 to 0.36), which remained significant after further adjustment for white matter and white matter hyperintensity volume.
Elevated SUA concentration was also associated with lower MoCA scores (beta=−0.20, 95% CI −0.33 to −0.06).
Conclusion: The SUA changes were associated with decreased WM volume. Progressively increased SUA levels detrimentally affect
brain health, manifested by smaller brain tissue volume, impaired microstructural integrity, and poorer cognitive performance. Long-
term prevention of SUA fluctuation is essential for protecting brain health and preventing early-stage dementia.
Limitations: The effects of SUA variance in a more recent period on neuroanatomical features were still unknown and warrant future
exploration.
Funding for this study: This study was supported by grants No. 62171297 and 61931013 from the National Natural Science
Foundation of China, No. [2015] 160 from the Beijing Scholars Program, No. ZYLX202101 from Beijing Hospitals Authority Clinical
Medicine Development of Special Funding Support, No. 2021-135 from Beijing Municipal Health Commission-Beijing Key Clinical
Discipline Funding.
Ethics committee - additional information: The META-KLS cohort study was ethically approved by the Medical Ethics Committee
of Kailuan General Hospital (No.2021002). Written informed consent was obtained from all participants. Participants did not receive a
stipend.
White matter abnormalities in patients with mild cognitive impairment and post-acute sequelae of COVID- 19:
preliminary results of a radiomics study (7 min)
Melania Stubos; Trieste / Italy
897
Author Block: M. Stubos, M. Ukmar, G. Pizzamiglio, I. Zorzenon, L. Bottaro, E. Zulian, N. Fiotti, G. Furlanis, M. A. A. Cova; Trieste/IT
Purpose: The purpose of this study was to assess the contribution of radiomics in the analysis of the normal appearing white matter
(NAWM) in patients with cognitive decline related to mild cognitive impairment (MCI) and to post-acute sequelae of COVID-19 (PASC)
in comparison to the normal white matter (NMW) of healthy controls. Furthermore, we compared the NAMW of the patients with MCI
to the one of patients with PASC.
Methods or Background: A retrospective study on ten patients with MCI, ten patients with PASC and 23 healthy controls who
underwent a 3T MRI was performed, and the 3D-FLAIR sequence was selected. Fourteen different brain areas (bilateral frontal lobes,
temporal lobes, parietal lobes, thalami, cerebellar peduncles, lateral ventricles, and genu and splenium of the corpus callosum) were
selected for the ROIs placement: features were extracted on a radiomics software and a statistical analysis was carried out through a
binary logistic regression model.
Results or Findings: In the comparison between the NAWM of MCI patients and the NWM of healthy controls, statistically significant
differences (p<0.05) were found in all the 14 analysed areas, with a good performance in the temporal, frontal and parietal lobes.
Significant differences were also found in the comparison between the NAWM of PASC patients and the NWM of healthy controls, with
a good performance in the frontal lobes, temporal lobes and thalami. No significant differences were found in the comparison between
the NAWM of patients with MCI and PASC.
Conclusion: This study shows that radiomics can be a useful tool in the analysis of the NAMW in patients with cognitive decline
related to MCI and PASC.
Limitations: The low number of patients included in the study and the heterogeneity of the MCI group of patients.
Ethics committee - additional information: The study is retrospective study, hence ethical approval was not sought.
COVID-19 induces grey matter atrophy in patients with cognitive but also with only olfactory disorders (7 min)
Simonetta Gerevini; Cremona / Italy
Author Block: S. Capelli1, A. Caroli1, A. Arrigoni1, A. Napolitano2, G. Pezzetti2, A. Remuzzi3, F. L. Lorini2, M. Sessa2, S. Gerevini2;
1
Ranica/IT, 2Bergamo/IT, 3Dalmine/IT
Purpose: The aim of this study was to evaluate grey matter (GM) structural alterations related to COVID-19 in two separate groups of
patients with the most frequent and distinctive COVID-19-related neurological manifestations - isolated olfactory disorders (COVID-
OD) and cognitive disorders (COVID-CM) – compared to a control group of healthy individuals.
Methods or Background: Sixty-one COVID-CM patients (57[60–63] years, 62% females), 84 COVID-OD patients (49[35–57] years,
60% females) and 17 controls (51[41–52] years, 41% females) were included in the study. To investigate differences between
patients and controls in terms of GM regional volume and voxel-wise density, Region-Based Morphometry (RBM) and Voxel-Based
Morphometry (VBM) were applied to T1-weighted MRI scans. Surface-Based Morphometry (SBM) was used to investigate changes in
cortical thickness.
Total intracranial volume and age were included as nuisance variables in the statistical model assessing group differences.
Results or Findings: The multi-morphometric analysis revealed statistically significant reduction in GM regional volume (RBM) and
density (VBM) as well as lower cortical thickness in both COVID-CM and COVID-OD groups compared to controls. Notably, COVID-CM
patients showed more widespread and severe tissue loss than COVID-OD patients.
The most affected GM regions were hippocampus, putamen, cingulate cortex, praecuneus, amygdala, lingual gyrus, and caudate
nucleus. Most of the atrophic regions are known to be involved in memory processes, in the sense of smell, or both.
Conclusion: Current MRI findings indicate that, with varying degrees of severity, both COVID-19-related olfactory and cognitive
disorders lead to GM atrophy, possibly reflecting neurodegeneration and neuroinflammation. The COVID-CM group showed more
pronounced GM changes, suggesting a stronger inflammatory response.
Limitations: This was a retrospective and monocentric study, with a small control population. Quantitative clinical data assessing the
severity of olfactory or cognitive disorders was not available.
Funding for this study: This study received partial support from Brembo SpA (Curno, Bergamo, Italy) through the "Progetto
TrexUno" initiative.
Ethics committee - additional information: The use of patient data was granted ethical approval by the local ethics committee as
part of a broader observational study protocol (Reg. 118/22). Informed consent was acquired from patients or from their next of kin
(in the case of ICU patients).
Brain age in healthy individuals and across multiple neurological disorders (7 min)
Li Chai; Beijing / China
898
Author Block: L. Chai, Z. Zhuo, Y. Duan, Y. Liu; Beijing/CN
Purpose: The aim of this study was to investigate brain age in healthy individuals and across multiple neurological disorders and its
association with MRI measures and clinical variables.
Methods or Background: MRI and clinical data of 2,913 HC and 331 MS patients, 189 NMOSD patients, 239 AD patients, 244 PD
patients, and 338 cSVD patients were collected. Brain age gap (BAG) was defined as the difference between brain age predicted using
3D T1w with deep learning and chronological age. Brain regions volumes and WMH and clinical measures were compared between HC
with advanced brain aging (BAG ≥ 5 years) and resilient brain agers (BAG ≤ −5 years). Associations between BAG, WMH and clinical
variables were examined in patients.
Results or Findings: Increased BAG was in patients with MS (10.30 ± 12.6 years), NMOSD (2.96 ± 7.8 years), AD (6.50 ± 6.6 years),
PD (4.24 ± 4.8 years), and cSVD (3.24 ± 5.9 years). WMH was higher and regional brain volume was lower in advanced brain agers (p
<0.001) than in resilient brain agers. The specific brain regions associated with increased BAG differed across the various neurological
disorders. Increased BAG was correlated with WMH and cognitive decline in neurological disorders. Increased BAG was correlated with
higher disability scores in MS patients but not in NMOSD patients.
Conclusion: The BAG shows utility as an imaging marker for monitoring cognitive and physical impairment across different
neurological disorders. Advanced brain age was associated with atrophy and WMH, suggesting an increased risk of neurological
problems.
Limitations: First, it was cross-sectional study. Second, deep learning methods are a “black box”, and the interpretability of brain
age predictions needs further improvement. Third, we did not consider other factors that influence brain age. Fourth, the cognitive
and clinical assessments were relatively limited.
Funding for this study: This study was funded by the Beijing Tiantan Hospital, Capital Medical University, No. KY-2019-050-02 and
KY-2019-140-02.
Ethics committee - additional information: This study was ethically approved by the Beijing Tiantan Hospital, Capital Medical
University, No. KY-2019-050-02 and KY-2019-140-02.
899
ESR eHealth 21 - How to critically appraise artificial intelligence (AI)/radiomics papers
Categories: Artificial Intelligence & Machine Learning, Imaging Informatics, Research

ETC Level: LEVEL II
Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Every day, many scientific manuscripts regarding artificial intelligence and radiomics are published. To readers, it is not always clear
how to best approach these papers, especially when many technical details are provided. This session will discuss strategies to
critically appraise such manuscripts, what qualities to look for and how to identify potential issues in the statistical analyses.
Currently, available reporting guidelines will be highlighted to provide readers, as well as potential authors and reviewers, with help
objectively assessing a manuscript's reporting quality.
Moderator:

On metrics and methods: pearls and pitfalls in statistical analyses (15 min)
Nathaniel Mercaldo; Boston / United States
Read avidly but think critically: must-have qualities to look for in radiomics research (15 min)
AI reporting guidelines: how to focus on quality and reproducibility (15 min)

Panel discussion: What can we learn from the literature so far? Where do we go next? (10 min)
900
RC 2112 - Pros and Cons: there is always a place for intervention in paediatric stroke
Categories: Interventional Oncologic Radiology, Neuro, Paediatric, Vascular

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:
Carlo Gandolfo; Genova / Italy

Carlo Gandolfo; Genova / Italy
Pro (10 min)

This house believes that there is always a place for intervention in paediatric stroke.
Con (10 min)

Manoelle Kossorotoff; Paris / France
This house believes that there is not always a place for intervention in paediatric stroke.
Panel discussion: What is the place for intervention in paediatric stroke? (25 min)
901
RC 2115 - Diagnosis and management of acute and chronic upper limb ischaemia
Categories: Imaging Methods, Interventional Radiology, Multidisciplinary, Vascular

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:
Jernej Lučev; Maribor / Slovenia

Jernej Lučev; Maribor / Slovenia
Pathophysiology (15 min)

Maria Tsitskari; Nicosia / Greece
1. To develop a working knowledge of inflammatory vasculopathy.

2. To arrange these entities in practical order and explain when do we encounter them.
3. To develop multidisciplinary teamwork in your institution.
Imaging of acute and chronic upper limb ischaemia (15 min)

Edit Dósa; Budapest / Hungary
1. To develop a practical strategy for preoperative imaging and planning and how to design a flow chart.
2. To learn how to recognise and understand common vascular conditions.
Endovascular treatment of acute and chronic upper limb ischaemia (15 min)
Elias Brountzos; Athens / Greece
1. To discuss whether there is enough evidence to advocate for specific treatment techniques.
2. To appreciate the morbidity and mortality risks.
3. To learn how to diagnose and manage vascular patients with varied clinical presentations.
Panel discussion: The role of endovascular treatment vs surgery? (10 min)
902
RC 2113 - How to communicate the risk and benefits of the use of shielding of patients and staff?
Categories: EuroSafe Imaging/Radiation Protection, Multidisciplinary, Physics in Medical Imaging, Professional Issues, Radiographers
ETC Level: LEVEL I
Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

Communication from the point of view of shielding patients (15 min)

1. To understand the feelings of patients when exposed to ionising radiation.

2. To understand differences in communication with patients in general, parents of paediatric patients and pregnant patients.
3. To appreciate the support of scientific papers and recommendations regarding the use of shielding.
Risks and benefits of the staff shielding in image (fluoroscopy) guided interventions (15 min)
Gabriel Bartal; Tel Aviv-Yafo / Israel
1. To understand issues of image guided interventionists (radiologists, cardiologists or surgeons), nurses and radiographers using
shielding in Cath labs or operating rooms.
2. To learn about the effects of shielding on personal doses and orthopaedic (including spine) issues.
3. To learn about existing and future scatter radiation reduction systems.
The role of the regulator in the use of shielding for staff and patients (15 min)
Jana Povolná; Prague / Czechia
1. To understand the position of regulators.

2. To understand how regulators can affect clinical practice.
3. To become familiar with the slow pace of changes made on the national and international levels.
Panel discussion: Shielding in the 21st century - science or sentiment? (10 min)
903
RPS 2105 - Exploring the latest frontiers in machine learning techniques for MDCT advancements
Categories: Artificial Intelligence & Machine Learning, Imaging Methods, Multidisciplinary

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Rebeca Mirón Mombiela; Herlev / Denmark
Coronary artery calcium scoring on the segment-level using deep active multi-task learning for time-efficient
annotation (7 min)
Bernhard Föllmer; Berlin / Germany
Author Block: B. Föllmer1, S. Tsogias1, F. Biavati1, M. Bosserdt1, K. F. Kofoed2, P. Maurovich-Horvat3, P. Donnelly4, T. Benedek5, M.
Dewey1; 1Berlin/DE, 2Copenhagen/DK, 3Budapest/HU, 4Belfast/UK, 5Targu Mures/RO
Purpose: The aim of this study was to develop and evaluate a time-efficient annotation strategy for multi-task segment-level
coronary artery calcium scoring (CACS) on non-contrast CT for the improvement of localisation and quantification of calcifications in
the coronary artery tree.
Methods or Background: This study included 1514 patients (mean age 60.0 ± 10.2 years, 55.7% female) with stable chest pain
from the multicentre DISCHARGE trial (NCT02400229), which were randomly divided into a training/validation set (1514), and a test
set (455). We developed a deep active learning strategy for time-efficient annotation of coronary artery segment-regions for auxiliary
task learning in a multi-task model for segmentation of CACs on segment-level. We compared the model with a baseline U-Net in
terms of micro-average sensitivity and micro-average specificity for assigning detected calcification to the correct segment and
analysed interobserver variability in a subset of 150 patients.
Results or Findings: The micro-average sensitivity and micro-average specificity for assigning detected calcification to the correct
segment improved from 0.581 (95% CI, 0.550-0.613) to 0.732 (95% CI, 0.711-0.754, p<0.001) and from 0.965 (95% CI, 0.962-0.968)
to 0.978 (95% CI, 0.976-0.980), respectively (p<.001), compared to the baseline model, with an additional annotation time of less
than 12 hours for annotation of coronary artery segment-regions. The agreement between the model and the reference standard (first
observer) for segment class assignment was good with a weighted Cohen’s κ of 0.806 (95% CI, 0.782-0.828) and only slightly lower
compared to the second observer (weighted Cohen’s κ of 0.819 [95% CI, 0.789-0.849] (p<.001).
Conclusion: Deep active learning can be used for time-efficient annotation of coronary artery regions to improve the performance of
a multi-task model for segment-level CAC scoring.
Limitations: A consensus between the two observers that can serve as a reference standard was not available.
Funding for this study: This work was funded by the German Research Foundation through the graduate program BIOQIC
(GRK2260, project-ID: 289347353), and the DISCHARGE project (603266-2, HEALTH-2012.2.4.-2) funded by the FP7 Program of the
European Commission.
Ethics committee - additional information: The DISCHARGE trial was approved by the ethics committee at
Charité–Universitätsmedizin Berlin as the coordinating center, by the German Federal Office for Radiation Protection, and by local or
national ethics committees.
CT-based artificial intelligence for individual lymph node metastasis assessment in gastric cancer (7 min)
Yun Qin; Chengdu / China
904
Author Block: Y. Qin1, D. Dong2, W. Zhang1, C. Li2, H. Jiang1, M. Bashir3, B. Song1, J. Hu1, J. Tian2; 1Chengdu/CN, 2Beijing/CN, 3Durham,
NC/US
Purpose: Accurate preoperative lymph node (LN) metastasis assessment on a per-LN basis is imperative but challenging in gastric
cancer (GC). Therefore, we aimed to develop an artificial intelligence (AI) model on preoperative CT images to predict individual LN
metastasis and to investigate whether the model could be used to improve radiologist interpretation accuracies.
Methods or Background: Consecutive GC patients of whom the resected LNs could be matched rigidly on preoperative CT images
were retrieved from a prospectively-collected database spanning 5 years. An AI model, integrating handcraft radiomic and deep
learning features, was generated based on both the LNs and corresponding primary tumours. In the testing dataset, LN metastasis
was predicted by the AI model and also by three independent radiologists without and with the assistance of AI against
histopathology. Diagnostic performances were characterized by area under the receiver operating characteristic curve (AUC),
sensitivity, and specificity.
Results or Findings: A total of 1381 LNs (739 [53.5%] metastasis positive) in 666 patients (median age, 61 years; range, 29-86
years; 462 males) were included. In the testing dataset, the AUC of the AI model was 0.804 for predicting individual LN metastasis.
Compared with radiologist consensus interpretation, the AI model demonstrated a higher specificity (70.6% vs. 48.5%, p<0.001) but a
lower sensitivity (73.9% vs. 85.8%, p<0.001). The specificity of radiologist consensus interpretation with AI assistance improved over
that without AI assistance (60.3% vs. 48.5%, p<0.001), whereas the sensitivities remained comparable (81.5% vs. 85.8%, p=0.052).
Conclusion: In GC patients, we developed a CT-based AI model with high accuracy for predicting the metastasis status of individual
LNs. The AI model also showed promise in assisting radiologists to improve their diagnostic accuracies.
Limitations: This was a single-centre retrospective study.
Ethics committee - additional information: This study was approved by the institutional review board (No.2014-215).
Auto-LSN: fully automated liver surface nodularity quantification for the diagnosis of advanced fibrosis in CT imaging
(7 min)
Sisi Yang; Paris / France
Author Block: S. Yang1, A. Bône2, T. Decaens3, A. J. Glaunes1; 1Paris/FR, 2Villepinte/FR, 3Grenoble/FR

Purpose: The liver surface nodularity score was proposed in 2016 to measure the irregularities of the surface of the liver in imaging
and was shown to positively correlate with fibrosis, graded by the Metavir score. The LSN software requires the user to manually draw
regions of interest before automatically segmenting the liver contour and measuring the score. We propose an entirely automated
alternative (auto-LSN), and we compare its potential for the diagnosis of advanced fibrosis with LSN.
Methods or Background: It was a monocentric retrospective study on portal phase CT images in patients with suspected primary
hepatic tumors and an available Metavir score between 2015 and 2020.
With respect to LSN, our auto-LSN method innovates on three key components: (i) automatic segmentation of the liver contour with
deep learning, (ii) exclusion of low-contrast, high-curvature or inner regions of this liver contour, (iii) refinement and smoothing of the
contours with Savitzky-Golay filter in the remaining regions of interest. The potential of auto-LSN to diagnose advanced fibrosis was
measured with the area under curve (AUC), and compared with LSN, measured with the commercially available software by a
radiologist (8 years of experience). The Delong's test was used for the comparison of AUCs.
Results or Findings: One hundred and two patients were included (88 males, age 72y ± 9.89),75 (74%) in the advanced fibrosis
group (Metavir F3/F4). Auto-LSN and LSN scores were 2.64 (± 0.47) and 3.50 (± 0.86) respectively in the advanced fibrosis group,
2.20 (± 0.51) and 3.00 (± 0.67) in the other. The AUC of auto-LSN and LSN were 76% and 68%, respectively (p = 0.18).
Conclusion: The performance of auto-LSN and LSN to diagnose advanced fibrosis were similar. Auto-LSN has the advantage to be
fully automated.
Limitations: Monocentric retrospective study.
Funding for this study: This work is supported by a public grant overseen by the French National Research Agency (ANR) as part of
the Investments for the Future programme (PIA) under grant agreement No. ANR-21-RHUS-01. This work is also funded by Guerbet.
Assessing the accuracy of an AI-based coronary artery calcium score algorithm on non-gated chest CT images with
varying slice thicknesses (7 min)
Dan Mu; Nanjing / China
905
Author Block: D. Mu1, K. Yin1, W. Chen1, X. Chen2, B. Zhang1; 1Nanjing/CN, 2Shanghai/CN
Purpose: This study aimed to evaluate the accuracy of an artificial intelligence-based coronary artery calcium score (AI-CACS)
algorithm when applied to non-gated chest computed tomography (CT) images with varying slice width thickness.
Methods or Background: A total of 112 patients who underwent both chest CT and simultaneous electrocardiogram (ECG)-gated
non-contrast enhanced cardiac CT were prospectively enrolled. Different image thicknesses (1 mm, 3 mm, and 5 mm) were
reconstructed from the same chest CT scan. The coronary artery calcium score (CACS) was obtained semi-automatically from ECG-
gated cardiac CT scans using a conventional CAD method, serving as the reference (ECG-CACS). An AI-based algorithm was
developed to automatically calculate CACS from non-gated chest CT images (AI-CACS). Agreement and correlation were assessed
using Bland-Altman analysis and Spearman correlation coefficients. Risk stratification was also performed and compared.
Results or Findings: AI-CACS demonstrated strong correlations with ECG-CACS for the three different slice thicknesses (1 mm:
0.973, 3 mm: 0.941, 5 mm: 0.834; all p < 0.001). AI-CACS with a 1 mm slice thickness showed no statistically significant difference
compared to ECG-CACS (p=0.085). The Bland-Altman plot revealed mean differences of -6.5, 15.4, and 53.1 for the AI-CACS 1 mm, 3
mm, and 5 mm groups, respectively, with 95% limits of agreement of -95.0 to 81.9, -96.6 to 127.4, and -187.8 to 294.0, respectively.
Agreement of risk categories for CACS was measured by kappa (κ) values (AI-CACS-1mm: 0.868; AI-CACS-3mm: 0.772; AI-CACS-5
mm: 0.412; all p < 0.001), and the concordance rate was 91%, 84.8%, and 62.5%, respectively.
Conclusion: The AI-based algorithm proved to be feasible for calculating CACS from chest CT scans, with images having a 1mm slice
width thickness yielding the best results.
Limitations: Only slice thicknesses of 1 mm, 3 mm, and 5 mm were evaluated in this study. A larger multi-centre, multi-vendor
cohort study shall be conducted.
Ethics committee - additional information: This study was approved by the IRB with approval number: 2022-547-01.
Deep learning decision-making research software prototype for automated inner ear classification (7 min)
Andras Kedves; Innsbruck / Austria
Author Block: A. Kedves1, A. A. Almasri2, S. Sugarova3, A. Alsanosi4, F. Almuhawas4, L. Hofmeyr5, F. Wagner6, K. Sriperumbudur1, A.
Dhanasingh1; 1Innsbruck/AT, 2Pecs/HU, 3St. Petersburg/RU, 4Riyadh/SA, 5Stellenbosch/ZA, 6Bern/CH
Purpose: The aim of this study was to create an efficient DICOM viewer program, that automatically crops the inner ear, and
classifies inner ear malformations (IEM), based on computed tomography (CT).
Methods or Background: Retrospectively we evaluated 2053 patients from three hospitals and extracted 1200 inner ear CTs to
import, crop, and the artificial intelligence (AI) to train, test, and validate. An automated cropping algorithm based on K-means
clustering was created to crop the inner ear volume, along with a simple graphical user interface (GUI). Using the crops as an input,
we created a deep learning convolution neural network (DL CNN) (5-fold cross-validation) to determine whether the inner ear
anatomy is abnormal or normal (data equally distributed). Abnormal anatomy consists of cochlear hypoplasia, ossification, incomplete
partition type I, incomplete partition type III, and common cavity (data equally distributed) were selected. Both the cropping tool and
the AI model were validated.
Results or Findings: We developed an efficient research software prototype that can read CT files and crops a volume that contains
the inner ear. Based on that volume of interest, the AI model makes the classification. The cropping was 92.25% accurate. The area
under the curve (AUC) is 0.86 (95% CI: 0.81-0.91) on DL. Accuracy, precision, recall, and F1 scores are 0.812, 0.791, 0.8, and 0.766,
respectively.
Conclusion: We present a fully automatised workflow of software development and validation tool. Our solution could provide good
diagnostic accuracy during risk stratification; however, must be supervised by the decision-maker.
Limitations: The most important limitation to make the AI model more robust is the number of samples available at the time of this
study.
Ethics committee - additional information: Approved bythe independent ethics committee of three hospitals IRB Nos.
22/0084/IRB, 23_001/IRB, and S_23_001/IRB, respectively.
Deep learning for estimating pulmonary nodule malignancy risk: how much data does AI need to reach radiologist level
performance? (7 min)
Bogdan Obreja; Nijmegen / Netherlands
906
Author Block: B. Obreja1, K. V. Venkadesh1, W. Hendrix1, Z. Saghir2, M. Prokop1, C. Jacobs1; 1Nijmegen/NL, 2Hellerup/DK
Purpose: Deep learning algorithms require large training datasets to achieve optimal performance. For many AI tasks, it is unclear
whether algorithm performance would improve further if more training data was added. The aim of this study is to quantify the
number of CT training samples required to achieve radiologist-level performance for a deep learning AI algorithm that estimates
pulmonary nodule malignancy risk.
Methods or Background: For estimating pulmonary nodule malignancy risk, we used the NLST dataset (malignant nodules:1249,
benign nodules:14828) to train a deep learning algorithm. The dataset was split: 80% training and 20% internal validation. The
algorithm was trained on random subsets of the training set with subset sizes ranging from 10% to 100%, with a class distribution of
malignant≈7.77% and benign≈92.23%. The trained AI algorithms were validated on a size-matched cancer-enriched cohort
(malignant:59, benign:118) from DLCST. The performance was compared against a group of 11 clinicians that also scored the test set,
which included 4 thoracic radiologists.
Results or Findings: Using training data subsets of 10%, 20%, and 30%, the AI achieved AUC values of 0.74 (95%CI:0.67-0.82), 0.79
(95%CI:0.72-0.85), and 0.81 (95%CI:0.74-0.87) respectively. When the training data set size reached 60% (malignant:602,
benign:7112), the performance saturated, reaching an AUC of 0.82 (95%CI:0.75-0.88). This was comparable to the average AUC of all
clinicians (0.82,95%CI:0.77-0.86,p>0.99) and of the four thoracic radiologists (0.82,95%CI:0.74-0.89,p>0.99).
Conclusion: The AI was able to reach the level of an experienced thoracic radiologist when it was trained on 7714 nodules
(malignant:602) from the NLST dataset. These findings have potential implications for the allocation of resources in developing deep
learning algorithms for lung cancer medical imaging diagnostics.
Limitations: The generalizability of these findings is constrained by heterogeneity and geographical limitations of the datasets used
in this study.
Funding for this study: Public private consortium with funding from NWO, Dutch Ministry of Economic Affairs, and MeVis Medical
Solutions, Bremen, Germany.
Ethics committee - additional information: This study included data collected from the National Lung Screening Trial (NLST) and
the Danish Lung Cancer Screening Trial (DLCST). For NLST, IRB approval was obtained at all 33 participating sites and all participants
provided informed consent. For DLCST, the Ethics Committee of Copenhagen County approved the study, and informed consent was
obtained from all participants.
SALT: Softmax for Arbitrary Label Trees (7 min)

Giulia Baldini; Essen / Germany
Author Block: G. Baldini1, S. Koitka1, C. M. Friedrich2, J. Haubold1, B. M. Schaarschmidt1, M. Forsting1, F. Nensa1, R. Hosch1; 1Essen/DE,
2
Dortmund/DE
Purpose: Segmentation networks treat anatomical structures as isolated entities, neglecting their inherent hierarchical relationships.
We aimed to develop Softmax for Arbitrary Label Trees (SALT) that leverages these properties to enhance segmentation speed and
interpretability.
Methods or Background: This study proposes a segmentation method for CT-imaging that employs conditional probabilities to
model the hierarchical structure of anatomical landmarks (such as, the lungs can be split in left/right and in upper/middle/lower lobe).
This study utilizes 900 body region segmentations (883 patients) of the SAROS dataset from The Cancer Imaging Archive (TCIA). The
TotalSegmentator was used to generate additional segmentations for a total of 117 labels. SALT was trained on 600 CTs, and 150 CTs
were used for validation and testing. The model was evaluated on SAROS and on the TCIA dataset LCTSC using the Dice-Similarity-
Coefficient (DSC).
Results or Findings: On the SAROS test set, the model obtained a DSC of 0.99 for abdominal and thoracic cavities, 0.98 for
mediastinum, bones, and pericardium, 0.97 for muscles, 0.93 for subcutaneous tissue, 0.86 for brain, and 0.81 for spinal cord. On the
LCTSC dataset, the model exhibited a DSC of 0.94 for right lung, 0.91 for left lung, 0.93 for both lungs, 0.89 for pericardium, 0.83 for
spinal cord. Furthermore, SALT demonstrated remarkable computational efficiency, being capable of segmenting a whole-body CT in
20 seconds. This feature allows for its integration into clinical workflows, as a full-body segmentation could be automatically and
efficiently computed whenever a CT scan is performed.
Conclusion: SALT used the hierarchical structures inherent in the human body to achieve high-quality segmentations while
delivering exceptional speed. Additionally, this method also allows training using multiple, incompatible datasets.
Limitations: SALT was only evaluated on two TCIA datasets and its performance should be assessed on more datasets and
conditions.
Funding for this study: This study did not receive external funding.
Ethics committee - additional information: Ethical approval was not required for this study, as it exclusively utilizes data already
publicly available on The Cancer Imaging Archive (TCIA).
Validation of an artificial intelligence software for automatic pulmonary nodule volumetry using micro-CT determined
ground truth nodule volumes (7 min)
Louise D'hondt; Ghent / Belgium
907
Author Block: L. D'hondt1, P-J. Kellens1, K. Torfs2, H. Bosmans2, A. Snoeckx3, K. Bacher1; 1Ghent/BE, 2Leuven/BE, 3Antwerp/BE
Purpose: Validation of pulmonary nodule volumetry in clinically available software for automatic nodule detection and volumetry is
currently either underrepresented or based on a consensus-driven ground truth in patient images, introducing uncertainties due to
intrinsic structural differences between readers. Purpose of the study was to validate the nodule volumetry of a computer-aided
detection (CAD) software using objective micro-CT determined ground truth nodule volumes.
Methods or Background: Eighteen 3D-printed solid pulmonary nodules, including six diameters and three morphology classes,
were subjected to high-resolution μCT scanning to establish objective ground truth volumes. The anthropomorphic Lungman phantom,
containing the nodules, was scanned using a 256-slice CT scanner at three CTDIvol-values (6.04, 1.54, 0.20 mGy), and subsequently
reconstructed with both iterative and deep learning image reconstruction, along with either soft or hard kernels. Volumetric accuracy
of a commercially available automatic volumetry software (AVIEW LCS+) was assessed through multiple linear regression, identifying
which predictors (reconstruction algorithm, kernel, dose, morphology, and diameter) significantly influence the outcome (% error).
Results or Findings: Volumes of nodules larger than six mm in diameter were accurate within 10% of their ground truth volume.
Accuracy of volume measurements was significantly influenced by smaller and irregular morphologies (p<0.001). Notably, variation in
reconstruction algorithm exerted no significant influence (p>0.05). Radiation dose and reconstruction kernel emerged as crucial
parameters for accuracy. However, regression analysis showed diminished impact of the latter through significant interaction with
nodule characteristics, resulting in measured volumes closer to the real volumes. Considerable discrepancies were observed between
objective pulmonary nodule volumes and volumes determined through consensus readings.
Conclusion: Robustness of the volumetry software for variations in CT acquisition parameters in a phantom makes it a valuable
candidate application for diverse imaging protocols across multiple centers.
Limitations: This is a phantom study.
Funding for this study: Funding was provided by the FWO “Kom op tegen Kanker”-project for lung cancer screening research in
Belgium. (Project number: G0B1922N).
Ethics committee - additional information: No ethics committee approval was needed, since this study used phantom images.
AI-assisted detection of abdominal pathologies in chest CT scans (7 min)

Mikhail Belyaev; London / United Kingdom
Author Block: E. Petrash, M. Dugova, R. Gareeva, E. Kochina, A. Shevtsov, V. Samokhin, F. Yaushev, M. Belyaev; London/UK
Purpose: Amid intensified focus on chest pathologies, abdominal findings on chest CT scans may be overlooked. Detecting non-
alcoholic fatty liver disease (NAFLD) is crucial due to its hepatocellular carcinoma risk. Early identification of liver, kidney, and adrenal
incidentalomas aids in timely cancer interventions. Swift urolithiasis recognition prevents ureteral complications. Identifying aortic
aneurysms is vital due to its high mortality. Our research assesses an AI solution's efficacy in enhancing these detection rates on
chest CT.
Methods or Background: The retrospective study used data from the national lung screening trial, consisting of 2408 chest CTs
from primary patients. We excluded 130 cases with a pathology below the diaphragm mentioned in the report. The remaining 2278
CTs without described abdominal findings were auto-analysed by a comprehensive AI product, AUCT-Abdomen, which detects adrenal
mass, NAFLD, liver & kidney lesions, urolithiasis, and abdominal aortic dilatation. CTs with AI findings were then independently
reviewed by two radiologists with 12 and 14 years of experience.
Results or Findings: AI identified 270 (11.2%) previously unreported patients on top of 130 (5.3%) reported initially. AI findings
include 125 adrenal mass, 21 NAFLD, 68 liver lesions, 17 kidney lesions, 5 urolithiasis, 20 aortic dilatation, and 5 aortic aneurysm. AI
results were false positives in 17 (0.7%) cases.
Conclusion: Integrating AI into the radiological evaluation of chest CT scans may ensure the thorough detection of abdominal
incidental findings, including adrenal mass, hepatic steatosis, liver and kidney lesions, urolithiasis, aortic dilatation and aneurysm. Our
research may highlight a potential deprioritisation of findings peripheral to the chest and the potential of AI in bridging this diagnostic
gap. The potential benefits from deploying AI products for incidental abdominal findings may be exceptionally high for departments
with strictly limited turnaround time.
Limitations: No limitations were identified for this study.
Funding for this study: Funding for this study was provided by AUMI AI Limited.
Ethics committee - additional information: The study is retrospective so no ethical approval was sought.
Fully automatic AI-driven assessment in coronary CT angiography for intermediate stenosis: a comparative study with
quantitative coronary angiography and fractional flow reserve (7 min)
Jung-In Jo; Seoul / Korea, Republic of
908
Author Block: J-I. Jo, H. J. J. Koo, J-W. Kang, D. H. Yang; Seoul/KR
Purpose: Limited data exists on direct comparison of AI-driven automatic coronary stenosis assessment in coronary CT angiography
(CCTA) to quantitative coronary angiography (QCA). This study aims to compare AI-based coronary stenosis evaluation in CCTA with
its quantitative counterpart of coronary angiography and invasive fractional flow reserve (FFR).
Methods or Background: In this single-center retrospective study, 215 intermediate coronary lesions, with QCA diameter stenosis
between 20% and 80%, were assessed from 195 symptomatic patients (mean age 61 years, 149 men, 585 coronary arteries). For
stenosis quantification in CCTA, an AI-driven research prototype (Siemens Healthineers, Germany) was used (AI-CCTA). Diagnostic
accuracy of AI-CCTA on per-vessel basis was assessed, using invasive coronary angiography stenosis grading (with > 50% stenosis) or
invasive FFR (< 0.80) as reference standards. AI-driven diameter stenosis was then correlated with QCA results and expert manual
measurements.
Results or Findings: Among 585 coronary arteries, disease prevalence as determined by invasive angiography (≥ 50%) was 46.5%.
AI-CCTA showed sensitivity of 71.7%, specificity of 89.8%, positive predictive value of 85.9%, negative predictive value of 78.5%, and
area under the curve (AUC) of 0.81. For 215 intermediate lesions assessed using QCA and FFR, diagnostic performance of AI-CCTA
was moderate, with AUC of 0.63 for both QCA and FFR. In measuring degree of stenosis, AI-CCTA demonstrated a moderate
correlation with QCA (r = 0.42, p < 0.001), which was notably better than results from manual quantification versus QCA (r = 0.26,
p=0.001).
Conclusion: The AI-powered automated CCTA analysis showed promising results when compared to invasive angiography. While AI-
CCTA demonstrated a moderate relationship with QCA in intermediate coronary stenosis lesions, its results appeared to surpass those
of manual evaluations.
Limitations: Individuals with stents or a history of coronary-artery bypass grafting were excluded from analysis.
Ethics committee - additional information: Approval from the Institutional Review Board of Asan Medical Center, Seoul, Korea
Using RegGAN to generate synthetic CT images from CBCT images acquired with different linear accelerators (7 min)
Kai Zhen Li; Jinan / China
Author Block: K. Z. Li; Jinan/CN

Purpose: The goal was to investigate the feasibility of the registration generative adversarial network (RegGAN) model in image
conversion for performing adaptive radiation therapy on the head and neck and its stability under different cone beam computed
tomography (CBCT) models.
Methods or Background: A total of 100 CBCT and CT images of patients diagnosed with head and neck tumours were utilised for
the training phase, whereas the testing phase involved 40 distinct patients obtained from four different linear accelerators. The
RegGAN model was trained and tested to evaluate its performance. The generated synthetic CT (sCT) image quality was compared to
that of planning CT (pCT) images by employing metrics such as the mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and
structural similarity index measure (SSIM). Moreover, the radiation therapy plan was uniformly applied to both the sCT and pCT
images to analyse the planning target volume (PTV) dose statistics and calculate the dose difference rate, reinforcing the model's
accuracy.
Results or Findings: The generated sCT images had good image quality, and no significant differences were observed among the
different CBCT modes. The conversion effect achieved for Synergy was the best, and the MAE decreased from 231.3±55.48 to
45.63±10.78; the PSNR increased from 19.40±1.46 to 26.75±1.32; the SSIM increased from 0.82±0.02 to 0.85±0.04. The quality
improvement effect achieved for sCT image synthesis based on RegGAN was obvious, and no significant sCT synthesis differences
were observed among different accelerators.
Conclusion: The sCT images generated by the RegGAN model had high image quality, and the RegGAN model exhibited a strong
generalization ability across different accelerators, enabling its outputs to be used as reference images for performing adaptive
radiation therapy on the head and neck.
Funding for this study: This work was supported in part by the National Key Research and Development Program of China
(2016YFC0105106); National Natural Science Foundation of China under Grant 82102173/12275162/82072094;
Ethics committee - additional information: This study was reviewed and approved by the Institution Review Board of Shandong
Cancer Hospital. All methods of this study were carried out in accordance with relevant guidelines and regulations. Informed consent
to participate in the study was obtained from all study participants for personally identifiable data.
Artificial intelligence-based non-contrast-enhanced CT images for diagnosis of hepatic lesions: a multicentre study (7
min)
Zhuangxuan Ma; Shanghai / China
909
Author Block: Z. Ma, L. Jin, M. Li; Shanghai/CN
Purpose: We aimed to realise the identification and classification diagnosis of intrahepatic space-occupying lesions on non-contrast
enhanced CT (NCCT).
Methods or Background: In this retrospective study, patients who suspected with space-occupying lesions in liver undergoing both
NCCT and enhanced CT/MRI multi-phase enhancement from January 2017 to March 2023 in our hospital and from January 2020 to
August 2023 in another medical center. Each liver lesions in NCCT were confirmed by enhanced CT/MRI multi-phase enhancement or
pathology as the golden reference. The lesion contours of the patients in NCCT images were manually delineated by radiologists, and
radiomics features were extracted in 3D Slicer. An automatic machine learning algorithm was used to screen out the most relevant
radiomics features and establish a classification model for differential diagnosis to classify type of intrahepatic space-occupying
lesions.
Results or Findings: A total of 252 liver lesions in 230 patients from our hospital including 79 hepatic cysts, 81 haemangiomas, 52
malignant tumours and 40 liver abscesses. A total of 33 liver lesions in 230 patients from another medical centre including 12 hepatic
cysts, 8 haemangiomas, 8 malignant tumours and 5 liver abscesses. The sensitivity of the nnDetection to detect the lesion was 0.81,
the AUC of classification model for lesions were 1.0 (hepatic cysts), 0.99 (hemangiomas), 0.9 8(malignant tumours) and 0.98 (liver
abscesses) in our internal test dataset while the AUC were 1.0 (hepatic cysts), 0.97 (haemangiomas), 0.9 3 (malignant tumours) and
0.92 (liver abscesses) in outside test dataset.
Conclusion: Our proposed model showed better performance of the identification and classification diagnosis of intrahepatic space-
occupying lesions on NCCT.
Limitations: The small sample size may lead the bias of this study.
Ethics committee - additional information: The study was approved by Huadong hospital (reference number: 20230089).
910
E³ 2121 - New directions in advanced cardiac imaging
Categories: Cardiac, Imaging Informatics, Imaging Methods, Physics in Medical Imaging

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Prognostic biomarkers in ischaemic heart disease (45 min)

1. To become familiar with different imaging biomarkers in ischaemic heart disease.

2. To learn about the importance of advanced plaque analysis for outcome and prognosis in ischaemic heart disease.
3. To discuss future directions for cardiac imaging to predict and improve outcomes in ischaemic heart disease.
Prognostic biomarkers in non-ischaemic cardiomyopathies (45 min)

Antonio Esposito; Milano / Italy
1. To become familiar with advanced MRI and CT tissue characterisation techniques in non-ischaemic cardiomyopathies, including
LGE, mapping, ECV, strain and more.
2. To learn about the role of imaging-derived biomarkers in the different phenotypes of non-ischaemic cardiomyopathies.
3. To discuss further the future role of MRI and CT in predicting and improving the outcome of these patients.
911
RC 2117 - Stroke and its mimics
Categories: Emergency Imaging, Neuro

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

Radiological algorithms in stroke (15 min)

Danielle Byrne; Dublin / Ireland
1. To review the role of perfusion imaging in stroke and its pitfalls.

2. To identify suitable candidates for endovascular treatment.
3. To outline a logical approach to acute stroke imaging with a succinct algorithm allowing the radiologist to interpret imaging and
communicate results in a timely manner.
Perfusion CT mimics (15 min)

Koenraad Hans Nieboer; Brussels / Belgium
1. To understand the importance of a correct interpretation of CTP to avoid unnecessary fibrinolytic therapies.
2. To review the patterns of abnormal findings on CTP in stroke mimics.
3. To recognise common stroke mimics on CT perfusion.
MRI: the problem solver (15 min)

Ronni Mikkelsen; Aarhus / Denmark
1. To distinguish between signs of infarction and ischemia on stroke MRI (core and penumbra).
2. To name the most common sequences and their use in stroke evaluation.
3. To identify some differential diagnoses of ischemic stroke (stroke mimics).
Panel discussion: Time is brain - getting the most from state-of-the-art imaging (10 min)
912
SF 21a - United we stand, divided we fall: a multidisciplinary approach to endometriosis
Categories: Genitourinary, Nuclear Medicine

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

The gynaecologist's perspective (20 min)

Cyril Touboul; Paris / France
1. To define the place of imaging in the diagnostic strategy of endometriosis.

2. To understand the role of pre-operative imaging and imaging classification in planning surgery.
3. To know how to use imaging in the suspicion of recurrence.
The radiologist's perspective (US/MR) (25 min)

Luciana Chamie; Sao Paulo / Brazil
1. To discuss the value of US and MRI to diagnose adnexal locations.

2. To examine the utility of the combination of US and MR for diagnosing deep pelvic locations.
3. To describe lateral locations using MR imaging, including beyond parietal fascia.
Cryoablation (15 min)

Leo Razakamanantsoa; Paris / France
1. To consider the advantages of percutaneous treatment of endometriotic wall nodule.

2. To describe the technique and main results in the literature.
3. To be aware of potential complications to correctly inform our patients.
The patient association's perspective (5 min)

Panel discussion: Endometriosis pathway: where are we at? (15 min)
913
OF 21T - Career development: building a research publication portfolio as a young radiologist

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:

How to make the most of research opportunities (12 min)

1. To discuss the benefits to young radiologists of being active in research.

2. To discuss opportunities provided by the ESR and subspecialty societies for residents to undertake research.
Building research publications: from the young radiologists' network perspective (12 min)
Anmol Gangi; Nottingham / United Kingdom
1. To describe how a research network strategy can improve resident engagement in research.
2. To discuss exemplars to show how residents in non-academic hospitals can be successful researchers.
Tips and tricks to get published (12 min)

1. To discuss what radiology journals are looking for.

2. To discuss the strategies that young radiologists should use to get published.
Open forum discussion: Making the most of research opportunities as a young radiologist (20 min)
914
RPS 2101 - Acute abdominal pathologies: diagnosis and prediction
Categories: Abdominal Viscera, Emergency Imaging, GI Tract

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Martina Scharitzer; Vienna / Austria
Dual-Energy CT of acute bowel ischaemia: influence on diagnostic accuracy and reader confidence (7 min)
Moritz Oberparleiter; Basel / Switzerland
Author Block: M. Oberparleiter, J. Vosshenrich, H-C. Breit, B. Friebe, D. Harder, D. Boll, C. J. Zech, M. Obmann; Basel/CH
Purpose: Current guidelines do not recommend the use of Dual-Energy CT (DECT) for suspected acute mesenteric ischaemia due to
a lack of clinical studies. The purpose of this study was to investigate the diagnostic accuracy, reader confidence and reading time of
DECT compared to conventional CT.
Methods or Background: 25 patients with surgically proven acute mesenteric ischaemia and 25 gender- and age-matched controls,
who underwent arterial and portal venous phase DECT of the abdomen were included in this retrospective study. Two fellowship-
trained abdominal radiologists evaluated all cases with and without the use of DECT-derived virtual non-contrast images and iodine
maps for mesenteric ischaemia. Reading time was recorded and diagnostic confidence was rated on a 10-point Likert scale. The inter-
reader agreement was assessed using Cohen's kappa. Sensitivity and specificity were compared using McNemar’s test, reading time
and reader confidence with the Wilcoxon rank-sum test.
Results or Findings: Inter-reader agreement was good ( =0.72). Sensitivity and specificity for diagnosing acute mesenteric
ischaemia were 78% and 100% using conventional image data alone. Utilising additional DECT data, sensitivity was significantly
higher at 94% (p=0.02), while specificity remained 100%. Diagnostic confidence increased significantly from 8 (IQR, 7-10) to 9 (IQR,
8-10) (p<0.01). Mean reading time per case increased significantly from 154 to 183 s (p=0.02) using additional DECT images.
Conclusion: Additional use of DECT should be considered when examining for suspected acute mesenteric ischaemia as DECT
increased reader diagnostic accuracy and confidence for mesenteric ischaemia with only a moderate increase in reading time.
Limitations: Only two DECT scanner types were used in this study and results may not be transferable to other DECT platforms.
Readers who are familiar with DECT technology may benefit while inexperienced readers may benefit less from DECT images.
Funding for this study: This research received no specific grant from any funding agency in the public, commercial or non-profit
sectors.
Ethics committee - additional information: The need for informed consent was waived due to the retrospective nature of this
study.
Hit and miss: locating the exact site of gastrointestinal tract perforations as a challenge (7 min)
Tarik Binasa Plojović; Belgrade / Serbia
915
Author Block: T. B. Plojović, A. Pavlovic, D. Markovic, D. Janjic, J. Vukmirovic, K. Lazarevic, B. Jovandić, D. Vasin, S. Hasanagic;
Belgrade/RS
Purpose: The objective of our study was to analyse the capacity of CT (Computerized Tomography) to identify the site of
gastrointestinal perforation (GI) and to determine which radiological signs, either direct or indirect, are the most predictive.
Methods or Background: Between September 2022 and September 2023, we retrospectively studied 100 patients presenting with
pneumoperitoneum on CT. All patients had surgically proven gastrointestinal tract perforation. Two expert radiologists, with no
previous knowledge of the clinical histories or the surgical results, evaluated the CT scans.
Results or Findings: The locations of the perforations found during surgery in the 100 patients were as follows: 36 stomach or
duodenum; 15 small intestine; 12 appendix; 16 ascending, transverse or descending colon; and 21 sigma/rectum. The Kappa
correlation coefficient between radiologists for predicting the localisation of the perforation in our study was high. The two most
frequent signs observed in our study were free extraluminal air in the supramesocolic space and gas bubbles adjacent to the wall. The
prediction of the perforation site in the gastrointestinal tract using CT coincided with the surgical findings in 80 out of 100 patients. In
20 patients, the prediction did not concur with the findings. In 15 cases, CT identified an incorrect perforation site while in 5 patients,
CT did not identify the location of the GI perforation. The most sensitive sign in our study was the presence of free extraluminal air in
the supramesocolic compartment. The most specific ones were the presence of abscesses and focal wall defects.
Conclusion: CT can locate gastrointestinal perforation sites with a high sensitivity and excellent interobserver correlation.
Ethics committee - additional information: This study was approved by the Institutional Ethics Committee.
CT prognostic signs in emergency surgery for acute obstructive colonic cancer (AOCC) (7 min)
Rosita Comune; Succivo (CE) / Italy
Author Block: F. Pezzullo1, S. Tamburrini1, R. Comune1, R. D'avino1, C. Liguori1, M. Scaglione2, F. Tamburro1; 1Naples/IT, 2Sassari/IT
Purpose: This study aimed to identify CT prognostic signs of poor outcomes of emergency surgery in AOCC.
Methods or Background: Demographic, clinical, laboratory, radiological and surgical data of 65 consecutive patients with AOCC
who underwent emergency surgery were analysed. CTs were reviewed to assess the diameters of the cecum, ascending, transverse,
descending, and sigmoid proximal to the tumour. Furthermore, colon segments’ CD/L1-VD ratios, continence of the ileocecal valve,
small bowel over-distension, presence of faecal signs and cecal pneumatosis were also analysed. Postoperative complications,
according to Clavien-Dindo classifications, were used.
Results or Findings: Preoperative transverse and descending colon CD/L1-VD ratios were significantly associated with the
development of postoperative complications with a cut-off value of >/=1.4 and 1.3 (p=0.157 and p=0.008 - Clavien-Dindo
classification major - grade ≥III-V), respectively. Postoperative complications within 30 days after surgery occurred in 18/65 patients,
with 12 patients developing surgical complications (18.5%), 3 patients developing medical complications (4.6%) and 3 patients dying
(4.6%). Of the 18 patients, 15 (23,1%) developed severe complications (grade ≥III-V).
A cecum distension >/= 9 cm represented the critical dimension beyond which perforation and cecal necrosis were found at surgery
(11/65 patients). Cecal pneumatosis was detected in 5/11 patients.
Conclusion: CT is a valid tool to select patients at higher risk of complications. A CD/L1-VD ratio with cut-off values of 1.4
(transverse) and 1.3 (descending) predicted major complications (grade ≥III-V). A cecum diameter >/= 9 cm and continence of ileo
cecal valve were predictive factors of poor outcome and cecal necrosis. The CT sign of cecal pneumatosis was not pathognomonic for
cecal necrosis.
Limitations: This was a retrospective study and prospective multicenter studies are needed to validate our results.
Ethics committee - additional information: This study was approved by the local Institutional Review Board. The Ethical
Committee approval was obtained (2022030). Patient consent was waived (Retrospective Observational Study).
Spectral parameters in dual-layer spectral detector CT for non-Invasive prediction of oesophageal varices bleeding risk
in cirrhotic patients (7 min)
Baoyuan Wang; Lanzhou / China
916
Author Block: B. Wang, J. Li, W. Jin, J. Lei, Y. Wang; Lanzhou/CN
Purpose: This study aimed to explore the potential of dual-layer spectral detector computed tomography (DLCT) for the non-invasive
assessment of oesophagal varices bleeding risk in cirrhotic patients.
Methods or Background: 31 cirrhosis patients who were complicated with oesophagal varices underwent contrast enhancement
DLCT and data were retrospectively collected from May 2023 to September 2023. Patients were divided into low-risk (mild-moderate
oesophagal varices, n=20) and high-risk (severe oesophagal varices, n=11) groups based on endoscopic findings within 2 weeks after
DLCT scans. DLCT quantitative parameters such as arterial enhancement fraction (AEF), Iodine concentration (IC) and effective atomic
number (Zeff) in the liver and spleen were measured. We measured the diameters of both the portal and splenic veins, as well as the
volumes of the liver and spleen, and calculated the relevant ratios. T-test and Mann-Whitney U-test were performed to compare
parameters between the two groups. Logistic regression analysis was performed to obtain the independent risk factors for
oesophageal varices bleeding. The performance was evaluated using ROC analysis.
Results or Findings: There were significant differences in hepatic DLCT-AEF, the portal vein diameter and spleen volume between
the two groups (all p<0.05). The high-risk group demonstrated significantly lower DLCT-AEF (median,10.3% vs 3.3%; p=0.02) than
the low-risk group. Logistic regression analysis showed that liver AEF (odds ratio [OR]=0.85, p<0.05) and portal vein diameter
(OR=1.36, p<0.05) as independent predictors of bleeding risk. The combination of hepatic AEF and portal vein diameter showed an
AUC of 0.79, with a sensitivity of 91% and specificity of 70%.
Conclusion: The incorporation of DLCT-AEF and portal diameter may be used as an effective alternative to endoscopy for predicting
the risk of oesophagal varices bleeding risk in cirrhotic patients.
Limitations: This study was a single-centre study with a small sample size included.
Ethics committee - additional information: The ethics committee approval was obtained.
(HU)nt for truth: psoas muscle evaluation on CT imaging predicts 30-day survival in variceal bleed patients (7 min)
Jędrzej Krawczyk; London / United Kingdom
Author Block: J. Krawczyk, D. Leon, L. D. Tyson, T. Haq, D. Saba, R. H. Thomas, A. Dhar; London/UK
Purpose: This study aimed to assess if sarcopenia defined on cross-sectional imaging predicts 30-day mortality in the setting of
variceal bleeding, independent of baseline liver disease severity.
Methods or Background: A retrospective study was performed of all patients with decompensated cirrhosis admitted with a
variceal bleed over a 24-month period to our tertiary centre. CT imaging acquired within 3 months of admission was reviewed to
determine psoas muscle thickness, width, area and Hounsfield Units (HU: a surrogate measure of psoas density) at the level of the
inferior L4 endplate. The area of the psoas was calculated using a generic lesion segmentation tool and the average HU value was
calculated from circular ROI in the area of the muscle. Association with 30-day mortality was tested by univariate and multivariate
binary logistic regression; and the utility of identified prognostic biomarkers by AUROC analyses (SPSS v 27).
Results or Findings: 104 patients with decompensated cirrhosis were identified with variceal bleed episodes. 15/104 died within 30
days. As expected, a higher MELD score was associated with higher mortality (OR 1.106, 95% CI 1.034-1.183, P=0.003). Higher Psoas
HU and PMTH were both associated with lower mortality, independent of MELD (Adjusted OR 0.891, 0.821-0.967, P=0.006; AOR 0.810,
0.665-0.987, P=0.036 respectively). Psoas HU was predictive of 30-day survival (AUROC 0.790, 0.673-0.907, P=0.004) as was PMTH
(AUROC 0.743, 0.571-0.914, P=0.017), and MELD (AUROC 0.719, 0.534-0.904, P=0.008).
Conclusion: Psoas HU and PMTH are negatively associated with mortality in patients with decompensated cirrhosis and variceal
bleeding. Crucially, this relationship appears independent of baseline liver disease severity (MELD score).
Limitations: Larger studies are needed to confirm the association of Psoas HU and PMTH with 30-day mortality and to further define
the role of these parameters as prognostic indexes in this clinical setting.
Risk factors of spontaneous rupture of primary liver cancer based on CT imaging features (7 min)
Chen Yong; Yinchuan / China
917
Author Block: C. Yong; Yinchuan, Ningxia/CN
Purpose: This study aimed to analyse and summarize the imaging signs of spontaneous rupture of primary liver cancer (PLC) and
explore the high-risk factors affecting tumour rupture.
Methods or Background: From September 2016 to August 2020, 81 patients who suffered from spontaneous PLC rupture were
included in this study. A control group of 81 patients was randomly selected by matching the age, sex and BCLC stage during the
same period. The clinical data and CT imaging characteristics of patients with spontaneous rupture of primary liver cancer were
analysed retrospectively.
Results or Findings: The main risk factors for spontaneous rupture of PLC include cirrhosis, tumour close to the diaphragm, biolobar
distribution (P<0.05), portal vein obstruction, tumour diameter >10 cm, invasion of the liver capsule (arc-to-chord ratio>1) and
tumour protrusion ≥25% (P<0.001). Furthermore, logistic regression analysis showed that cirrhosis (OR 0.278, 95% CI: 0.078-0.990),
portal obstruction (OR 3.586, 95% CI: 1.272, 10.107), and tumour protrusion (≥25%) (OR 2.831, 95% CI: 1.668-4.806) were relatively
independent predictive factors of spontaneous rupture of PLC.
Conclusion: Tumour protrusion ≥25%, cirrhosis and portal vein obstruction are closely related to spontaneous rupture of PLC and
can be used as independent risk factors to predict the rupture of primary liver cancer.
Limitations: This is a retrospective single-centre study.
Funding for this study: This study was funded by the National Natural Science Foundation of China.
Ethics committee - additional information: This study was approved by the institutional ethics committee of General Hospital of
Ningxia Medical University.
Diagnostic impact of AAST injury scale in the assessment of hepatic traumatic lesions (7 min)
Cesare Maino; Monza / Italy
Author Block: C. Maino, C. Talei Franzesi, M. Ragusi, D. G. Gandola, T. P. Giandola, P. N. Franco, D. Ippolito; Monza/IT
Purpose: This study aimed to determine the diagnostic impact of computed tomography in the evaluation and management of
patients with traumatic liver lesions following the American Association for Surgery of Trauma (AAST) scale and its relation to clinical
and laboratory data.
Methods or Background: A total of 103 hemodynamically stable patients, with traumatic liver injury, who underwent contrast-
enhanced CT scan to assess and quantify liver damage, were enrolled. Imaging data were independently evaluated by a general
surgeon and a radiologist (both with more than 15 years of experience). The reviewers first graded liver lesions, according to the
AAST scale, blinded to the clinical data. During the second revision session, the reviewers reconsidered the CT findings along with the
support of clinical data. The primary study outcome was to determine the patient's management [operative (OM) or not-operative
(NOM)] based exclusively on imaging CT findings and by adding laboratory data.
Results or Findings: A good inter-reader agreement was found for AAST grades I, II, III, and V (k= 0.870, k=0.880, k=0.900, and
k=1); while in grade IV the agreement was fair (k=0.455). According to the first revision section, the accuracy in determining the
management was higher for the radiologist (AUC=0.850, 95% CI 0.770-0.950) than the surgeon (AUC=0.700 95% CI 0.550-0.820),
achieving a statistically significant difference (p=0.025). During the second revision session, after correlation with clinical and
laboratory data, the overall accuracy between the two readers was statistically comparable (AUC=0.880 and AUC=0.850, p>0.05).
Conclusion: The CT liver damage score, according to the AAST scale, represents a useful and fast approach to correctly address the
management of liver trauma patients.
Ethics committee - additional information: The study was retrospective.
Are non-contrast-enhanced abdominal CT scans more effective for the diagnosis of choledochal stones than MRCP in
patients with acute biliary pancreatitis? (7 min)
Sercan Kiremitçi; Nevşehir / Turkey
918
Author Block: Y. Aksu, S. Kiremitçi; Nevşehir/TR
Purpose: This study aimed to verify the diagnostic performance of non-contrast-enhanced abdominal CT scans to demonstrate
choledocholithiasis in contrast to MRCP in patients with acute biliary pancreatitis.
Methods or Background: In this case-controlled study, we evaluated 54 eligible patients for acute biliary pancreatitis. Endoscopic
Retrograde Cholangiopancreatography (ERCP) was performed for all patients and choledocolithiasis was verified with ERCP. Basic
laboratory parameters, ultrasound, non-contrast-enhanced abdominal CT scans, MRCP and ERCP findings were evaluated.
Results or Findings: The patient group mean age was 63,9±17,4. The mean amylase level was 266,3±587,4. The mean lipase level
was 381,8±944,5. The mean total bilirubin level was 3,07±6,51. The mean direct bilirubin level was 1,75±2,12. The mean
choledochal diameter was 11,2±4,2 mm. In ultrasound examinations, the mean was 12,06±4,8 mm; and in CT scans, the mean was
11,2±4,34 mm. There was no statistically significant difference between choledochal diameter and all modalities (p>0,05). 43
(79,6%) of 54 choledochal stones were shown with CT scans, while 25 (46,3%) of 54 choledochal stones were shown with MRCP. 17 of
54 choledochal stones were isodense. 27 of 54 CS were hyperdense. 7 of 54 CS were hypodense in CT scans. Logistic regression
analysis demonstrated choledochal stone sensitivity with 96% specificity. 34% with non-contrast-enhanced CT scans. PPV was 55,81%
and NPV 90,91% (the odd ratio was 12,632 [AUC;0,65 [95%].
Conclusion: We validated that the diagnostic performance of a non-contrast-enhanced CT scan was superior to MRCP for
demonstrating choledochal stone in patients with acute biliary pancreatitis. We verified this result with logistic regression analysis and
a cross-tabulation method. We concluded when physicians suspect choledochal stones after the ultrasound, the next step should be
to employ non-contrast-enhanced abdominal CT scans.
Limitations: This study had a small sample size, lacked stone biochemical assessment, and lacked control USG, CT scans and MRCP
examinations.
Ethics committee - additional information: This study was approved by the Nevşehir Hacıbektaş University Ethical Committee.
Evaluation of detectability for radiolucent gallstones by multi-parameter imaging on spectral CT (7 min)

Jin-ge Zhang; Chengdu / China
Author Block: J-g. Zhang; Chengdu/CN

Purpose: This study aimed to explore the clinical value of multi-parameter imaging of dual-layer detector spectral CT in the diagnosis
of gallstones and assess the ability to differentiate radiolucent gallstones (negative gallstones) from bile.
Methods or Background: 38 patients with suspected negative gallstones who underwent spectral CT examinations were
retrospectively enrolled. All patients had ultrasounds or surgical results as the gold standard. Conventional CT attenuation, virtual
monochromatic spectral (VMS) image, effective atomic number map, electron density map, iodine density map and spectral curves
were generated and quantitatively analysed for all negative gallstones. The statistical analysis used ANOVA and paired T-tests. The
sensitivity, specificity and diagnostic accuracy of spectral images for negative gallstones were calculated using ultrasound or surgical
results as reference standards.
Results or Findings: The CT attenuations of negative gallstones were significantly different from those of bile in VMS images at
different energy levels (all P< 0.05) and were more different at low energy levels (40-60 keV, P<0.001). The spectral curve of
negative gallstones represented an upward curve with a slope of 1.76±0.34, while the curve of bile in the adjacent area showed a
downward curve with a slope of -1.03±0.28, which indicated a significant difference between the two curves (P<0.05). There were
significant differences in the effective atomic number and electron density between negative gallstones and bile for 6.72±1.13 vs.
7.95±0.54 and (104.2±1.2) %EDW vs. (100.3±0.8) %EDW, respectively (Both P<0.05). The sensitivity, specificity and accuracy of
spectral CT in the diagnosis of negative gallstones were 93.75%, 100% and 94,44%, respectively.
Conclusion: The multi-parameter spectral images of dual-layer detector CT could provide more quantitative parameters for the
diagnosis of negative gallstones, which would improve the detectability of negative stones.
Limitations: There were a small number of patients in the study.
Ethics committee - additional information: This study was approved by the Ethics Committee of West China Hospital (No.
2022-447).
Early IVIM-DWI for the prediction of post-pancreatectomy acute pancreatitis (7 min)

Luca Fortuna; Vicenza / Italy
919
Author Block: L. Fortuna, L. Costa, B. Maris, G. Zamboni; Verona/IT
Purpose: This study aimed to evaluate early DW-MRI radiological findings and texture analysis parameters that will predict the
development of post-pancreatectomy acute pancreatitis (PPAP) in patients undergoing pancreaticoduodenectomy.
Methods or Background: In this IRB-approved prospective study, 65 patients underwent MRI on the third postoperative day after
pancreaticoduodenectomy. Scan protocol included standard sequences, post-gadolinium acquisitions, and IVIM-DWI. IVIM DICOM
images were analysed with in-house software that produced F, D, and D* maps and allowed us to calculate texture parameters of
three different ROIs (stump, tail, entire pancreatic remnant). By retrospectively applying the 2021 ISGPS definition of PPAP, patients
were defined as with or without POH/PPAP. Texture parameters and radiological findings were compared between the two groups
(Kruskal-Wallis, ANOVA tests).
Results or Findings: The patient population included 33 females and 32 males. 20 patients developed postoperative
hyperamylasemia (POH) and 6 of these grade B or C PPAP. Significant differences in texture parameters were identified between the
POH/PPAP and the non-POH/PPAP groups for mean ADC (1.33±0.22 vs 1.56±0.28) × (10−3 mm2/s; p=0.006) and D value (0.11±0.14
vs 0.22±0.15; p=0.03), F entropy (4.5±0.10 vs 4.1±0.08 (SE); p=0.004) and D* entropy (2.7±0.73 vs 1.3±0.38 (SE); p=0.01) of ROIs
including the pancreatic stump. Similar results were found evaluating ROIs of the tail (mean D, p=0.01; entropy F, p=0.02; entropy
D*, p=0.01) and entire pancreatic remnant (mean D, p=0.01; entropy F, p=0.0001; entropy D*, p=0.0001). No macroscopic features
consistent with PPAP were identified.
Conclusion: Early postoperative MRI texture analysis of IVIM-derived parameters might predict who will develop PPAP after
pancreaticoduodenectomy.
Limitations: The limitations of the study are the limited sample size, the difference in slice thickness between post-contrast T1
images and those in DWI IVIM; the exclusion of patients unable to perform MRI due to early clinical worsening; and POH and PPAP
groups union.
Ethics committee - additional information: This study was approved by the Ethics Committee. The notification can be found
under the number 2130 CESC VR-RO.
Time dependency and risk factors of splanchnic vein thrombosis development in the early phase of acute pancreatitis:
a systematic review and meta-analysis (7 min)
Zsolt Zsolt Borbély Ruben; Budapest / Hungary
Author Block: Z. Z. Borbély Ruben1, B. M. Philip1, E. Á. Szalai1, B. Gellért1, D. Veres Sandor1, B. Teutsch1, B. Erőss1, P. Hegyi1, N.
Faluhelyi2; 1Budapest/HU, 2Pécs/HU
Purpose: Splanchnic vein thrombosis (SVT) is a local complication of acute pancreatitis (AP) that may lead to subsequent
complications such as portal hypertension, gastrointestinal bleeding, and mesenteric ischemia. This study aimed to analyse the
temporal progression and contributing risk factors of SVT occurrence during the early phase of AP.
Methods or Background: We systematically searched medical databases (Embase, MEDLINE via PubMed, Scopus, and CENTRAL) on
27.10.2022. Inclusion criteria were studies using appropriate diagnostic modalities to identify SVT in patients from the early phase of
AP. We performed a random-effects meta-analysis, calculated SVT-affected patient proportions with 95% confidence intervals (CI) and
conducted subgroup analyses. The protocol was prospectively registered in PROSPERO: CRD42022367578.
Results or Findings: The proportion of patients with SVT within 12 days after symptom onset was 0.13 (CI 0.07-0.23). The
occurrence was lowest at 0.06 (CI 0.03-0.1) between 0-3 days after symptom onset and increased fourfold to 0.23 (CI 0.16-0.31)
between 3-11 days. The proportion of patients affected on hospital admission was 0.12 (CI 0.02-0.49), and it was 0.17 (CI 0.03-0.58)
1-5 days after admission. Alcoholic aetiology (0.31, CI 0.13-0.58) and pancreatic necrosis (0.55, CI 0.29-0.78, necrosis above 30%)
correlated with increased SVT prevalence.
Conclusion: The risk of developing SVT is significant in AP, affecting up to a quarter of patients. The risk of occurrence increases with
time in the early stages of AP. Alcoholic aetiology and pancreatic necrosis elevate the risk for SVT in AP. Our findings highlight the
need for anticoagulation therapy and advanced imaging (CT, MRI) to become a routine component of AP therapy.
Limitations: The limitations were the lack of individual patient data and aggregate data from published studies limited our ability to
control for potential confounders and explore effect modifiers beyond subgroup analyses.
Funding for this study: The research was supported by the Hungarian Ministry of Innovation and Technology, National Research,
Development and Innovation Fund (TKP2021-EGA-23 to PH), Translational Neuroscience National Laboratory program
(RRF-2.3.1-21-2022-00011 to PH), a project grant (K131996 to PH) and the Translational Medicine Foundation. Funding for Brigitta
Teutsch was provided by the ÚNKP-22-3 New National Excellence Program of the Ministry for Innovation and Technology from the
source of the National Research, Development and Innovation Fund (to BT - ÚNKP-22-3-I-PTE-1693). The funders did not affect the
concept, data collection, analysis, or writing of the manuscript.
Ethics committee - additional information: No ethical approval was required for this systematic review with meta-analysis, as all
data were already published in peer-reviewed journals. No patients were directly involved in the design, conduct, or interpretation of
our study.
Extracellular volume fraction derived from dual-energy CT: a potential predictor for postpancreatectomy acute
pancreatitis after pancreatoduodenectomy (7 min)
920
Xiaohan Bai; Nanjing / China
Author Block: X. Bai, J. Yin, H. Shi, K. Jiang, Q. Xu; Nanjing/CN

Purpose: This study aimed to investigate the value of extracellular volume (ECV) fraction and fat fraction (FF) derived from dual-
energy CT (DECT) for predicting postpancreatectomy acute pancreatitis (PPAP) after pancreatoduodenectomy (PD).
Methods or Background: This retrospective study included patients who underwent DECT and PD between April 2022 and
September 2022. PPAP was determined according to the International Study Group for Pancreatic Surgery (ISGPS) definition. Iodine
concentration (IC) and fat fraction of the pancreatic parenchyma were measured on preoperative DECT. The ECV fraction was
calculated from iodine map images of the equilibrium phase. The independent predictors for PPAP were assessed by univariate and
multivariate logistic regression analyses and receiver operating characteristic (ROC) curve analyses.
Results or Findings: 69 patients were retrospectively enrolled (median age, 60 years; interquartile range, 55-70 years; 47 men). Of
these, nine patients (13.0%) developed PPAP. These patients had lower portal venous phase IC, equilibrium phase IC, FF and ECV
fraction, compared with patients without PPAP. After multivariate analysis, ECV fraction was independently associated with PPAP (odd
ratio [OR], 0.87; 95% confidence interval [CI]: 0.79, 0.96; p<0.001), with an area under the curve (AUC) of 0.839 (sensitivity 100.0%,
specificity 58.3%).
Conclusion: A lower ECV fraction is independently associated with the occurrence of PPAP. ECV fraction may serve as a potential
predictor for PPAP.
921
RC 2110 - Imaging of the knee
Categories: EuroSafe Imaging/Radiation Protection, General Radiology, Musculoskeletal

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:
Ana Catarina Vieira; Porto / Portugal

Ana Catarina Vieira; Porto / Portugal
Imaging the menisci pre- and post-operative (15 min)

Andrea Alcala-Galiano Rubio; Madrid / Spain
1. To describe the normal anatomy and MRI appearances of the pre-and post-operative menisci.
2. To explain the challenging findings of the post-operative meniscus.
Knee ligament (15 min)

1. To describe the normal anatomy and MRI appearances of an ACL repair and graft reconstruction.
2. To explain the imaging appearances of normal maturation and pathological conditions involving ligament repair.
3. To list the most valuable imaging findings that correlate with clinical complications.
Oedema in bone and soft tissues of the knee (15 min)

Edwin Oei; Rotterdam / Netherlands
1. To learn how to recognise bone oedema patterns and related soft tissue injuries.
Panel discussion: Are we imaging too many knees with MRI? (10 min)
922
RPS 2104 - Pulmonary nodules and lung cancer screening
Categories: Artificial Intelligence & Machine Learning, Chest, Oncologic Imaging

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Quality assurance of national lung cancer screening in Taiwan (7 min)

Yeun-Chung Chang; Taipei / Taiwan, Chinese Taipei
Author Block: Y-C. Chang, Y-S. Huang, H-H. Hsu, A. M-F. Yen, H. J. Chiou, W. P. Chan, P-C. Yang; Taipei/TW
Purpose: This study aimed to describe the preliminary experience in quality assurance of LDCT in national lung cancer screening in
Taiwan.
Methods or Background: From 1st July 2022, the National Lung Cancer Screening (LCS) program in Taiwan was launched on a
biennial basis. Enrollment criteria include heavy smokers (>30 py, 50-74 years old) and subjects with a family history of lung cancer
(male 50-74 years and female 45-74 years old) according to international guidelines and evidence from TALENT (Taiwan Lung Cancer
Screening for Never Smoker Trial) study. Modified Lung-RADS (Version 1.1) was used for categorising the LDCT interpretation results
after adjusting the size criteria of non-solid nodules >20mm diameter as Category 3 (probably benign findings). This is the first
national lung cancer screening to include non-smoker subjects with a family history in the world. Quality evaluation parameters
included radiation exposure dose, interpretation results based on modified Lung-RADS, cancer detection rate, positive predication
rate (PPV) in different groups, and proven lung cancer stages in 2 groups.
Results or Findings: Until 30th June 2023 (data estimated in June 2023), there were a total of 48,372 subjects (male 56.27%,
female 43.73%) receiving LDCT LCS in 163 hospitals. 531 lung cancers (1.11%) were identified (data from the Health Promotion
Administration, Ministry of Health and Welfare). The majority of lung cancers detected were in the early stage (stage 0 12.5%, stage I
72.58%). Quality assurance (QA) was performed according to the result of positive findings of LDCT (category 3, 4) (11%), positive
evaluation after visiting chest specialists (9.23%), cancer detection rate, radiation exposure, etc.
Conclusion: The preliminary results of QA showed the importance of screening subjects with a family history of lung cancer in
addition to heavy smokers in Taiwan.
Limitations: Data analysis is subjected to change because of different time for statistical estimation.
Funding for this study: Funding for this study was received from the Health Promotion Administration, Ministry of Health and
Welfare, Taiwan.
Impact of perifissural nodules on false positive rates in lung cancer screening with AI as the initial reader (7 min)
Daiwei Han; Groningen / Netherlands
923
Author Block: D. Han1, M. A. Heuvelmans1, H. L. Lancaster1, J. W. C. Gratama2, M. Silva3, J. Field4, M. Oudkerk1; 1Groningen/NL,
2
Apeldoorn/NL, 3Parma/IT, 4Liverpool/UK
Purpose: The primary objective of this study is to assess the false positive rate in lung cancer screening attributable to PFNs, at the
participant level. Perifissural nodules (PFNs) have been definitively established as benign in lung cancer screening (LCS) trials. Their
prevalence, accounting for 20-30% of all nodules, could significantly impact the false positive rate of lung cancer screening,
potentially leading to an unnecessary number of follow-ups. This issue is particularly relevant when AI systems are employed as the
primary readers, as they currently struggle with accurate classification of PFNs.
Methods or Background: We selected 1,253 baseline scans from the UK Lung Cancer Screening Trial based on the presence of
pulmonary nodules exceeding 15 mm³ in volume. We employed the AI-based software AVIEW to automatically detect and
volumetrically quantify solid pulmonary nodules. Subsequently, all AI-detected pulmonary nodules with a volume of ≥30 mm³
underwent visual classification by an experienced reader, distinguishing between PFNs and non-PFNs. Pulmonary nodules measuring
<100 mm³ were considered negative, while those ≥100 mm³ were categorized as positive.
Results or Findings: At the nodule level, a total of 375 pulmonary nodules were classified as PFNs, comprising 296 (78.9%)
measuring <100 mm³ and 79 (21.1%) measuring ≥100 mm³. At the participant level, out of 1253 participants, 316 (25.2%) were
found to have PFNs. Among these, 250 (20.0%) participants had only negative PFNs, while 66 (5.2%) participants had positive PFNs.
Notably, 33 (2.6%) participants with positive PFNs did not have concurrent pulmonary nodules measuring ≥100 mm³.
Conclusion: Using AI-based software as the primary reader results in a few false positive PFNs in lung cancer screening.
Limitations: The false positive rate attributable to PFNs may be influenced by the performance of AI.
Ethics committee - additional information: This study was waived by the ethics committee due to the retrospective nature of this
study.
Prevalence of bronchiectasis, airway wall thickening and emphysema in Chinese low-dose CT Screening (7 min)
Zhenhui Nie; Groningen / Netherlands
1 1 1 1 2 1 1 1 1 2
Author Block: M. Vonder , X. Yang , H. Groen , M. Oudkerk , Z. Ye , M. Dorrius , G. De Bock , Z. Nie ; Groningen/NL, Tianjin/CN
Purpose: This study aimed to assess the prevalence of lung CT findings in a general Chinese population. In lung cancer CT screening,
other lung findings like bronchiectasis, airway wall thickening and emphysema are associated with more exacerbations and
hospitalisations, as well as increased mortality rate.
Methods or Background: This study included Nelcin-B3 participants aged 40-74 years in China who received low-dose CT lung
cancer screening. Baseline characteristics of participants were described. Fleischner criteria were applied to assess bronchiectasis,
airway wall thickening, and emphysema (at least mild). The prevalence and combined prevalence for lung findings were determined.
Multivariable logistic regression analysis was performed to examine factors associated with the prevalence of these lung CT findings.
Results or Findings: In total, 978 participants (mean age 61.3 years ±6.8; 54.6% women) were included. Bronchiectasis was
identified in 9.2% of participants, 35.7% showed airway wall thickening, and 19.9% had emphysema. 2.1% of participants showed all
three CT findings. 50% of participants with emphysema were more likely to be current smokers. Multivariable logistic regression
showed age (OR=1.04; CI: 1.01-1.07), smoking (OR=3.03; CI: 1.87- 4.93), bronchiectasis (OR=1.68; CI: 1.00-2.83) and airway wall
thickening (OR=2.06; CI: 1.46-2.92) were positively associated with the presence of emphysema.
Conclusion: In the general Chinese population, at least 48% have one lung CT finding for lung cancer screening. Only 2% have all
three lung CT findings. Smoking is the strongest predictor for the presence of emphysema. The relevance of these CT findings should
be considered in future lung cancer screening guidelines.
Limitations: Firstly, no detailed clinical information could be provided. Secondly, CT diagnosis of mild bronchiectasis and airway wall
thickening remains challenging which potentially leads to false-positive diagnoses. Lastly, it only analysed patient data collected at
one singular medical centre, which limits generalisability to the whole Chinese population.
Funding for this study: Funding for this study was received from The Royal Netherlands Academy of Arts and Sciences and the
Ministry of Science and Technology of the People’s Republic of China.
Ethics committee - additional information: This study was approved by the Ethics of Biomedicine Research of Second Military
Medical University.
External validation of an AI algorithm for pulmonary nodule malignancy risk estimation on a dataset of incidentally
detected pulmonary nodules (7 min)
Renate Dinnessen; Eindhoven / Netherlands
924
Author Block: R. Dinnessen1, K. V. Venkadesh1, D. Peeters1, H. A. Piggelen-Gietema2, E. Scholten1, C. M. Schaefer-Prokop3, C. Jacobs1;
1
Nijmegen/NL, 2Maastricht/NL, 3Amersfoort/NL
Purpose: An AI algorithm for malignancy risk estimation was developed and validated on screen-detected pulmonary nodules. We
aimed to test the AI algorithm in clinical data and compare the results to the Brock model.
Methods or Background: A size-matched dataset of solid incidentally detected pulmonary nodules with a diameter range between
5-15 mm was collected, consisting of 53 malignant nodules from CT scans performed at least two months prior to a lung cancer
diagnosis, and 53 benign nodules. Differences in patient and nodule characteristics between the malignant and benign groups were
assessed. AUCs and 95% confidence intervals were determined and compared using the DeLong method. Sensitivity and specificity
were determined at a 10% malignancy risk threshold for the AI algorithm and Brock model, according to the British Thoracic Society
guidelines.
Results or Findings: No statistical difference in size was detected between the malignant and benign nodules (median [range]: 10.8
[5.8, 15.4]; 10.4 [5.8, 15.1]; respectively). Cases with malignant nodules had a significantly lower number of nodules (p=0.001). The
AI algorithm significantly outperformed the Brock model (p<0.001). AUC [95% CI] of the AI algorithm and Brock model were 0.87
[0.80-0.94] and 0.59 [0.48-0.69], respectively. The AI algorithm had a higher sensitivity (0.60 [0.46-0.74]) and specificity (0.87
[0.75-0.95]) than the Brock model (0.42 [0.28-0.56]; 0.75 [0.62-0.86]; respectively).
Conclusion: The AI algorithm outperformed the Brock model in a clinical dataset with a more heterogeneous population than a
screening population. The AI algorithm demonstrated the potential for nodule risk stratification in a clinical setting, which can aid
clinicians in decisions in nodule management, thereby potentially reducing unnecessary follow-up.
Limitations: This is a retrospective validation on a single-centre dataset. More research is needed to test the performance in larger
and multi-centre data.
Funding for this study: Funding was provided by the Dutch Cancer Society (KWF Kankerbestrijding, project number 14113).
Ethics committee - additional information: This study included data collected retrospectively from one university medical centre.
The local IRB board waived the need for informed consent because of the retrospective design and the use of anonymized data in this
study.
Deep learning for growth prediction of pulmonary ground-glass nodules (7 min)

Yingli Sun; Shanghai / China
Author Block: Y. Sun, L. Jin, C. Li, M. Li; Shanghai/CN

Purpose: The diagnosis of pulmonary ground glass nodules(GGNs) remains a challenge in clinical practice. The growth rate of
nodules is heterogeneous, although generally slower than that of other diseases. The purpose of this study is to attempt to predict the
long-term stability or growth of GGNs using deep learning based on baseline CT imaging.
Methods or Background: In this retrospective study, 575 GGNs from 456 patients were recruited. Five hundred and seventy-five
GGNs were randomized into training (70%) and validation sets (30%). A deep learning-based algorithm was developed and validated
using baseline CT imaging and clinical features. The deep learning prediction network model was compared with the traditional
radiographic features. Also, the first follow-up imaging was also added as input to improve the performance of the deep learning
model.
Results or Findings: The growth and stable groups contained 233 and 342 GGNs, respectively. Traditional radiographic features
have limited value in the prediction of growth or long-term stability of GGNs(AUC=0.70 ± 0.06). Size, density, and age were
independent predictors of GGN growth. Comparing with traditional radiographic model, our deep learning model yielded a significant
higher AUC value of 0.80 ± 0.05 (P < 0.01). The addition of first follow up CT images improved the model performance (AUC=0.84 ±
0.06).
Conclusion: We developed and validated a deep learning model to predict the natural growth pattern of GGNs basing on baseline
and first follow up CT imaging. The model achieved good performance and may provide a basis for the improvement of follow-up
management of GGNs.
Limitations: First, in this retrospective study, the variety of CT scan protocols may have affected the characteristics of GGNs and
deep features. Second, the conclusions require further validation using external large-scale datasets.
Funding for this study: Funding was received from these agencies: National Natural Science Foundation of China 61976238(Ming
Li), Science and Technology Planning Project of Shanghai Science and Technology Commission 20Y11902900 (Ming Li), Shanghai
“Rising Stars of Medical Talent” Youth Development Program “Outstanding Youth Medical Talents” SHWJRS [2021]-99 (Ming Li),
National key research and development program 2022YFF1203301(Ming Li), Cancer Society of Shanghai SACACY21C12 (Yingli Sun),
Emerging Talent Program of Huadong Hospital Grant numbers XXRC2213(Ming Li), Leading Talent Program of Huadong Hospital Grant
numbers LJRC2202(Ming Li) , and Excellent Academic Leaders of Shanghai 2022XD042(Ming Li).
Ethics committee - additional information: This study was approved ethically with the approval code: 2019K134.
Real world impact of deep-learning supported CAD for routine thoracic CT showed higher agreements with expert peer
review on management recommendations for incidental lung nodules (7 min)
Rishi Ramaesh; Edinburgh / United Kingdom
925
Author Block: R. Ramaesh1, M. P. Engbersen2, M. Javidi1, J. Rodrigues3, E. Van Beek1, M. Bernabeu1; 1Edinburgh/UK, 2Amsterdam/NL,
3
Bath/UK
Purpose: Incidental lung nodules at CT provide an opportunity for the timely detection of early-stage lung cancer. Computer-aided
detection (CAD) supported by deep learning aims to assist radiologists in the detection and further assessment of nodules. This study
investigates the effect of CAD on agreements in management recommendations between the reporting radiologist and an expert peer
reviewer.
Methods or Background: In this multicentre implementation study, participating radiologists from four centres reported on chest
CTs in adult patients without a history of malignancy or nodules, during routine practice with the CAD as ‘second reader’. The
reporters documented management recommendations concerning lung nodules sequentially, without CAD and with CAD. All cases
with nodules found either by CAD or the reporter were independently reviewed by a thoracic radiologist with at least eight years
experience. Recommendations of the reporter, with and without CAD, were compared to the recommendations of the reviewer.
Agreement between recommendations was assessed with quadratic weighted kappa, and the difference was tested with a William’s
test.
Results or Findings: Nineteen percent (237/1264) of CTs had lung nodules found and confirmed by the reviewer. In 11.4% (27/237)
of nodule cases, the reporting radiologist had changed their recommendation after CAD. The most frequent reason given for this
change was an initially missed nodule (8.4%; 20/237). The weighted kappa between the reporters’ and the reviewers’
recommendations was 0.49 and 0.55, unaided and aided by CAD (p = 0.04).
Conclusion: Aided by CAD, reporting radiologists found more incidental lung nodules and provided management recommendations
which are more in agreement with an expert reviewer, suggesting a significant contribution of CAD to the clinical management of
incidental nodules.
Limitations: Limitations include a lack of reference standard by longitudinal data or a consensus of multiple reviewers.
Funding for this study: This study was funded by the NHS AI in Health and Care Award (grant number: 2119-C25043).
Ethics committee - additional information: The study was approved by the EM-REC (Edinburgh Medical School Research Ethics
Committee). Reference number: 22-EMREC-015 dd. 14-04-2022.
Comparison of two different DL-based CAD systems regarding pulmonary nodule detection, localisation and
classification: a multi-reader study (7 min)
Nina Wiescholek; Bern / Switzerland
Author Block: A. A. Peters1, N. Wiescholek1, J. Klaus1, F. Strodka1, A. Macek1, E. C. Primetis2, D. D. Drakopoulos2, A. Christe1, L. Ebner1;
1
Bern/CH, 2Muri bei Bern/CH
Purpose: The aim of this study was to evaluate and compare the performance of two DL-CAD systems regarding detection,
localisation and classification of pulmonary nodules.
Methods or Background: The main study cohort contained 122 proven T1 tumors of the lung and was extended by 83 cases
(subsolid, n=13; solid<6mm, n=40; controls, n=30), resulting in a primary cohort of n=205. Two different DL-CAD systems analyzed
all cases. Five independent blinded readers with different experience levels (residents, n=3; seniors, n=2) performed two readout
sessions, first stand-alone and then with access to the results of one of the DL-CAD systems. Two readers used software 1 and the
three readers used software 2 and scored nodule size, density and localisation. LungRADS categories were calculated and compared.
Results or Findings: After application of the eligibility criteria, the final cohort consisted of 198 subjects with 221 pulmonary
nodules. Residents' mean detection rate increased from 64% to 77% (p<0.001) using the respective DL-CAD (table 2), while the
seniors’ detection rates did not improve (p=0.25). Regarding the correct localization of the nodules, the residents' rates for lobar
(73% vs. 77%; p<0.001) and segmental (64% vs. 68%; p<0.001) nodule localisation improved significantly, the seniors showed no
significant benefit. Regarding software comparison, software 2 lead to a slightly higher increase in detection rates (software 1, 80% to
86% and software 2, 67% to 77%; both p<0.001). Both systems showed no significant effect on the rate of correct LungRADS
classification.
Conclusion: Less experienced readers have more benefits from using DL-CAD systems regarding detection and localisation of
pulmonary nodules. There is no effect on correct LungRADS classification. Both systems performed comparably, software 2 lead to a
higher increase in detection rates.
Limitations: Selection bias (high cancer prevalence). Nodule size groups categorized instead of exact measures.
Funding for this study: No funding was obtained for the current study.
Ethics committee - additional information: This study was approved by the Kantonale Ethikkommision (KEK) Bern.
Incidental detection of ground-glass nodules and primary lung cancer in patients with primary breast cancer: incidence
and long-term follow-up on chest CT (7 min)
Hye Sun Ryu; Seoul / Korea, Republic of
926
Author Block: H. S. Ryu, H. N. Lee, J. I. Kim, J. K. Ryu, Y. J. Lim; Seoul/KR
Purpose: Patients with breast cancer have a higher risk of developing lung cancer than the general population. The study aimed to
evaluate the incidence of GGN and risk factors for GGN growth in patients with breast cancer and to evaluate the incidence and
pathologic features of lung cancer.
Methods or Background: We retrospectively reviewed the clinical data and chest CTs of 1384 patients diagnosed with breast
cancer who underwent chest CT between January 2008 and December 2022. We evaluated the incidence and size change of GGNs
during follow-up and identified independent risk factors for their growth using multivariate analysis. Furthermore, the incidence and
pathologic features of lung cancer were also evaluated.
Results or Findings: We detected persistent GGNs in 69 of 1384 (5.0%) patients. The initial diameter of GGNs was 6.3 ± 3.6 mm on
average, with primarily (85.5%) pure GGNs. Among them, 27 (39.1%) exhibited interval growth with a median volume doubling time
of 1006.0 days (interquartile range, 622.0–1528.0 days) during the median 959 days (interquartile range, 612.0–1645.0 days) follow-
up period. Older age (P = 0.026), part-solid nodules (P = 0.006), and total number of GGNs (≥2) (P = 0.007) were significant factors
for GGN growth. Lung cancer was confirmed in 13 of 1384 patients (0.9%), all with adenocarcinoma, including one case of minimally
invasive adenocarcinoma. The cancers demonstrated a high rate of epidermal growth factor receptor mutation (69.2%).
Conclusion: Persistent GGNs in breast cancer patients with high-risk factors should be monitored for early detection and treatment
of lung cancer.
Limitations: This retrospective study was conducted at a single centre with a small sample size, manual measurement of GGNs was
subject to errors, and chest CT was not dedicated to automated volume measurement.
Ethics committee - additional information: The study was approved by institutional review board of Kyung Hee University
Hospital at Gangdong (2023-02-016) and informed consent was waived owing to the retrospective nature of the study.
Radiologic features of nodules attached to the mediastinal or diaphragmatic pleura on chest CT (7 min)
Yeqing Zhu; New York / United States
Author Block: Y. Zhu, R. Yip, D. Yankelevitz, C. I. Henschke; New York, NY/US

Purpose: The aim of this study was to determine whether NCNs attached to mediastinal or diaphragmatic pleura (M/DP-NCNs) having
the same features as fissural or costal pleura NCNs at low-dose CT can follow the same recommendations.
Methods or Background: This retrospective study reviewed chest CTs from patients in two separate databases. Group A included
1451 patients who had lung cancer which was first present as solid nodule with size 3.0-30.0 mm. Group B included 345 consecutive
screening participants who had at least 3 follow-up scans and at least one solid nodule with size 3.0-30.0 mm. Radiologists reviewed
CTs for identifying solid M/DP-NCNs, defined as nodules 0 mm in distance from the mediastinal or diaphragmatic pleura, and recorded
average diameter, margin, and shape. General descriptive statistics were used.
Results or Findings: Among 1451 lung cancer patients, 163 patients (median age, 68 years [IQR: 61.5–75.0 years]; 92 males) had
164 malignant M/DP-NCN, 3.0-30.0 mm in diameter. None of the 164 malignant M/DP-NCN had smooth margins and triangular or LOS
shapes. Among 345 consecutive lung cancer screening participants, 146 participants (median age, 65 years (IQR: 59–71 years]; 81
females) had 240 M/DP-NCNs with size 3.0-30.0 mm in diameter. None of the M/DP-NCN with smooth margins and triangular or LOS
shapes were malignant after median follow-up of 57.8 months (IQR: 46.3–68.1 months).
Conclusion: For solid M/DP-NCNs with smooth margins and triangular or LOS shapes on low-dose CT recommend annual follow-up,
instead of immediate workup, may be appropriate.
Limitations: Although all diagnosed lung cancers were reviewed for this study, this study did not provide an estimate of the overall
prevalence of M/DP-NCNs, including benign ones.
Ethics committee - additional information: All participants signed institutional review board–approved, Health Insurance
Portability and Accountability Act compliant consent forms.
Construction and validation of a risk score system for diagnosing invasive adenocarcinoma presenting as pulmonary
pure ground-glass nodules: a multicentre cohort study in China (7 min)
Qing Cheng Meng; Zhengzhou / China
927
Author Block: Q. C. Meng, P. Gao; Zhengzhou/CN
Purpose: Pure ground-glass nodule (pGGN) invisibility drives clinical intervention. Radiomics and radiogenomics aid pGGN diagnosis
but lack of standardised acquisition parameters, reproducibility and inconsistent methods. We aim to evaluate a risk score system for
diagnosing invasive adenocarcinoma presenting as pGGN .
Methods or Background: Seven hundred and seventy-two pGGNs from 707 individuals were grouped into training (509
patients/558 observations) and validation (198 patients/214 observations) sets. A test set with 143 observations was also analysed.
The quantitative parameters were obtained using AI. The positive pGGN cutoff score was ≥3. Risk score systems3 were calculated as
the history of carcinoma*1+chronic obstructive pulmonary disease (COPD)*1 + long diameters*1 + volume of nodule*1 + mean CT
values*1 + type II vascular supply sign*1 or type III*2 + other variables of radiographic characteristics*1. The risk score system and AI
model were evaluated using areas under the receiver operating characteristics curve (AUCs), accuracy, sensitivity, specificity, and
positive predictive values.
Results or Findings: Risk score system 3 (AUC, 0.840) performed better than the AI model (AUC, 0.553), risk score system 1 (AUC,
0.802, and risk score system 2 (AUC, 0.816), with 88.0% (0.850–0.904) accuracy, 95.6% (0.932–0.972) PPV, 89.6% (0.864–0.920)
sensitivity, and 80.6% (0.717–0.872) specificity in the training sets. Risk score system 3 yielded the best performance in the
validation and test set, with AUCs of 0.769 and 0.801.
Conclusion: The risk scoring system based on AI-based quantitative image parameters, clinical features, and radiographic
characteristics can effectively predict the invasive adenocarcinoma of pulmonary pGGNs.
Limitations: First, the study was retrospective, with some selection bias. Second, the improved AI algorithm may enhance the
diagnosis of the invasiveness of pGGNs in clinical practice, and iterative upgrading of the algorithm is needed in the future.
Funding for this study: Funding of this study was received from the Key project of Medical science and Technology of Henan
Province in China (No: SBGJ202102057).
Ethics committee - additional information: This multi-centre, retrospective cohort study was conducted in accordance with the
Declaration of Helsinki, and the protocol was approved by the The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer
Hospital Medical and the The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital Ethics Committee (2021-
KY-0022).
Differentiation of malignant grade of non-mucinous pulmonary adenocarcinomas in subsolid nodules using enhanced
dual-energy computed tomography (7 min)
Changjiu He; Chengdu / China
Author Block: C. He, J. Liu, P. Zhou; Chengdu/CN

Purpose: The aim of this study was to explore the ability of dual-energy computed tomography (DECT) in distinguishing malignant
grade of non-mucinous pulmonary adenocarcinomas (NMPAs) in subsolid nodules.
Methods or Background: Three hundred and forty-three subsolid nodules from 302 consecutive patients, who underwent dual-
phase enhanced DECT, were retrospectively enrolled. The nodules were pathologically diagnosed with non-mucinous adenocarcinoma
in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive pulmonary adenocarcinoma (IPA) and divided into low-risk (AIS,
MIA, and well-differentiated IPA; n = 196) and high-risk groups (moderately/poorly differentiated IPA; n = 147) based on the novel
grading system. The following features were analysed: gender, age, nodule type, diameter, volume, mean attenuation, and DECT
quantitative parameters including electron density (Rho), iodine concentration (IC), normalized iodine concentration (NIC), and slope
of the spectral curve (λhu) in arterial and venous phases. The combined model was constructed with multivariable logistic regression.
The diagnostic performance was evaluated using area under curve (AUC).
Results or Findings: The independent factors for distinguishing high-risk group from low-risk group included age, nodule type,
diameter volume, mean attenuation, Rho in arterial phase (Rho_A), and Rho in venous phase (Rho_V) (all P < 0.05). The AUC of
combined model including nodule type, diameter, and Rho_V was 0.836, which outperformed age, nodule type, diameter, volume,
mean attenuation, Rho_A, and Rho_V (all P < 0.05).
Conclusion: The combined model including nodule type, diameter, and DECT-derived Rho_V could differentiate malignant grade of
NMPAs in subsolid nodules.
Limitations: Our study had several limitations. First, this was a single-centre study, therefore, the reliability of our results need to be
confirmed by external validation datasets. Second, the VNI rather than true non-enhanced image were used to assess the nodule
type, nodule size, and nodule attenuation. Third, this study did not include comprehensive morphological characteristics.
Funding for this study: This research was funded by the National Natural Science Foundation of China (grant number 82202141),
the Fundamental Research Funds for the Central Universities (grant number ZYGX2021YGCX017), and the Sichuan Science and
Technology Program (grant number 2022YFSY0006).
Ethics committee - additional information: The study was approved by the ethics committee of Sichuan Cancer Hospital.
928
JIIQ - Radiology Feud
Categories: Abdominal Viscera, Cardiac, Education, Emergency Imaging, General Radiology, Neuro
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Martin Reim; Tartu / Estonia
Game Show Host (90 min)

The Simpsonographers Team

Antonella Del Gaudio; Durham, NC / United States
Nick Janssen; Berchem / Belgium
Carlos Manuel Baracaldo Silva; Valencia / Spain
The Radiogriffins Team

Mariana Andreea Calota; Paris / France
Davis Simanis Putrins; Riga / Latvia
Luca Dóra Laszip; Budapest / Hungary
Closing words
929
HW 21Cd - Which cardiac findings every general radiologist should know
Categories: Cardiac, Chest, General Radiology, Imaging Methods

ETC Level: LEVEL II
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:


Sue Thomas; Bournemouth / United Kingdom
1. To become familiar with typical and atypical imaging findings of cardiac findings to be reported in non-cardiac CT and MRI exams.
3. To discuss the limits and technical drawbacks of non-cardiac CT and MRI for the identification and characterisation of cardiac
findings.
4. Understand the additional diagnostic value of ECG gating.
5. To learn how to report cardiac findings in non-cardiac CT and MRI.
930

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1


Tutors (6 min)
931
ST 20 - SOLACE Insight: Strengthening Lung Cancer Screening Across Europe
Date: March 2, 2024 | 16:00 - 16:30 CET
The SOLACE project assesses the current state of policies, needs, and best practices of Lung Cancer Screening (LCS) in EU member
states and produces a comprehensive guideline and implementation package covering all steps of the lung cancer screening
pathway: evidence-based guidelines, technical papers, documents regarding quality assurance, methodology, benefit-harm balance,
and cost-effectiveness. This package is used to showcase the new implementation methodology suggested. Moreover, SOLACE
designs, plans, and rolls out three pilot projects in 10 member states with more than 12,000 participants with the aim to increase
participation considering: gender aspects, inequalities regarding hard-to-reach populations (social, ethnic, geographical), and higher-
risk individuals.
Moderator:
Interview (30 min)

932
OF 21R - Unlocking research potential: empowering radiographers to make a difference through

funded projects
Categories: Education, Management/Leadership, Professional Issues, Radiographers, Research

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
This session is dedicated to fostering a culture of research engagement among radiographers. Comprising three illuminating talks,
this session offers invaluable insights into the pursuit of funded projects, the associated benefits, challenges, and strategies to
transform setbacks into achievements in the realm of radiography research. This empowering session should be of interest to
radiographers, health professionals, educators and researchers. By providing practical guidance, addressing challenges, and sharing
inspirational stories of success, this session aims to equip attendees with the knowledge and motivation needed to actively contribute
to research endeavours and make a meaningful difference in the field of medical imaging and radiotherapy.
Moderator:
Claudia Sa Dos Reis; Lausanne / Switzerland

Claudia Sa Dos Reis; Lausanne / Switzerland
How to get involved in funded projects (16 min)

Benefits, challenges and pitfalls for seeking research funding (16 min)
Ian C. Simcock; St Albans / United Kingdom
Turning rejection into success:insights from the SAFE Europe Project (16 min)
933
EIBIR 21 - EUCAIM at the forefront: innovating radiology with Europe's cancer image infrastructure for
artificial intelligence (AI)
Categories: Artificial Intelligence & Machine Learning, Oncologic Imaging, Professional Issues, Research
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

The EUCAIM project: why is it needed? (12 min)

1. To learn the bases of this leading European hub for the cancer research community and AI innovation.
2. To appreciate the relevance of unifying fragmented and temporally limited datasets into an extensive Atlas of Cancer Images.
3. To understand how unlocking the potential of big data in oncology will improve the reproducibility and robustness of AI-based
prediction models based on medical imaging.
The European Cancer Imaging Initiative (10 min)

Saila Rinne; Luxembourg / Luxembourg
EUCAIM and the EU Cancer Imaging Initiative: a look at the future (12 min)
Mario Aznar; Madrid / Spain
1. To understand the fit of EUCAIM in the current and future landscape of EU programmes, initiatives and legal frameworks focussed
on widening the health data space.
2. To share a vision of how EUCAIM will operate in the future, a legal entity acting as the hub for oncology research communities
fostering innovation in the EU.
The view on EUCAIM: the ESR perspective (12 min)

1. To understand the ESR perspective on EUCAIM.

2. To learn about the ESR's vision for large infrastructures.
3. To determine how will EUCAIM influence and impact radiology.
EUCAIM's Atlas of Cancer Images: present and future (12 min)

Ignacio Blanquer; Valencia / Spain
1. To understand the concept of EUCAIM as a federation of providers and communities.

2. To learn the current developments of the EUCAIM's Atlas of Cancer Images, its architecture and functionality.
3. To be informed about future developments and workplan.
934
How to collaborate across Europe? (12 min)
Katrine Riklund; Umeå / Sweden
1. To learn about possibilities to collaborate at distance.

2. To appreciate a common structure for collaboration.
3. To understand differences and similarities between countries.
Discussion (15 min)
935
PC 21 - Nothing about me without me: communication skills required for radiologists when
communicating with their patients
Categories: Breast, Interventional Radiology, Oncologic Imaging, Paediatric, Professional Issues

ETC Level: LEVEL II
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

Communicating with patients in the context of breast imaging (12 min)

Ruchira Sinnatamby; Cambridge / United Kingdom
1. To present specific points to be observed when interacting with patients in the context of mammography, breast ultrasound and
breast MRI, and image-guided biopsy.
2. To explain how to communicate unexpected findings.
3. To stress the importance of coordination between radiologists and treating physicians.
How to communicate uncertainty to patients with BIRADS 3 or 4 and share decision-making? (12 min)
1. To understand the impact of patient communication before, during and after radiology procedures for BIRADS 3 or 4.
2. To learn practical steps to help patients understand the procedures and what to expect concerning the results.
Paediatric radiology: communicating with parents (24 min)

Julian H.W. Jurgens; Hamburg / Germany
Ola Kvist; Stockholm / Sweden
1. To discuss the specific context of paediatric imaging in the presence of the parents.
2. To explain some key points to observe when communicating with parents about imaging findings.
Interventional radiology (12 min)

Ruediger Schernthaner; Vienna / Austria
1. To explain the indication and the procedure of obtaining informed consent.

2. To explain the result of the procedure and, if necessary, to disclose adverse events.
Panel discussion: How should we best teach clinical communication skills during radiology training? (25 min)
936
E³ 2120 - Complications and follow-up imaging after percutaneous ablation
Categories: Abdominal Viscera, Chest, Interventional Oncologic Radiology, Musculoskeletal, Oncologic Imaging
Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
The session summarises the most common expected complications and their management after thermal ablation in several different
organs. Moreover, the audience will familiarise themselves with the clinical and imaging follow-up needed after ablation.
Moderator:

Complications and follow-up imaging after lung ablation (24 min)

Georgia Tsoumakidou; Lausanne / Switzerland
Complications and follow-up imaging after liver and kidney ablation (24 min)
Fernando Gomez Muñoz; Valencia / Spain
Complications and follow-up imaging after bone and soft-tissue ablation (24 min)
Panel discussion: Which actions should be taken to prevent complications? (13 min)
937
SF 21b - Cardiac: clinical use of T2 mapping techniques
Categories: Cardiac, Imaging Methods, Research, Translational Imaging, Vascular

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:
Julian Alexander Luetkens; Bonn / Germany

Julian Alexander Luetkens; Bonn / Germany
T2 mapping in myocardial disease (20 min)

Moritz Christian Halfmann; Mainz / Germany
1. To define imaging protocols for parametric T2 mapping.

2. To discuss clinical applications of parametric T2 mapping for myocardial tissue characterisation.
3. To reflect on current challenges and future directions for parametric T2 mapping.
Assessment of intracardiac shunts (20 min)

Aurelio Secinaro; Roma / Italy
1. To list anatomical landmarks of main L-R cardiac shunts and their clinical relevance.
2. To define the pathophysiological principle of intracardiac shunts and how they can influence T2 signal.
3. To describe potential clinical application of T2 mapping to detect and quantify cardiac shunts.
BOLD imaging (20 min)

Kady Fischer; Bern / Switzerland
1. To obtain an introduction into the basics of BOLD/OS imaging.

2. To describe the advantages and potential applications of BOLD/OS imaging in cardiovascular cohorts in comparison to other tissue
characterisation techniques.
3. To understand the limitations and lack of standardisation keeping BOLD/OS from entering routine clinical protocol.
Panel discussion: T2 mapping: all I need for cardiac MRI? (25 min)
938
E³ 2123 - Head and neck
Categories: Head and Neck

Date: March 2, 2024 | 16:00 - 17:30 CET
CME Credits: 1.5
Moderator:

Temporal bone and skull base (28 min)

1. To differentiate between the anatomy, normal variants, and congenital disorders of the temporal bone.
2. To understand the causes and imaging features of hearing and vestibular disorders.
3. To describe the imaging presentation of the most common tumours of the skull base.
Nose, paranasal sinuses, and nasopharynx (28 min)

Tilak Das; Cambridge / United Kingdom
1. To describe the anatomy and normal variants of the nose, paranasal sinuses, and nasopharynx.
2. To differentiate between the imaging features of acute and chronic inflammatory changes of the nose and paranasal sinuses.
3. To understand the imaging features of benign and malignant tumours of the nose, paranasal sinuses, and nasopharynx.
Oral cavity, oropharynx, hypopharynx and larynx (28 min)

1. To describe the normal imaging anatomy of the oral cavity, oropharynx, hypopharynx, and larynx.
2. To understand the imaging features of tumours of the oral cavity and oropharynx.
3. To understand the imaging features of tumours of the hypopharynx and larynx.
939
BS 21 - Making a difference: towards person-centred imaging, treatment, and care
Categories: Education, Evidence-Based Imaging, Multidisciplinary, Professional Issues, Radiographers

Date: March 2, 2024 | 16:00 - 17:00 CET
CME Credits: 1
Moderator:
ELENI GEORGIADOU; Keratsini / Greece

ELENI GEORGIADOU; Keratsini / Greece
Celebrating diversity in radiography: tips to support LGBTQ+ peoples' needs (14 min)
Gareth Robert Hill; Dundee / United Kingdom
1. To increase awareness and understanding of the unique challenges faced by LGBTQ+ individuals in the field of radiography.
2. To enhance knowledge of practical strategies for creating an inclusive and supportive environment.
3. To improve cultural competence and communication skills when interacting with LGBTQ+ individuals.
Enhancing older peoples' experiences of imaging and radiotherapy (14 min)

1. To describe current knowledge and attitudes towards dementia patients, including how these may be influenced by age,
experience, gender, qualification, grade and workplace.
2. To describe student radiographers'/radiation therapists' opinions towards patients with dementia.
3. To demonstrate the benefits of an interprofessional collaboration workshop for students working with patients with dementia.
Eliminating weight stigma and enhancing the experience of overweight or obese patients (14 min)
Daniele Di Feo; Florence / Italy
1. To explore dose optimisation in overweight or obese patients.

2. To discuss radiological technologies that can influence the management and experience of obese patients.
3. To outline suggestions of how radiographers can optimise radiological protocols in overweight or obese children.
It's time to talk about breast density in mammography (14 min)

Deborah Mizzi; Msida / Malta
1. To describe the concept of breast density and its significance in mammography.

2. To recognise the challenges associated with breast density in mammography screening.
3. To explore the advancements and options for supplemental screening in dense breasts.
940

Date: March 2, 2024 | 18:00 - 18:15 CET
Moderators:
Interview (30 min)
941

Date: March 3, 2024 | 07:50 - 08:00 CET
Moderators:
Interview (30 min)

942
RPS 2215 - Aortic aneurysm and dissection research: from classification to management

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Katharina Mueller-Peltzer; Freiburg / Germany
Uncommon anatomy, uncommon ailment: right-sided aortic arch with aberrant retroesophageal subclavian artery and
dissecting aortic aneurysms in youth (7 min)
Ahmed Monier Sherif; Abu Dhabi / United Arab Emirates
Author Block: A. M. Sherif; Abu Dhabi/AE

Purpose: To pose the question of whether there exists any association between right-sided aortic arch and aberrant left
retroesophageal subclavian artery and dissecting aortic aneurysms in young patients. To encourage innovative thinking and the
exploration of new theories regarding potential links between this unique anatomical variant and aortic pathology in the absence of
traditional risk factors. To discuss imaging findings and management strategies for these unique patients.
Methods or Background: We present three compelling cases of young patients, aged 25 to 40, each afflicted with dissecting aortic
aneurysms involving the aortic arch. They exhibited no known predisposing factors for aortic pathology, such as hypertension or
connective tissue disorders. What distinguishes these cases is the presence of a right-sided aortic arch with an aberrant
retroesophageal left subclavian artery. The literature has yet to establish any association between right-sided aortic arch and
dissecting aneurysms.
Results or Findings: Imaging studies consistently revealed a right-sided aortic arch and an aberrant retroesophageal left subclavian
artery in all cases. We pose the question of whether this unique anatomical configuration may introduce hemodynamic stress, thereby
potentially contributing to the development of aortic dissection in these patients.
Conclusion: This series invites exploration of novel ideas and associations, opening doors for fresh discussions and theories
surrounding the possible connection between right-sided aortic arch with aberrant retroesophageal left subclavian artery and
dissecting aortic aneurysms in young, otherwise healthy patients. While the exact etiological mechanisms remain elusive, this
observation underscores the importance of innovative thinking and further research to elucidate the embryological origins and
hemodynamic implications of this distinctive anatomical variant. Additionally, it underscores the need for comprehensive imaging and
multidisciplinary management approaches to guide patient care in this intriguing context.
Ethics committee - additional information: The study is educational and so no ethical approval was required.
Comparative analysis of computational workstations for stent planning of thoracic aortic aneurysms (7 min)
Vitali Koch; Frankfurt a. Main / Germany
943
Author Block: V. Koch, L. D. Grünewald, S. Mahmoudi, S. Martin, C. Booz, J. Gotta, K. Eichler, T. Vogl, T. Gruber-Rouh; Frankfurt a.
Main/DE
Purpose: The aim of this study was to evaluate the accuracy of three different computer workstations in measuring thoracic aortic
aneurysms (TAAs) both in live patients and in laboratory settings, using either pre-interventional computed tomography angiography
scans (CTAs) or a specially designed phantom model.
Methods or Background: This retrospective study involved 23 patients with confirmed TAAs detected during routine CTA scans.
Alongside measurements of the phantom, an experienced radiologist, unaware of the true dimensions, assessed TAA sizes using three
different workstations in two separate rounds. Measurement accuracy was determined by calculating measurement errors, and a
Pearson correlation analysis was conducted.
Results or Findings: Measurements obtained from the Siemens workstation had a deviation of 3.54% (range, 2.78 to 4.03%;
p=0.14) from the true size, while those from General Electric deviated by 4.05% (range, 1.46 to 7.09%; p<0.0001), and TeraRecon
had a deviation of 4.86% (range, 3.22 to 6.45%; p<0.0001). Siemens demonstrated the highest precision among the workstations,
despite having the most variable measurements (with a range of 4.46%). TeraRecon showed the least variability (with a range of
2.83%) but had the largest deviation from the true phantom size. General Electric's workstation exhibited a variability range of 2.94%.
Siemens displayed the strongest correlation between measurements from the 1st and 2nd rounds (r=0.898), followed by TeraRecon
(r=0.799) and General Electric (r=0.703). Repeating measurements reduced processing times by 40% with General Electric, 20% with
Siemens, and 18% with TeraRecon.
Conclusion: In summary, all three workstations demonstrated accurate dimension assessment in the majority of cases with
consistently high reproducibility. This ensures precise pre-interventional planning for thoracic endovascular aortic repair.
Limitations: Retrospective study. Vendor-specific setup.
Ethics committee - additional information: The institutional ethical review board approved this retrospective study that complies
with the Declaration of Helsinki. The need for written informed consent was waived.
Assessment of AAA morphology using relaxation-enhanced angiography without contrast and triggering (REACT):
comparison with CTA (7 min)
Wen Zeng; Chengdu / China
Author Block: W. Zeng, Z. Li, C. Xia; Chengdu/CN

Purpose: The aim of the study was to assess the accuracy of relaxation-enhanced angiography without contrast and triggering
(REACT) for measuring abdominal aortic aneurysm.
Methods or Background: Contrast-enhanced computed tomography (CT) is usually the gold standard for AAA diagnosis. A novel
REACT is proposed, which can be used to diagnose aorta diseases. In this study, patients were recruited consecutively and received
REACT using 3T MRI and CTA. Two radiologists independently assessed scans for the morphology of AAA, including the diameter of
neck of AAA, the maximum diameter of AAA, and the thickness of lateral thrombus.
Results or Findings: A total of 15 consecutive AAA patients (53.27±13.68 years, 11 males) were included. Considering CT as the
standard of reference, REACT provided a good correlation. There was no significant difference between REACT and CT in measuring
the diameter of neck of AAA(25.30±4.71 mm vs. 26.93±5.06 mm, P=0.028) and the maximum diameter of AAA (46.75±19.43 mm vs.
48.52±20.46 mm, P=0.009). When measuring the thickness of lateral thrombus, REACT slightly underestimated it than CT (6.48±2.17
mm vs. 7.86±3.18 mm, P=0.073). The interobserver agreement for contrast-enhance CT and REACT is relatively good (0.846 for neck
of AAA, 0.751 for maximum diameter of AAA, and 0.56 for thickness of lateral thrombus). Further, REACT achieved a substantial
accordance with contrast-enhanced CT regarding the morphology of AAA [Cohen’s Kappa 0.776 (95% CI: 0.88–0.93)].
Conclusion: REACT has high accuracy in AAA morphological measurements without any contrast agent. This novel sequence is also
sensitive to the thickness of thrombosis, which provides a new diagnostic method for AAA patients.
Limitations: First, this was a single-center study with a relatively small number of patients enrolled. Second, the readers were not
blinded to the CTA and REACT, which might have influenced observer bias.
Funding for this study: This study was supported by the National Key R&D Program of China (2022YFC2009905)
Ethics committee - additional information: All the subjects provided written informed consent, and this study was approved by
the ethics committee of West China Hospital.
Revealing aortic health: a radiomic approach for early acute aortic syndrome detection on non-contrast computed
tomography (7 min)
Chiara Zanon; Padua / Italy
944
Author Block: C. Zanon, A. Toniolo, A. Spertino, C. Bini, G. Cabrelle, M. Antonello, E. Quaia, A. Pepe; Padua/IT
Purpose: Acute Aortic Syndrome (AAS) is a life-threatening condition that frequently leads to a high mortality rate for delayed
diagnosis. Routine computed tomography (CT) screening programs depend mainly on subjective and qualitative analyses, but Texture
Analysis (TA) can extract quantitative data that are not visible to the human eye.
This study aimed to evaluate the use of Aortic TA parameters on non-contrast computed tomography (CT) to differentiate between
patients with AAS and healthy subjects.
Methods or Background: We retrospectively included 22 patients (mean age, 58 ± 11 years; 13 (59%) male) with a confirmed
diagnosis of ASS (12 Type B dissection, 6 intramural hematomas, 4 penetrating aortic ulcers) who underwent emergency CT. Two
expert radiologists used specific software to draw aortic wall volumes of interest (VOIs) on unenhanced CT images (axial plane:
ascending aorta, isthmus, descending aorta) in patients with AAD and a healthy control group (n=10). We extracted 118 texture
parameters (first- and second-order parameters). The Mann-Whitney Test was used, and a p-value of 0.05 was considered statistically
significant.
Results or Findings: The groups showed statistically significant differences in 17 features, 4 in the ascending aorta (p < 0.04), 4 in
the isthmus (p <0.02), and 9 in the thoracic aorta (p<0.04).
Conclusion: Texture Analysis may provide valuable insights into the possibility of detecting early aortic wall deterioration to prevent
AAS.
Limitations: This study is retrospective and involves a limited number of patients.
Ethics committee - additional information: The institutional review board of the University of Padua approved this retrospective
study.
Dual-source photon-counting CTA of the thorax: impact of low energy virtual monoenergetic imaging on image quality,
vascular contrast and diagnostic assessability (7 min)
Christian Booz; Frankfurt a. Main / Germany
Author Block: C. Booz1, I. Yel1, V. Koch1, L. D. Gruenewald1, L. S. Alizadeh1, S. Martin1, T. J. Vogl1, D. P. Overhoff2, S. Waldeck2;
1
Frankfurt a. Main/DE, 2Koblenz/DE
Purpose: The aim of this study was to evaluate the impact of low energy VMI+ reconstructions on quantitative and qualitative image
quality, vascular contrast and diagnostic assessability of thoracic arteries in photon-counting CTA.
Methods or Background: A total of 120 patients (66 male) who had undergone dual-source photon-counting CTA scans of the
thorax were retrospectively analyzed. Standard 120 kV CT images and low keV VMI+ series from 40 to 100 keV with an interval of 15
keV were reconstructed. Quantitative analyses included evaluation of vascular CT numbers, signal-to-noise ratio (SNR) and contrast-
to-noise ratio (CNR). CT number measurements were performed in the ascending and descending aorta, the aortic arch, the common
carotid artery, the subclavian artery and the coronaries. Qualitative analyses were performed by three board-certified radiologists
independently using five-point scales to evaluate image quality, vascular contrast and diagnostic assessability of thoracic arteries.
Results or Findings: Mean attenuation, CNR and SNR values were highest in 40 keV VMI+ reconstructions (HU, 1205 ± 11; CNR, 29
± 7; SNR, 30 ± 9) followed by 55-keV VMI+ reconstructions (HU, 679 ± 8; CNR, 23 ± 6; SNR, 24 ± 7); all three mean values at these
keV levels were significantly higher compared with the remaining VMI+ series and standard 120 kV CT series (HU, 169 ± 7; CNR, 19 ±
5; SNR, 27 ± 7) (p<.0001). The qualitative analysis showed highest rating scores for 55 keV VMI+ reconstructions followed by 40 keV
and 70 keV VMI+ series with a significant difference compared to standard 120 kV CT images series (p<.0001).
Conclusion: Low keV VMI+ reconstructions at a level of 40-55 keV significantly improve image quality, vascular contrast and the
diagnostic assessability of the thoracic arteries compared with standard CT series in photon-counting CTA.
Limitations: Single-center retrospective study design
Funding for this study: No funding was receidved for this study.
Ethics committee - additional information: The local IRB approved this study.
Early insights into the use of dynamic computed tomography angiography for classifying challenging endoleaks (7 min)
Ákos Bérczi; Budapest / Hungary
945
Author Block: Á. Bérczi, Z. Jokkel, A. Jermendy, M. Berczeli, F. Szablics, Z. Szeberin, B. Merkely, C. Csobay-Novák; Budapest/HU
Purpose: The aim of this study was to implement and validate our dynamic CTA (d-CTA) protocol, designed to characterize endoleaks
that were previously unclassified or of uncertain origin with triphasic CTA approach.
Methods or Background: Between January 2022 and January 2023, d-CTA scans were performed on specific patients who had
uncertain endoleak types or showed ongoing expansion of the aneurysm sac following endovascular aneurysm repair. A total of 12-18
scans were completed, with two different scan protocols: one with 16 cm cranio-caudal coverage and 4.4 seconds between
acquisitions, and another with 8 cm coverage and acquisition times ranging from 0.8 to 2 seconds. The scanning protocol was
individually tailored based on the region of interest (ROI) and suspected endoleak types. Quantitative data analysis was focused on
changes in Hounsfield units within a defined ROI across multiple contrast-enhanced scans.
Results or Findings: A total of 18 patients underwent d-CTA scans, out of which 15 individuals met the inclusion criteria. In two
cases type V endoleaks were reclassified as type II by identifying inflow arteries. Additionally, for five patients with confirmed type II
endoleaks, the quantitative analysis identified the inflow vessels as either the inferior mesenteric artery (IMA) or lumbar arteries.
Furthermore, indistinct types of I, II, and/or III endoleaks in seven patients were successfully characterized. In one patient with
suspected type II endoleak from IMA on triphasic CTA, the d-CTA scans showed an occluded IMA with no endoleak a month later.
Conclusion: D-CTA has the potential to serve as a valuable supplement to the standard triphasic CTA follow-up, aiding in the
characterization of difficult-to-diagnose endoleaks and offering valuable information for the development of precise and targeted
treatment strategies.
Limitations: The main limitation of the study is the small sample size.
Ethics committee - additional information: The present study has been approved by the regional and institutional committee of
science and research ethics of our University (SE-RKEB 96/2023).
Value of dark-blood computed tomography angiography for the assessment of aortic wall disease in patients with
acute aortic syndrome (7 min)
Na Li; Wuhan / China
Author Block: N. Li, W. Sun, Q. Luo, S. Gui, H. Shuishi; Wuhan/CN

Purpose: The objective of this study was to assess the capability of dark-blood images derived from contrast-enhanced dual-energy
CT scans in visualizing aortic wall disease in patients with acute aortic syndrome (AAS).
Methods or Background: Ninety-seven patients with computed tomography angiography (CTA)-confirmed AAS, including acute
aortic dissection (n = 35), intramural hematoma (n = 31), and atherosclerotic aortic ulcers (n = 31), were enrolled using IQon spectral
CT (Philips Healthcare). Dark-blood images, polyenergetic images (PI), and 40–120 keV virtual monoenergetic images (VMI) were
reconstructed. Objective image analysis was performed on PI and VMI images using quantitative values, such as image signal-to-noise
ratio (SNRimage), contrast-to-noise ratio (CNRimage), lesion signal-to-noise ratio (SNRlesion), wall-to-fat ratio (WFR), and wall-to-
lumen ratio (WLR). Subjective image analyses were conducted by two readers who assessed the diagnostic quality on 4-point Likert
scales and the inner/outer vessel wall conspicuity on 5-point Likert scales for dark-blood images, PI, VMI40keV, VMI70keV, and
VMI100keV.
Results or Findings: The SNRimage, CNRimage, SNRlesion, WFR, and WLR gradually decreased as the keV levels increased, while
the highest values were found in VMI40keV (all P < 0.05). The subjective scores of the different images were statistically significant (P
< 0.05), with the dark-blood image ranked highest in Likert scores regarding diagnostic quality and both inner and outer wall
conspicuity, followed by VMI40keV, VMI70keV, and PI, with the lowest scores at VMI100keV. Inter-observer agreement of dark-blood
image ranged between moderate and substantial (weighted kappa values 0.60-0.68).
Conclusion: Similar to previous studies, low keV levels of VMI enhanced the identification and outlining of wall lesions. Compared to
other enhanced images derived from spectral detector CTA, dark-blood images improve the subjective identification of aortic wall
lesions and enhance the contrast of wall lesions.
Limitations: It was a single-centre trial with a small sample size.
Funding for this study: This research was supported by the National Natural Science Foundation of China under Grant 81271570
Ethics committee - additional information: The study was approved by the Institutional Review Board of Tongji Medical College,
Huazhong University of Science and Technology.
Relaxation-enhanced angiography without contrast and triggering (REACT) MRA with deep learning-constrained
compressed sensing in aorta aneurysm: a large field of view angiography (7 min)
Wen Zeng; Chengdu / China
946
Author Block: W. Zeng, C. Xia, Z. Li; Chengdu/CN
Purpose: In this study, free-breathing relaxation-enhanced angiography without contrast and triggering (REACT) MRA was applied to
aortic diseases to achieve large field imaging of the entire aorta. Deep learning-constrained compressed sensing was used to reduce
scanning time and improve image quality, and this study aimed to examine the feasibility of DLCS for shortening examination time
and improving image quality.
Methods or Background: Prospectively recruited patients with aortic disease, including aneurysms and dissections. All the patients
underwent 3.0T MR including REACT-CS6, REACT-CS9 and REACT-CS12 between June 2023 and September 2023. The images of
REACT-DLCS6, REACT-DLCS9 and REACT-DLCS12 were reconstructed. ROIs were placed at ascending aorta, arch, descending aorta,
abdominal aorta and liver. The signal-to-noise ratio of the aorta and the contrast-to-noise ratio between the aorta and the liver were
calculated. Quality grading was performed on a 5-point scale for the assessment of overall image quality, artifact removal and
background suppression.
Results or Findings: Twenty-three patients were included in this study. Acquisition times were CS6 (4 min 46 s), CS9 (3 min 14 s),
and CS12 (2 min 06 s). DLCS6 showed the highest SNR (p < 0.001) and CNR (p < 0.001). There was no evidence of differences among
DLCS6, DLCS9 and DLCS12 for CNR (DLCS6 vs. DLCS9, p = 1.00; DLCS6 vs. DLCS12, p = 1.00; DLCS9 vs. DLCS12, p = 1.00). DLCS6
obtained higher scores than other sequences (all p < 0.01) and the AICS9 followed.
Conclusion: In conclusion, REACT is suitable for large FOV angiography to aid in the diagnosis of patients with aortic disease without
the hassle of radiation and gadolinium contrast agents. DLCS9 reduces scanning time to 3 minutes while maintaining relatively good
image quality.
Limitations: The current work did not fully evaluate the diagnostic performances of these sequences.
Funding for this study: The study was funded by the National Key R&D Program of China (grant number: 2022YFC2009905).
Ethics committee - additional information: Ethics approval has been obtained from the Ethics Committee on Biomedical
Research, West China Hospital of Sichuan University (approval number: 2021-1171).
947
RPS 2210 - Bone density and spine imaging
Categories: EuroSafe Imaging/Radiation Protection, Imaging Methods, Musculoskeletal

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Üstün Aydingöz; Ankara / Turkey
Application of a novel ultra-high resolution photon counting detector CT in assessing thoracolumbar vertebral
anatomic structures (7 min)
Yufei Huang; Shanghai / China
Author Block: Y. Huang; Shanghai/CN

Purpose: The purpose of this study was to evaluate the feasibility of ultra-high resolution (UHR) scanning on photon counting
detector CT (PCD-CT) for assessing thoracolumbar vertebral anatomic structures.
Methods or Background: During September 2023, 20 patients underwent the UHR thoracolumbar scanning on PCD-CT utilising a
standard resolution (SR) reconstruction with a thickness of 1 mm and an increment of 0.7 mm, and a UHR reconstruction of 0.2 mm
for both thickness and increment. All images were reconstructed by the kernel of Br60, Br80, Br84 and Br89. The radiomic features
were extracted and the slope of grey values across the vertebral cortex were calculated. The image quality was subjectively analysed
by a Likert 5-point scale.
Results or Findings: The radiomic features associated with the demonstration of anatomical structures indicated that UHR images
outperformed SR images with statistical significance (p<0.001). The slope of grey values across the vertebral cortex of UHR images
was highest in UHR images with Br89 (1553.4 [1169.0], p<0.001) and Br84 (1643.5 [1402.5], p<0.001). Image noise significantly
increased in UHR images with a high level of kernel. The subjective evaluation revealed that UHR images with Br84 had the best
display of bone trabeculae and cortex (5 [0], p<0.001). Moreover, Br84 (5 [0]) and Br89 (5 [0]) had the highest demonstration of
spatial resolution (p<0.001).
Conclusion: It is feasible to perform thoracolumbar UHR scanning on PCD-CT, and Br84 is the best kernel to improve the spatial
resolution and visualisation of vertebral trabecular microstructure. Thoracolumbar vertebral imaging seems particularly likely to
benefit from the advantages of PCD-CT in the clinical scenario in the future.
Limitations: Due to ethical issues, it is not possible to compare the image quality of conventional scanning images on energy-
integrating detector CT and UHR scanning images on PCD-CT in the same patient.
Ethics committee - additional information: The study was approved by the ethics committee reference number: 186
Association between haemochromatosis genetic variants and vertebral fracture: a UK Biobank study (7 min)
Lucy Banfield; Exeter / United Kingdom
948
Author Block: L. Banfield, K. Knapp, D. Melzer, J. Atkins; Exeter/UK
Purpose: Haemochromatosis is caused by the p.C282Y homozygous variant resulting in iron overload. Previous studies demonstrate
increased risks of osteoporosis, back pain and femoral fracture in male homozygotes. The objective of this study was to review the
prevalence of vertebral fracture (identified on iDXA) by genotype, in a large community cohort.
Methods or Background: Lateral spine iDXA scans (GE-Lunar) acquired from UK Biobank were assessed. A subset of 292
participants; 146 pC282Y homozygotes and 146 age, BMI and sex matched controls were randomly selected. Images were reviewed
for radiological evidence of vertebral fracture using the Genant Semi-Quantitative Scale. Descriptive analysis and logistic regression
models assessed for associations between genotypes and incidence of vertebral fracture, adjusting for age and stratified by sex. Bone
mineral density (BMD) scores were also reviewed for participants with reported fractures.
Results or Findings: 37 vertebral fractures (12.9%) were identified with 20 (14.0%) in the p.C282Y homozygotes and 17 (11.8%) in
the wildtype. Vertebral fractures were twice as common in C282Y homozygous men (n=10, 18.2% vs n=5, 8.9%) but not higher in
women (n=10, 11.4% vs n=12, 13.6%) however, age-adjusted logistic regression suggests this is not statistically significant in males
(OR: 2.34 [0.73-7.42, p=0.15]) although numbers were small.
There was also no significant difference in BMD between the homozygotes and the wildtypes: Lumbar spine T-score –0.21 versus 0.34;
p=0.50, femoral neck T-score –1.03 versus -1.09 respectively; p=0.94.
Conclusion: There may be an increased prevalence of vertebral fracture in the male p.C282Y homozygotes in the absence of
diminished BMD when compared to those with no genetic mutations, but this was not statistically significant and further work is
needed in a larger cohort.
Limitations: The study used a volunteer population, so may be biased towards healthier individuals.
Funding for this study: Janice L. Atkins is supported by a National Institute for Health and Care Research (NIHR) Advanced
Fellowship (NIHR301844). David Melzer, Karen M. Knapp, and Lucy R. Banfield are supported by the University of Exeter.
Ethics committee - additional information: Ethical approval for UK Biobank was obtained from the North West Multi-Centre
Research Ethics Committee.
MRI or low-dose CT in suspected axial spondyloarthritis: preliminary results of a comparative study (7 min)
Dominik Deppe; Berlin / Germany
Author Block: D. Deppe, S. T. Ulas, M. Koka, F. N. Proft, M. Protopopov, L. Spiller, V. Rios Rodriguez, D. Poddubnyy, T. Diekhoff;
Berlin/DE
Purpose: This study aimed to evaluate and compare the diagnostic efficacy of magnetic resonance imaging (MRI) and low-dose
computed tomography (CT) in patients with suspected axial spondyloarthritis (axSpA). In a standard setting, patients undergo
radiography and subsequent MRI, when the radiograph is negative. CT is used in unclear cases, however, it’s gaining more attention
since the introduction of low-dose techniques.
Methods or Background: 86 patients with suspected axSpA were randomised into two arms: 1. MRI-first (followed by CT if MRI was
negative) and 2. CT-first (followed by MRI if CT was negative). Positive imaging was defined by the unambiguous identification of
active inflammatory (bone marrow oedema according to the updated ASAS definition suggestive of axSpA) or structural lesions (e.g.
extensive erosions or ankylosis suggestive of axSpA), rated in consensus reading by two radiologists specialised in musculoskeletal
imaging.
Results or Findings: In the MRI-first arm (48 patients), MRI was positive in 11/48 (23%) patients. Subsequent CT in 37 patients
showed no positive results (0/37; 0%). In the CT-first arm (38 patients), CT was positive in 12/38 (32%). Subsequent MRI (26 patients)
was positive in one (1/26; 4%).
Conclusion: MRI and low-dose CT demonstrated comparable diagnostic performance in patients with suspected axSpA.
Limitations: If initial imaging was positive, subsequent imaging was not performed which limits the analysis.
Ethics committee - additional information: The study was approved by the local ethics committee (EA1/145/22).
Straightening out scoliosis: how artificial intelligence assisted Cobb angle measurements on spine radiographs can
support the clinical user (7 min)
Willem Grootjans; Leiden / Netherlands
949
Author Block: W. Grootjans1, S. Challiui1, A. Bubberman1, C. Salzlechner2, D. Ehinger2, K. van Langevelde1; 1Leiden/NL, 2Vienna/AT
Purpose: Quantitative assessment of scoliosis using Cobb angles on spine radiographs is, albeit labour-intensive, commonly
performed in radiology practices. This study aimed to assess the added value and clinical impact of automating Cobb angle
measurements using artificial intelligence (AI).
Methods or Background: A total of 100 anteroposterior spine radiographs were retrospectively analysed with a fully automated
software solution based on deep learning (SQUIRREL, version 1.0, ImageBiopsy Lab, Vienna, Austria). Automatically determined Cobb
angles and range of the spinal curvature were compared to manual measurements. User interaction for AI (un)assisted Cobb angle
measurements was monitored using specially designed mouse-tracking software in 20 cases. Time per case, number of mouse clicks,
and mousing distance were recorded. Differences were reported as mean±SD (change %).
Results or Findings: The acceptance rate of automated Cobb angle measurements was 93%. Rejected cases were due to the
presence of metal implants or unrealistic measurements. For accepted cases, the mean difference in automatic and manually
determined Cobb angles was 3.7°±3.4, while the mean difference in upper and lower vertebral position was 1.2±1.5 and 0.8±1.1
respectively. Without AI, the mean number of left, right and middle mouse button (LMB, RMB, MMB) clicks was 32.2±7.0, 1.9±0.5, and
1.9±2.0 respectively. Mean mousing distance and time per case were 48620±22698 pixels and 66.9±18.0 seconds respectively.
Supporting the user with AI resulted in a mousing distance and time per case of 13868±2486 (-29%) and 29.2±4.7 (-44%) seconds
respectively. The mean number of LMB and RMB clicks were 11.3±2.2 (-35%) and 0.4±0.9 (-21%) respectively, while MMB was not
used.
Conclusion: Automated analysis of Cobb angles by AI-software showed excellent performance. Automation with AI modifies user
behaviour and reduces the physical effort required to report a case in clinical practice.
Limitations: Data on a single experienced user was reported. Further studies extend these experiments to different users.
Ethics committee - additional information: The study was approved by the institutional ethics commitee (reference number:
20231010001).
Assessment of tumour burden using quantitative lumbar magnetic resonance imaging in patients with multiple
myeloma (7 min)
Rui Xin Yan; Beijing / China
Author Block: R. X. Yan; Beijing/CN

Purpose: This study aimed to evaluate the potential usefulness of quantitative parameters derived from quantitative lumbar
magnetic resonance imaging (MRI) in quantifying tumour burden in multiple myeloma (MM) patients.
Methods or Background: 40 newly diagnosed MM patients were enrolled. The MM group was subsequently staged I-Ⅲ according to
the ISS. The healthy control group was comprised of 16 healthy subjects. All groups underwent MRI examinations, including plain MRI,
multi-echo chemical shift imaging (ME-Dixon) and diffusion Kurtosis imaging (DKI). Mean fat fraction (FF)/ mean diffusion coefficient
(MD), mean Kurtosis (MK), and apparent diffusion coefficient (ADC) were obtained by placing manual ROI on the lumbar vertebral
body three times. One-way ANOVA and the Kruskal-Wallis test were used to compare the differences in MR quantitative parameters
between different stages. The Pearson or Spearman rank correlation test was used to analyse the correlation between MR
quantitative parameters and prognostic biomarkers.
Results or Findings: Significant FF/MD/MK/ADC differences were shown between the MM group and the healthy control group
(P<0.05), and FF/MK/ADC showed significant differences between different ISS stages (P<0.05). FF was significantly different in
comparing ISS-I vs ISS-Ⅱand ISS-I vs ISS-Ⅲ, MK was only significantly different in comparing ISS-I vs ISS-Ⅲ. ADC only significantly
differed in comparing ISS-II vs ISS-Ⅲ. Besides, MK/ADC was negatively/positively correlated with β2-MG, Ca, and CCa (r=-0.431,
-0.354, -0.334/0.370, 0.338, 0.539, respectively).
Conclusion: Results suggest that quantitative parameters derived from lumbar MRI examination can provide potentially more
objective markers in assessing bone marrow lesions and tumour burden in patients with MM.
Limitations: The number of enrolled patients was relatively small, and the ISS stage cohort was distributed unbalanced. Secondly,
the ROI was selected and drawn manually. Therefore, we have drawn ROI three times and calculated the average as the final results
to minimise bias.
Ethics committee - additional information: This study was approved by the Affiliated Hospital of Qingdao University.
Assessment of the predictability of vertebral fractures in multiple myeloma in dual-layer CT (DLCT) with virtual non-
calcium (VNCa) CT images and calculation of virtual calcium-only (VCa) images (7 min)
Thuy Duong Do; Heidelberg / Germany
950
Author Block: S. Brandelik1, S. Rahn1, M. Merz2, W. Stiller1, S. Skornitzke1, C. Melzig1, H-U. Kauczor1, T. F. Weber1, T. D. Do1;
1
Heidelberg/DE, 2Leipzig/DE
Purpose: This study aimed to assess the predictability of vertebral fractures in multiple myeloma in dual-layer CT (DLCT) with virtual
non-calcium (VNCa) CT images and the calculation of virtual calcium-only (VCa) images.
Methods or Background: 81 patients with plasma cell dyscrasia and whole-body DLCT at the time of diagnosis and follow-up spine
imaging were included in the study. Conventional CT images (CI), VNCa images with calcium suppression (CaSupp) indices 25 and
100, and the novel method of calculated VCa images by subtraction of CaSupp100 - CaSupp25 were quantitatively analysed using
region-of-interests in the vertebral bodies L1-L5 and all vertebral bodies with fractures on baseline or follow-up imaging. Logistic
regression analyses were performed to assess the predictability of imminent spine fractures. For model comparisons, the Akaike
information criterion and R² were consulted.
Results or Findings: New fractures were seen in 24 patients' follow-up imaging. Predictability of new vertebral fractures was
significant for baseline assessment of CT numbers in CI, CaSupp 25 VNCa, and VCa (p=0.01, respectively) with a higher risk for new
fractures in case of lower CT numbers in CI and VCa (Odds ratio 0.982 [0.969;0.994], 0.987 [0.978;0.995]) and in case of higher CT
numbers in CaSupp 25 (Odds ratio 1.015 [1.006;1.026]). Direct model comparisons implied that CT numbers in CaSupp 25 and VCa
might show better fracture prediction than in CI (R2=0.18 both vs. 0.15; AICc=91.95, 91.79 vs. 93.62). Neither age, gender nor pre-
existing fractures improved the fracture predictability when included in the calculation.
Conclusion: VNCa and calculated VCa images in DLCT are feasible to predict imminent vertebral fracture risk in MM patients.
Limitations: This study was limited by the number of patients.
Ethics committee - additional information: This retrospective exploratory single-centre study was approved by the local review
board (application number: S-348/2019). The need for written informed consent was waived.
The role of dynamic magnetic resonance imaging in improving the diagnostic accuracy of degenerative diseases of the
lumbosacral spine (7 min)
Abdelfattah Saoud; Cairo / Egypt
Author Block: N. Ibrahim, A. Saoud, M. Yosry, A. Yassin; Cairo/EG

Purpose: This study aimed to evaluate the role of dynamic MRI in the assessment of lumbar spine degenerative diseases compared
to conventional MRI regarding the detection of spinal instability and evaluating the degenerative changes in flexion and extension in
comparison to a neutral position to determine the best clinical management and outcome for the patients.
Methods or Background: A prospective study conducted on symptomatic patients with degenerative spine referred to MRI. They
underwent imaging in neutral, flexion and extension positions. AP diameter of the spinal canal, foraminal height and area,
lumbosacral angle and lordotic angle were measured and used as a comparative index. Correlation with patients' symptomatology
and outcome were analysed.
Results or Findings: Dynamic MRI evoked radiological findings explaining the patients’ clinical picture in 32.4% of the patients. 25%
showed a narrowing of the neural foramina dimensions that could be elicited on dynamic images only. 25% of spinal canal stenosis
was induced by the positional change and 50% showed spinal instability either in flexion or extension position. 70.3% had
conservative management with pain improvement after 3 months of compliance. In 24.3% of the patients, the treatment plan was
changed from conservative to surgical intervention after performing dynamic MRI of which 75% showed improvement after surgery.
The evaluation of the patient’s pain level and outcome after management taken according to the dynamic MRI results showed
significant improvement (p-value 0.01).
Conclusion: Dynamic MRI of the lumbar spine allows better detection of hidden pathologies in conventional MRI in patients with
clinical/radiological mismatches. It gives the surgeon a tool to understand patients’ symptomatology. It also provides evidence of
neural compression and facilitates accurate preoperative planning regards the procedure and the target level, and accordingly better
postoperative outcome.
Limitations: This study had a small patient population and there was a lack of long-term follow-up of some of the patients.
Ethics committee - additional information: This study has the approval number MS 85/2023.
Opportunistic screening of osteoporosis using routine lumbar spine MRI: a retrospective comparative study (7 min)
Naser Mohammad Issa Obeidat; Irbid / Jordan
951
Author Block: L. Rousan, N. M. I. Obeidat, N. Abdo, L. Alshraa, T. Ajam, H. Al-Zoubi, H. Saif, A. Albaqshi; Irbid/JO
Purpose: Multiple studies compared T-scores obtained from BMD with different MRI protocols depending on the SNR ratio of the
lumbar vertebral bodies. This study aimed to assess whether routine sequences of the lumbar spine MRI can predict the presence of
osteoporosis by visually assessing the alteration of the signal intensity alone.
Methods or Background: A 322 cohort of patients who underwent both lumbar spine MRI and DEXA scan within a 6-month interval
between January 2015 and December 2022 were selected. Three radiologists graded L4 (or L3) vertebral body fat content into ≤50%
or >50% based on visual estimation of its bone marrow signal intensity on sagittal T1 weighted sequences. Average vertebral body
bone marrow signal intensity was also obtained. MRI interpretation results were compared with DEXA scan reports after grouping
patients into three categories (normal, osteopenia, and osteoporosis). Statistical analysis was performed and a p-value of <0.05 was
considered significant.
Results or Findings: 322 patients were enrolled in the study (80.4% females, average age 57 years). 10.2% of which had
osteoporosis, but the majority had a normal T-score (54.7%). MRI interpretation yielded 167 and 155 patients with a score of ≤50%
and >50%, respectively, with an average signal intensity of 727. There was a significant statistical correlation between each of the
MRI readings (estimated fat percentage and bone marrow signal intensity) with the diagnosis of reduced bone density on the DEXA
scan (p-value of 0.0008 and 0.028, respectively).
Conclusion: Our results showed that routine lumbar spine MRI studies can predict reduced bone density and, hence proved to be
beneficial in opportunistic screening for osteoporosis. These results are of considerable interest in suggesting the silent disease of
osteoporosis in the most requested musculoskeletal MRI scan.
Ethics committee - additional information: The institutional review board approved this study and waived the need for written
consent (Approval No. 38/123/2019).
952
RPS 2207 - Non-oncological genitourinary diseases
Categories: Abdominal Viscera, Genitourinary, Imaging Methods

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Revolutionising the diagnosis of congenital cervical malformations: a novel magnetic resonance imaging-based
classification for streamlined standardised assessment (7 min)
Zhilin Yuan; Beijing / China
Author Block: Z. Yuan, Y-L. He, Y. Li, J. Ren, M. Huang, X. Liu, C. Wang, Z. Y. Jin, H-D. Xue; Beijing/CN
Purpose: The aim of this study was to develop an innovative MRI-Based classification method to optimise standardised diagnosis of
congenital cervical malformations (CCMs) and summarise their unique MRI Features.
Methods or Background: A total of 79 consecutive patients with CCMs, pre-treatment pelvic MRI scans were conducted and
subsequently reviewed by three experienced gynaecological radiologists. Various characteristics, such as signal patterns,
morphological classifications, and other relevant factors, were meticulously documented. Furthermore, any accompanying
abnormalities were noted for subsequent analysis.
Results or Findings: Morphologically, CCMs can be categorised into three types as follows: Type I (42,53%), characterised by the
presence of a cervix with visible cervical canals; Type II (18,23%), featuring an existing cervix with concealed cervical canals; and
Type III (19,24%), indicating cervical aplasia, which involves a blind end in the lower part of the uterine corpus. Haematometra was
significantly more prevalent in Type I compared to Type II patients (p<0.001), and Type I patients exhibited a significantly longer
mean sagittal length than Type II patients (p<0.001). Cervical detection had three signal patterns: no signal (27%), no-evident layer
differentiation (21%), and multiple-evident layer differentiation with haematocele (52%). Most patients (94%) had complete vaginal
atresia.
Conclusion: The MRI-based classification system proposed for CCMs, along with the highlighted key MRI features, demonstrates
substantial potential to improve the effectiveness of clinical diagnoses for individuals with CCMs.
Limitations: The current study has some inherent limitations. Firstly, it was a retrospective observational study conducted at a single
centre, which may have introduced certain sampling biases. Second, the number of enrolled patients was relatively small. Given the
rarity of CCMs, collecting larger datasets from multiple centres would be advantageous for the continued improvement of this
classification system.
Funding for this study: This work was supported by grants from National High Level Hospital Clinical Research Funding (grant No.
2022-PUMCH-A-004) and Natural Science Foundation of China (grant No. 82271886)
Ethics committee - additional information: Ethical approval for this study was obtained from the Institutional Review Board.
Correlation between total fibrotic kidney volume and parasympathetic activity in autosomal‐dominant polycystic
kidney disease patients: a pilot study (7 min)
Antonella Borrelli; Rome / Italy
953
Author Block: A. Borrelli, L. Laschena, S. Lucciola, A. Dehghanpour, C. Catalano, V. Panebianco; Rome/IT
Purpose: In patients with autosomal-dominant polycystic kidney disease (ADPKD) cystic enlargement may cause parenchymal
hypoxia, renin secretion, and endothelial dysfunction, causing hyperactivation of renin–angiotensin system and early hypertension,
affecting blood pressure circadian rhythm. Aim of this study is to evaluate renal damage progression, indicated by MRI-based
parameter, in ADPKD patients, correlating with sympathetic/parasympathetic balance using heart rate variability (HRV) parameters.
Methods or Background: Sixteen adult ADPKD patients were enrolled, undergoing MRI with three Tesla scanner, to evaluate
especially total kidney volume (TKV) and total fibrotic volume (TFV). A post-processed slice-by-slice renal segmentation, with
colorimetric maps was obtained. A software was used for three-dimensional volume rendering reconstruction, which resulted in
semiquantitative estimation of parenchymal perfused tissue and fibrotic areas, giving indication of functional parenchymal areas.
These data were correlated with HRV parameters, calculated using 24 h-ECG Holter (low frequency, LF, and high frequency, HF).
Results or Findings: A statistically significant positive linear correlation was observed between length of kidneys and LF (r = 0.595,
p < 0.05), and LF-day (r = 0.587, p < 0.05). Moreover, a statistically significant positive linear correlation exists between HF and TFV
(r = 0.804, p < 0.01) and height- adjusted (ha) TFV (r = 0.801, p < 0.01). Finally, we found a statistically significant positive linear
correlation between HFnight and TKV (r = 0.608, p < 0.05), ha-TKV (r = 0.685, p < 0.01), TFV (r = 0.594, p < 0.05) and ha-TFV (r =
0.615, p < 0.05).
Conclusion: The increase in TKV and TFV could lead to parasympathetic tone hyperactivation, probably in response to hypoxic stress
and vasoconstriction due to cystic enlargement.
Limitations: Limited sample size: however, ADPKD represents a rare disease, therefore the sample may be considered
representative
Funding for this study: No funding was available for this study.
Ethics committee - additional information: Ethical approval for this study was obtained from the Institutional Ehtics committee.
MRI of pelvic endometriosis: evaluation of the mr#Enzian classification and the importance of adenomyosis subtypes (7
min)
Antonia-Maria Pausch; Zurich / Switzerland
Author Block: A-M. Pausch, V. Filleböck, M. Benli, I. Witzel, A. M. Hötker; Zurich/CH

Purpose: This study aimed to investigate the utility of the #Enzian classification in magnetic resonance imaging (MRI) for
endometriosis assessment, focusing on inter-reader agreement, diagnostic accuracy, and the correlation of adenomyosis with deep
infiltrating endometriosis (DIE).
Methods or Background: This IRB-approved retrospective single-centre study included 412 women who underwent MRI evaluation
for endometriosis between February 2017 and June 2022. Two experienced radiologists independently analysed MRI images using the
#Enzian classification and assessed the type of adenomyosis, if any. The surgical #Enzian classification served as the gold standard
for evaluating preoperative MRI results. Statistical analysis was performed to assess inter-reader agreement and diagnostic accuracy.
Results or Findings: Inter-reader agreement was substantial to excellent (Cohen’s kappa 0.75 – 0.96) for most compartments
except peritoneal involvement (0.39). The preoperative MRI showed mostly substantial to excellent accuracy (0.84 – 0.98), sensitivity
(0.62 – 1.00), specificity (0.87 – 1.00), positive (0.58 – 1.00) and negative predictive values (0.86 – 1.00) for most compartments,
except for peritoneal lesions (0.36, 0.17, 1.00, 1.00, 0.26 respectively). A trend with a higher prevalence of concordant DIE in women
with external compared to those with internal adenomyosis was visible (p = 0.067).
Conclusion: The mr#Enzian showed mostly high inter-reader agreement and diagnostic accuracy for various endometriosis
compartments. MRI's role is particularly significant in the context of the current paradigm shift towards medical endometriosis
treatment. The inclusion of information about the type of adenomyosis in the mr#Enzian classification could enhance diagnostic
accuracy and inform treatment planning.
Limitations: The study may have selection bias due to the highly subspecialised gynaecology department as referral source,
moreover, the use of laparoscopy with histological confirmation as the gold standard, despite its known limitations, could introduce
further bias.
Ethics committee - additional information: Cantonal Ethics Committee Zurich gave the ethical approval for this study.
The hidden connections: the relationship between endometrioma size, adenomyosis and deep infiltrating
endometriosis (7 min)
Hande Özen Atalay; Istanbul / Turkey
954
Author Block: H. Özen Atalay, M. S. Sogut, U. Kalkan, A. Durur Karakaya; Istanbul/TR
Purpose: The purpose of this study was to evaluate the association between endometrioma size, the presence of adenomyosis, and
deep-infiltrating endometriosis.
Methods or Background: The present retrospective study was performed at a single center, involving a total of 118 patients who
were diagnosed with endometrioma. The longest diameter of the biggest endometrioma, the presence of adenomyosis, and the
presence of deep-infiltrating endometriosis were evaluated in the pelvic MRI by two radiologists with consensus. The Kruskal-Wallis
test, Pearson correlation coefficient and Fisher's exact test were used to calculate the correlation coefficient, and to detect possible
relationships.
Results or Findings: Seventy-one patients have deep-infiltrating endometriosis, and thirty-one patients have adenomyosis, in total
of 118 patients with endometrioma. The mean diameter of the endometriomas was 40.79±18.9 mm. The mean diameter of the
endometrioma did not have a statistically significant effect on the presence of adenomyosis or deep-infiltrating endometriosis
(p=0.812). There were weak correlations between the mean diameter of the endometrioma and the presence of adenomyosis and DIE
(r=0.028 and r=0.060, p>0.05 respectively). However, there was a significant association between the presence of adenomyosis and
deep-infiltrating endometriosis (p=0.001). The Pearson’s r correlation test yielded a moderately positive correlation (r=0.301,
p=0.001).
Conclusion: This study indicates that endometrioma size is not significantly correlated with the presence of adenomyosis or deep-
infiltrating endometriosis. However, a significant association exists between the presence of adenomyosis and deep-infiltrating
endometriosis. These findings suggest that while endometrioma size may not be a predictive factor, the co-occurrence of
adenomyosis and deep-infiltrating endometriosis should be considered in clinical evaluations.
Limitations: This retrospective study included a limited number of patients from a single institution.
Ethics committee - additional information: Koc University Biomedical Research Ethics Committee, Istanbul/Turkey ethically
approved this study.
Photon-counting detector CT for kidney stone detection in contrast-enhanced urography: a comparison between
virtual non-contrast and virtual non-iodine reconstructions in a 3D printed kidney phantom (7 min)
Andre Euler; Baden / Switzerland
Author Block: P. S. Breiding1, M. Turrion2, K. Martini2, D. N. Nakhostin2, H. Alkadhi2, A. Euler3; 1Bern/CH, 2Zurich/CH, 3Baden/CH
Purpose: The aim of this study was to compare contrast media subtraction, kidney stone visibility and subjective image quality
between a novel virtual non-iodine reconstruction algorithm (VNI; PureCalcium) and a virtual non-contrast (VNC) algorithm on photon-
counting detector CT (PCD-CT).
Methods or Background: Calcium oxalate and uric acid stones were placed into contrast-filled calyces of a 3D printed kidney
phantom and imaged on PCD-CT. Two blinded readers assessed quality of PC and VNC images using a five-point visual scale. The
readers were asked to rate contrast media subtraction and erroneous kidney stone subtraction in each cylinder using a three-point
visual scale.
Results or Findings: Overall image quality was rated higher for VNI images compared to VNC images by one reader (4.9 vs. 4.0; p <
0.05). Inter-reader agreement for how well contrast was subtracted was substantial for both VNC and PC images (Krippendorff's alpha
= 0.628 and 0.731). Incomplete contrast subtraction was observed more frequently for VNI images compared to VNC images by one
reader (29% vs. 15%; p < 0.05). Inter-reader agreement for kidney stone subtraction was substantial for both VNC and PC images
(Krippendorff's alpha = 0.748 and 0.668). Kidney stones were subtracted more frequently with VNI compared to VNC (Reader 1: 22%
vs. 16%; Reader 2: 25% vs. 10%; p < 0.05). Smaller stones were subtracted more frequently than larger stones irrespective of
reconstruction method for both readers (p < 0.05).
Conclusion: Overall subjective image quality was higher for VNI images compared to VNC images. Incomplete contrast subtraction
and subtraction of kidney stones was more frequent for VNI images.
Limitations: The phantom model may be prone to artifacts and does not reflect true tissue heterogenicitiy or dynamic tissue
properties. We only used two keV levels to reconstruct the monoenergetic PC images.
Ethics committee - additional information: This study was ethically approved the Ethics committee - University of Zürich.
Quantitative measurement of renal artery spin labelling imaging: a non-invasive index to evaluate perfusion in
patients with renal artery stenosis (7 min)
Xiaoxiao Zhang; Beijing / China
955
Author Block: X. Zhang, G. Zhang, L. Xu, X. Bai, J. Zhang, L. Chen, H. Sun, Z. Jin; Beijing/CN
Purpose: The objective of this study was to evaluate the effect of interventional therapy on renal blood flow (RBF) in renal artery
stenosis (RAS) patients and to explore the correlation between RBF and kidney glomerular filtration rate (GFR).
Methods or Background: Eleven RAS patients underwent preoperative arterial spin labeling mapping (ASL) exam and interventional
therapy were enrolled, and seven patients underwent postoperative ASL examination. We analysed 22 renal arteries and kidneys as
independent subjects, dividing them into no renal artery stenosis, mild renal artery stenosis (≤75%), and severe renal artery stenosis
(＞75%) groups based on the degree of luminal stenosis. We evaluated the correlation between preoperative cortical RBF and
preoperative single kidney GFR, and differences between preoperative and postoperative RBF were compared.
Results or Findings: All RAS patients experienced a decrease in systolic and diastolic blood pressure post-interventional therapy.
The RBF without renal artery stenosis was higher than that in the renal artery stenosis group (273.68 ±62.77 vs 172.05±62.44), and
the RBF in the mild renal artery stenosis group was higher than that in the severe renal artery stenosis group (187.33±49.66 vs
156.77±75.21). A positive correlation was found between preoperative cortical RBF and preoperative single kidney GFR, with no
significant correlation between preoperative cortical RBF and preoperative estimated GFR. The renal cortical RBF was higher than
preoperative RBF (203.19±51.42 vs.164.00±68.10). The postoperative RBF in the region of abnormal perfusion was also higher than
the preoperative RBF (159.92±46.10 vs. 108.55±39.57).
Conclusion: Renal RBF obtained by ASL images was significantly correlated with single renal function and could be used to evaluate
the perfusion improvement in RAS patients after interventional therapy.
Limitations: The main limitations of this study are that the sample size is too small and that some patients did not undergo
postoperative review.
Ethics committee - additional information: Ethical approval was sought from the Institutional Review Board of Peking Union
Medical College Hospital.
Follow-up in acute pyelonephritis (PNA) using DWI sequences in MRI: our experience (7 min)
Giuseppe Di Guardia; Verduno / Italy
Author Block: G. Di Guardia, P. Lasciarrea, F. Testa, V. Verna, S. Rubiolo, X. Kaci, F. Lucio, A. Cappelletti; Verduno/IT
Purpose: Several studies have highlighted the role of MRI and diffusion-weighted sequences (MRI-DW) in the diagnosis of PNA and its
impact regarding the planning of an adequate therapeutic strategy by nephrologists.
The aim of our study is to evaluate the changes in signal alteration in diffusion-weighted sequences (DWI) and ADC values in patients
(pts) with PNA, by comparison with the mean ADC value (ADCm) of both the pathological kidney and the contralateral, in acute and
during follow-up.
Methods or Background: From June 2016 to February 2020, 101 patients were studied (10 males, 91 females, average age 40
years, minimum age 18 years, maximum age 68 years) with clinical-laboratory diagnosis of PNA. 26/101 patients were evaluated
retrospectively by evaluating the DWI sequences, demarcating circular regions of interest (ROI) on the areas of impaired signal and
comparing them both with the ADCm value of the ipsilateral healthy renal parenchyma, and with the contralateral, obtaining an ADC
ratio ( ADCr).
Results or Findings: In the acute phase (t0), 94 outbreaks of PNA were identified, with ADCm 1.26 ± 0.17 x 10-3 mm2 / s, ADCr0 =
0.64 ± 0.08. At follow-up after therapy (mean follow-up time 36.4 ± 11.7 days) we observed clinical-laboratory resolution in all
patients, but only 39/94 foci demonstrated complete resolution on MRI (41%) in agreement with the data in the literature.
Conclusion: RM-DW allows a valid monitoring of PNA outbreaks, also identifying the different degrees of evolution. For the
calculation of the ADCr we did not detect statistically significant changes using the free parenchyma of the pathological kidney
compared to the contralateral kidney.
Limitations: No limitations have been identified for this study.
Funding for this study: No funds were sought for this study.
Ethics committee - additional information: No ethics committee approval was required.
Reliable volume measurements in polycystic kidneys: a comparison between 3D volumetry and Mayo clinic’s kidney
volume calculator based on the ellipsoid equation (7 min)
Claudia Gerlotti Slusnys; Madrid / Spain
956
Author Block: C. Gerlotti Slusnys1, M. Paniagua González1, D. DÉlia Torrence2, P. López Gómez1, M. Camarena Gea1, M. Poida1, A.
Garcia Perez1, C. Calles Blanco1, J. Amorim Sortino1; 1Madrid/ES, 2Logroño/ES
Purpose: This study focuses on the precise determination of total kidney volume (TKV) in individuals with polycystic kidney disease
(PKD). Accurate TKV measurement is important for predicting renal functional decline and assessing Tolvaptan treatment candidates.
The widely used Mayo clinic ellipsoid equation is simpler but tends to underestimate TKV. We aim to assess its accuracy compared to
the more precise 3D volumetry method.
Methods or Background: We retrospectively studied 15 PKD patients (30 kidneys) who had CT scans from 2019 to 2023. TKV was
calculated using the Mayo Clinic method (MCM) and 3D volumetry method (3DM) with manual segmentation. In this study, a
predetermined threshold was set, deeming a difference of less than 10 percent between the two volumes as clinically not relevant
according to expert nephrologists. The paired Student’s T-test, dispersion, and Pearson’s correlation analysis were employed for
hypothesis testing.
Results or Findings: The paired Student’s T-test revealed a statistically significant difference between the means of both methods,
with an 8.55% variation. Dispersion analysis showed a range from -140 cc to +286 cc. Pearson’s correlation analysis demonstrated a
strong linear correlation, enabling the creation of a formula to predict 3DM volumetric values based on MCM calculations.
Conclusion: Our study, consistent with existing literature, found that the statistical difference between methods is below 10%.
Although 3DM is more precise, MCM remains a simpler and practical TKV calculation option. The linear correlation allows us to predict
3DM values based on MCM results. In conclusion, both tests are functional, and we introduce a tool that could potentially reduce the
need for 3D reconstruction.
957
E³ 22E - Traumatic lesions in the hand
Categories: Emergency Imaging, General Radiology, Musculoskeletal

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Finger injuries in sports (25 min)

1. To learn about pulley and tendon injury assessment.

2. To learn about the evaluation of collateral ligament injuries.
3. To recognise patterns regarding emergency treatment and report the essential information to the clinician.
Traumatic TFCC and ECU injuries (25 min)

Filippo Del Grande; Lugano / Switzerland
1. To learn about the Palmer classification.

2. To become familiar with associated injuries.
3. To detect signs for ECU pathology and instability.
958
RPS 2211 - Brain tumour imaging: from radiology to pathology
Categories: Artificial Intelligence & Machine Learning, Molecular Imaging, Neuro, Oncologic Imaging, Research
Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Jerzy Walecki; Warsaw / Poland
Default mode network functional connectivity is sensitive to glioma WHO-grade (7 min)

Ahmed M. Radwan; Leuven / Belgium
Author Block: A. M. Radwan, H. Vandermeulen, F. Samardzic, J. M. Sousa, S. Sunaert; Leuven/BE

Purpose: This study examined the impact of glioma characteristics (WHO grade and volume), patient age, and gender on the
functional connectivity (FC) of seven canonical resting-state networks (RSNs). Specifically, FC difference in the default mode (DMN),
dorsal attention (DAN), fronto-parietal (FP), language (Lang), salience (SAL), sensory-motor (SMN), and visual (Vis) networks.
Methods or Background: BOLD resting-state functional magnetic resonance imaging (rs-fMRI) data were collected at 3 Tesla from
40 neurosurgery-naı̈ve patients, newly diagnosed with cerebral glioma. Data were preprocessed using fmriprep, and denoising and
seed-to-voxel functional connectivity mapping (SBA-FC) were performed with CONN. Hypothesis testing was conducted using FSL’s
Randomise for nonparametric two-group t-tests. This excluded subject-specific lesioned voxels and controlled for lesion volume,
patient age, and gender. SBA-FC beta maps were compared between high-grade gliomas (HGG, WHO grades III-IV, N=21) and low-
grade gliomas (LGG, WHO grades I-II, N=19). Separate statistical contrasts also evaluated the effects of lesion volume, patient age,
and gender.
Results or Findings: Global DMN connectivity was significantly reduced in HGG patients compared to LGG (579 voxels, t-value
mean=3, PFDR<.05), involving the posterior cingulate and precuneus gyri. A small focus (18 voxels) of significantly increasing FC with
age was identified in the right supplementary motor area (SMA) (t-value mean=4.45, PFDR<0.05). Tumour volume and patient gender
did not significantly influence FC in any of the seven RSNs.
Conclusion: DMN functional connectivity is notably sensitive to glioma WHO grade. FC in the right SMA appeared to increase with
patient age, while lesion volume and patient gender did not impact FC in the other RSNs.
Limitations: This study had a limited sample size and a lack of neuropsychological data.
Ethics committee - additional information: This study was approved by the UZ/KU Leuven ethics committee study number is
S61759.
IDH status prediction in gliomas using machine-learning analysis of multiparametric MRI (7 min)
Vojtěch Sedlák; Humpolec / Czechia
959
Author Block: V. Sedlák, T. Belsan, D. Netuka, A. Kavková; Prague/CZ
Purpose: This study aimed to explore the efficacy of machine-learning algorithms in accurately predicting Isocitrate Dehydrogenase
(IDH) mutation status in adult-type diffuse brain gliomas, utilising quantitative data extracted from multiparametric MRI, to enhance
diagnostic precision and potentially guide personalized treatment strategies.
Methods or Background: A cohort of 100 patients underwent comprehensive multimodal MRI, encompassing ASL perfusion, DSC
perfusion, advanced diffusion imaging (including DKI, SMT and other models) and standard morphological imaging (i.e. T2, FLAIR, SWI,
pre and postcontrast T1). Quantitative features were then extracted from these scans and fed into machine-learning algorithms, with
the objective of developing a predictive model for IDH status in gliomas. Investigated algorithms included random forest, XGBoost,
AdaBoost, logistic regression and support vector machine models.
Results or Findings: Various performance metrics were assessed for each model with emphasis on accuracy and AUC. The
investigated machine-learning models achieved high diagnostic accuracies in determining the IDH mutation status, with areas-under-
the-curve ranging from 89% for Random Forrest to 97% in the case of the Logistic Regression model.
Conclusion: The integration of machine-learning algorithms with multiparametric MRI data demonstrates a promising avenue for the
accurate prediction of IDH status in glioma patients. This approach not only substantiates the pivotal role of advanced imaging
techniques in diagnostic neuro-oncology but also underscores the transformative impact of machine-learning in medical diagnostics
and patient stratification.
Limitations: The main limitation of the study is the still relatively modest sample size in combination with the inherent heterogeneity
of glioma characteristics, which in combination might introduce potential bias in algorithm training. Further studies with larger cohorts
and external validation are imperative to ascertain the generalisability of these models.
Funding for this study: This research was financially supported by the Charles University Grant Agency (project no. 222623)
entitled “Advanced Diffusion MR Imaging in Diagnosis of Brain Tumors”, implemented at the Second Faculty of Medicine of Charles
University.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Military University Hospital in
Prague.
DSC-PWI presurgical differentiation of Grade 4 astrocytoma and glioblastoma in young adults: unsupervised percentilic
rCBV analysis across enhancing and non-enhancing regions (7 min)
Ady Mildred Viveros; Barcelona / Spain
Author Block: A. M. Viveros, P. Naval-Baudin, S. Flores Casaperalta, F. A. Garay Buitron, S. Septién Rivera, M. Cos Domingo, C. Majós,
A. Pons Escoda; Barcelona/ES
Purpose: This research aimed to evaluate the differentiation ability of relative-cerebral-blood-volume (rCBV) percentile values for the
enhancing and non-enhancing tumour regions compared to the more commonly used mean or maximum preselected rCBV values.
The presurgical differentiation between IDH-mutant astrocytoma-grade-4 and IDH-wildtype-glioblastoma is relevant for patient
management, especially in young adults. It provides prognostic information and aids in guiding the molecular diagnostic work-up or in
identifying patients for trials on IDH-directed treatments. While DSC-PWI has demonstrated potential for this task, its full capabilities
may not yet have been realised.
Methods or Background: Patients with grade 4 astrocytic tumours, known IDH-mutation status, available presurgical MR with DSC-
PWI, and under 55 years old (threshold below which IDH-mutations are evenly balanced) were retrospectively retrieved from 2016-
2023. Both enhancing and non-enhancing regions were 3D-segmented. Voxel-level rCBV was calculated to derive mean, maximum,
and percentile values. Statistical comparisons were performed using the Mann-Whitney U test and AUC-ROC.
Results or Findings: The study comprised 59 patients: 11 astrocytoma-4, and 48 glioblastoma. The enhancing regions of
glioblastoma displayed a higher rCBV, though the differences were not statistically significant. The non-enhancing components of
astrocytoma-4 exhibited significantly higher rCBV, more pronounced when assessing lower percentiles. The 30th rCBV percentile
values for the non-enhancing region were 0.705 in astrocytoma-4 and 0.458 in glioblastoma, with a p-value of 0.001 and AUC-ROC of
0.811. This outperformed the results derived from the commonly used mean and maximum.
Conclusion: An unsupervised percentile-based approach to select rCBV values enhances the differentiation outcomes over the
traditional mean and maximum. The non-enhancing region offers more valuable insights than the enhancing region. Elevated rCBV
values in the lower percentiles of the non-enhancing component in astrocytoma-4 are the most distinguishable characteristic and may
represent very-low vascularized infiltrated tissue, versus pure oedema in glioblastoma.
Limitations: This was a single-site and retrospective investigation.
Ethics committee - additional information: This study was approved by the Ethics Committee of Hospital Universitari de
Bellvitge.
Tumour-ipsilateral hemisphere T2 relaxometry predicts progression-free survival in patients with primary glioblastoma
(7 min)
Josef Vymazal; Prague / Czechia
960
Author Block: J. Vymazal, A. Rulseh; Prague/CZ
Purpose: This study aimed to assess whether repeated MR T2 relaxometry of the tumour-ipsilateral hemisphere can predict the
progression-free survival (PFS) of patients with primary glioblastoma (GBM).
Methods or Background: 299 MR examinations in 32 GBM patients were included in this study. T2 in the tumour ipsilateral
hemisphere was calculated and plotted against PFS. The evaluation period ranged between 4 months and 19 years. Patients with no
disease progression were included only if they exceeded 2.5 years of follow-up; only data from the first 2.5 years were used if PFS
was longer. 1/T2 was plotted against PFS with linear regression and the slope was calculated. No post-progression data were used.
Results or Findings: Seven patients with PFS longer than 6 years (6-19 years, three of them still with no progression) had an
average slope of 0.0018 (-0.1294 to 0.4256). Seven patients with PFS less than 1 year (0.329 to 0.944) had an average slope of
-1.2659 (-3.7109 to -0.1964), p=0.025. Eight patients with PFS between 1-2 years had an average slope of -0.619 (-0.1412 to -1,3223)
p=0. Ten patients with PFS between 2-5 years had an average slope of -0.4032 (-0.6595 to -0.1331), compared with PFS longer than 6
years p=0.0015.
Conclusion: T2 relaxometry from the tumour-ipsilateral hemisphere reliably predicted PFS longer than 6 years, based on data from
the first 2.5 years. The average linear regression slope was dependent on PFS: No patients with PFS longer than 6 years had a slope
lower than -0.12, and no patients with a slope less than -0.8 had PFS longer than 1.67 years. This methodology may select patients
with a high risk of early recurrence and also those where long-term PFS can be expected.
Limitations: This was a partially retrospective study.
Funding for this study: This study was supported by MH CZ – DRO (Na Homolce Hospital – NNH, 00023884) IG204301, IG204302.
Ethics committee - additional information: This study was approved by the Ethics Committee of the Na Homolce Hospital,
Prague, The Czech Republic.
Quantitative histopathological analysis of the periphery of glioblastoma based on ADC and rCBV (7 min)
Juan Romero Coronado; Madrid / Spain
Author Block: J. Romero Coronado, A. Ramos Gonzalez, A. Hilario Barrio, E. Salvador Alvarez, C. Lechuga Vázquez, Z. H. Chen Zhou,
A. C. Martinez De Aragón Calvo, A. Cardenas; Madrid/ES
Purpose: In glioblastoma, non-enhancing areas with FLAIR hyperintensity represent both vasogenic oedema and tumour infiltration.
The purpose of this study is to correlate the degree of tumour infiltration in histological samples obtained from the FLAIR hyperintense
area, targeted by ADC and rCBV in a pre-surgical analysis.
Methods or Background: A total of 33 biopsies performed on 11 patients diagnosed with glioblastoma were analysed. Samples
were obtained in their first surgical procedure in patients without any prior treatment.
MRI with DWI and DSC sequences were performed and analyzed with Olea Sphere software. We obtained different quantitative
parameters using ROIs in the periphery of the non-enhancing tumour. During surgery, before the enhancing tumour was removed,
biopsies were obtained from the periphery, previously targeted based on ADC and rCBV. To assess the proliferative potential of our
samples, we performed immunohistochemical staining of p53 and MIB1. All slides underwent digital scanning using a NanoZoomer-SQ
scanner, and positive cells were quantified using QuPath-0.4.2 software.
Results or Findings: In this study, we have obtained a valuable correlation between ADC and rCBV data, and the presence of
tumour infiltration in the non-enhancing peripheral tumour with high signal in T2 and FLAIR sequences. The rCBV does not correlate
when the value is above 1, in which case we always found tumour infiltration. The ADC correlates with infiltration when the values are
low, while we have unexpectedly found areas with high ADC with dense tumour infiltration presumably due to the presence of large
vasogenic oedema.
Conclusion: Our findings provide valuable insights into the nature of the peripheral zone of glioblastomas. Future studies will help us
to correlate radiomic parameters with the degree of tumour infiltration, considering pathological, biological and immune biomarkers.
Funding for this study: This study was funded by the Fondo de Investigación en Salud (FIS).
Differentiating brain metastases, glioblastoma and primary central nervous system lymphoma non-invasively using
artificial intelligence-based multiparametric MRI (7 min)
Junjie Li; Beijing / China
961
Author Block: J. Li, L. Chai, Z. Zhuo, Y. Duan, Y. Liu; Beijing/CN
Purpose: This study aimed to differentiate brain metastases, glioblastoma and primary central nervous system lymphoma non-
invasively using artificial intelligence-based multiparametric MRI. Accurate differentiation of brain metastases (BM), glioblastoma
(GBM), and primary central nervous system lymphoma (PCNSL) is crucial in clinical practice. However, most studies on artificial
intelligence (AI) only focus on the differentiation of two types of tumours, lacking research on AI methods for the simultaneous
differentiation of all three tumours.
Methods or Background: This study included preoperative multiparametric MRI images of BM (n=375), GBM (n=391), and PCNSL
(n=361). The MRI sequences consisted of T1w, T2w, FLAIR, ADC, and contrast-enhanced T1 images. The data were randomly divided
into a training set (n=788) and a test set (n=339) in a 7:3 ratio. A fully automated differentiation model was developed based on the
multiparametric MRI for tumour differentiation. The results of the model were compared with those of junior and senior neuro-
radiologists.
Results or Findings: The accuracy of the model in differentiating BM, GBM, and PCNSL was 0.73, with corresponding AUC values of
0.82, 0.88, and 0.88. The results were similar to those of senior neuro-radiologists (accuracy: 0.74; AUC: 0.87, 0.90, 0.90) and higher
than those of junior neuro-radiologists (accuracy: 0.60; AUC: 0.60, 0.73, 0.71). The accuracy improved for the junior neuro-radiologists
after re-evaluating the cases using the model's results (accuracy: 0.70; AUC: 0.71, 0.81, 0.77).
Conclusion: The model utilises multiparametric MRI for non-invasive differentiation of BM, GBM, and PCNSL. The results of the model
are similar to those of senior neuro-radiologists and superior to those of junior neuro-radiologists. Thus, the diagnostic proficiency of
junior neuro-radiologists was improved.
Limitations: Firstly, all data were obtained from a single centre. Additionally, The number of PCNSLs was relatively lower in all data
sets, and this class imbalance may impart statistical bias in model performance. Lastly, the model was trained only with axial MRI
slices.
Funding for this study: This work was supported by the Beijing Municipal Natural Science Foundation for Distinguished Young
Scholars (No. JQ20035), Capital Health Development Research Project (NO. 2022-1-2042); Radiographic Standard Database
Construction Project (NO. YXFSC2022JJSJ004).
Ethics committee - additional information: This study was approved by the Institutional Review Board of our hosipital, and
written informed consent was obtained from all patients or their legal guardians.
The clinical application of MATRIX technology in the detection of brain metastases (7 min)
Xuejun Chen; Zhengzhou / China
Author Block: X. Chen, Y. Junhui; Zhengzhou/CN

Purpose: This study aimed to explore the ability to detect brain metastases of the three-dimensional fast spin echo sequence
modulated flip angle technique in refocused imaging with an extended echo train (MATRIX).
Methods or Background: Data from 59 patients suspected of having brain metastases were collected. At the same time, routine
two-dimensional fast spin echo (2D FSE) sequences (enhanced 5 mm-T1FLAIR and enhanced 1 mm-T1FLAIR) and MATRIX sequence
scans (enhanced black blood) were performed based on 3.0T magnetic resonance (1 mm-MATRIX). 28 patients with clinically
confirmed brain metastases were followed up for 3-6 months after surgery or treatment. Divided into 3 groups according to the
diameter of the lesions (≤3 mm, 3-10 mm, ≥10 mm), and compared the number of brain metastases detected in each sequence with
the detection rate according to the diameter of the lesions.
Results or Findings: The results were (1) 28 patients had a total of 752 brain metastases. The total number of brain metastases
detected by enhanced 5 mm-T1FLAIR, enhanced 1 mm-T1FLAIR, and enhanced dark blood. 1 mm-MATRIX were 604, 656, and 752,
respectively. Among them, there were 258, 263, 347, 291, 337, 349, 55, 56 and 56 with diameters ≤3 mm, 3-10 mm, and ≥10 mm,
respectively. Furthermore, (2) the number of detections of enhanced 5 mm-T1FLAIR, enhanced 1 mm-T1FLAIR and enhanced black
blood 1 mm-MATRIX was compared in pairs. Finally, (3) the detection rates of the three series of lesions with diameters ≤3 mm, 3-10
mm, and ≥10 mm were tested by chi-square tests. Lesions with diameters of ≤3 mm, 3-10 mm, and ≥10 mm were detected.
Comparison of the linear trend of the rate.
Conclusion: The detection rate of enhanced dark blood 1 mm-MATRIX is higher than that of enhanced 5 mm-T1FLAIR and 1 mm-
T1FLAIR, especially for the detection of brain metastases with a diameter of ≤3 mm.
Limitations: Only some brain metastases were confirmed pathologically and there were no cases of meningeal metastases.
Ethics committee - additional information: This study was reviewed and approved by the Medical Ethics Committee (No. 2022-
KY-0063-001).
Histogram analysis of perfusion and diffusion MR metrics in predicting consistency of meningiomas (7 min)
Lingmin Zheng; Fuzhou / China
962
Author Block: L. Zheng1, H. Zheng1, D. Lin1, Y. Song2, Y. Xue1, L. Lin1; 1Fuzhou/CN, 2Shanghai/CN
Purpose: The consistency of meningiomas is a critical factor in determining the required surgical instruments as well as affecting the
outcome of surgery. This study aimed to compare MR metrics derived from different modalities (DKI, DTI, ASL and DSC) in predicting
the consistency of meningiomas.
Methods or Background: 77 consecutive patients with histopathologically confirmed meningiomas were prospectively enrolled in
this study. Two neurosurgeons evaluated the tumour consistency and classified them as soft and hard groups. A volume of interest
was placed on the preoperative MR diffusion images to outline the whole tumour area. Histogram parameters were extracted from
perfusion and diffusion maps. Histogram parameters that were found to be related to tumour consistency in the univariate analysis
would then be further included in backward stepwise logistic regression analysis to build combined models for each modality. The
diagnostic performance of each model was evaluated by receiver operating characteristic analysis. The DeLong test was used to
compare AUCs.
Results or Findings: Histogram parameters of DTI metrics (DA, and FA), ASL (CBF) and DSC (Tmax, rCBV and rCBF) were found to
be significantly related to tumour consistency and then included in combined models (P<0.05). None of the DKI metrics can
significantly differentiate soft and hard meningiomas. The DSC combined model yielded the highest AUC of 0.858. The DTI combined
model had a relatively lower AUC value of 0.810, while the AUC of the ASL combined model was only 0.648. The DeLong test indicated
that there was no significant difference between DTI and DSC model diagnostic performance.
Conclusion: DSC and DTI metrics are feasible in predicting meningioma consistency via histogram analysis.
Limitations: This study was limited by the small sample size, the subjective classification of tumour consistency, and the lack of
histopathological evidence.
Funding for this study: This research was supported by the Fujian Provincial Health Technology Project (Grant number:
2020GGA039).
Ethics committee - additional information: This study was approved by the Ethics Committee of Fujian Medical University Union
Hospital assessed and approved this study (Grant number: 2020WSJK036)
963
RPS 2214 - A focus on MRI and mammography practice and care
Categories: Breast, Imaging Methods, Radiographers

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderators:
Cecilia Aynes; Sabadell-Barcelona / Spain
Eugen Divjak; Zagreb / Croatia
Comparing radiologists' and radiographers' assessment of quality and justification of lower back MRI referrals (7 min)
Catherine Chilute Chilanga; Drammen / Norway
Author Block: C. C. Chilanga1, M. Heggelund1, E. Kjelle2; 1Drammen/NO, 2Gjøvik/NO

Purpose: This study aimed to compare radiologists' and radiographers' assessment of the quality of Magnetic Resonance Imaging
(MRI) lumbosacral spine referrals for low back pain (LBP).
Methods or Background: A total of N=300 lumbosacral spine MRI referrals for LBP for adults aged 16–90 years were obtained. A
registration form was designed consisting of seven statements on relevant information in the referral rated on a three-point scale:
'Yes', 'No' or 'Uncertain'. In addition, the referral was rated as 'Justified', 'Unjustified' or 'Need more information'. The registration form
was pilot-tested on two radiologists to check for validity and reliability. Four radiologists and two radiographers were recruited and
assigned the task to assess the same 300 referrals individually and complete the registration form. STATA Statistics packages were
utilised to perform mixed model analysis to show variability in ratings between groups.
Results or Findings: Overall, an average of 65% (n=196) of the referrals were assessed as justified. There was a difference
between radiographers and radiologists in the assessment of referral justification with 73.5% (95% CI [69.4%-77.7%]) and 58.4%
(95% CI [54.5%-62.1%]) rated 'Justified' respectively. This variation was statistically significant (p-value <0.001). In referral quality, an
average of 7% (n=22) received a ‘good’ quality score. Compared to the radiologists, the radiographers rated a statistically significant
(p<0.001) higher percentage of the referrals as good quality, with radiologists rating 6.2% (95% CI [4.5%-8.0%] as good quality and
radiographers rating 10% (95% CI [7.3%-12.6%]) to be of good quality.
Conclusion: The radiographers rated more LBP MRI referrals as 'Justified' and gave the referrals an overall higher quality score than
radiologists.
Limitations: The limitations of the study are that data is only from a few raters, with an added unequal amount representation of
professional groups, limiting generalising of the findings.
Funding for this study: Funding was provided by The Research Council of Norway (Project number 302503).
Ethics committee - additional information: The study was approved by the Regional Committees for Medical and Health
Research Ethics (REK) reference number 378396 and the Norwegian Agency for Shared Services in Education and Research (SIKT)
reference 261461.
The most common patient-related and technical artefacts on breast MRI: how to recognise and overcome? (7 min)
Zeljka Ljepoja; Belgrade / Serbia
964
Author Block: Z. Ljepoja, M. M. Nadrljanski, I. B. Krušac, D. Dimitrijevic, L. Raspopovic, A. Djajic, M. Mihajlovic; Belgrade/RS
Purpose: This study aimed to recognise the most common artefacts and misinterpretations of the objects in the field of view: patient-
related (motion artefacts, artefacts due to positioning, metal artefacts) and technical (zipper artefacts, wrap-around artefacts,
chemical shift artefacts and zebra artefacts).
Methods or Background: 300 consecutive breast MRI exams, all realised with a full diagnostic protocol (T2W-STIR, T2W-TSE, T1W-
TSE, FLASH 3D with the application of the same contrast medium: gadobutrol) on 1.5T MRI unit (realised: October 2020 –October
2023) were analysed for the preselected sets of artefacts (patient-related and technical).
Results or Findings: Artefacts were detected on 19.67% of all analysed MRI breast exams. Artefacts predominantly belonging to
patients were 84.74%. Motion artefacts were 45.76%; artefacts due to positioning were 23.73%; metal artefacts were 15.25%. Only
15.25% of all artefacts were technical artefacts. Zebra artefacts were 5.08%; wrap-around artefacts were 3.39%; zipper artefacts
were 3.39%, and chemical shift artefacts were 3.39%. Significantly more patient-related artefacts were detected (motion artefacts,
p=4.45e-7), with the pval for distribution of 0.046, favouring the presence of motion artefacts. The diagnostic interpretation was
affected by patient-related motion artefacts with 7 exams being rescheduled and repeated, i.e. 11.86% of all detected artefacts.
Conclusion: Technologists and radiologists need to recognise and understand the artefacts on breast MRI in order to provide the
satisfactory and permanent quality of the images. Adequate patient preparation is important for the adequate image quality.
Limitations: This was a retrospective analysis.
Ethics committee - additional information: This was technical research, exempted from the decision of the ethical committee.
A safety questionnaire to assess participant (paediatric and adult) experience of ultra-high-field MRI: theory, design,
and implementation considerations (7 min)
Philippa Bridgen; London / United Kingdom
Author Block: P. Bridgen1, K. Colford1, B. Hansson2, I. M. Björkman-Burtscher3, T. Arichi1, S. Malik1, S. Giles1, C. Malamateniou1, G.
Turner1; 1London/UK, 2Lund/SE, 3Gothenburg/SE
Purpose: The aim of this research was to create a robust, user-informed, evidence-based questionnaire for patient safety aspects at
7T, using local expertise and previous literature.
Methods or Background: 7T MRI increases signal-to-noise-ratio (SNR) and improves contrast in comparison to standard magnetic
field strengths, giving the potential for additional information clinically. However, patients can experience transient sensory effects
during 7T examinations, which may impact patient experience and acceptability. Previous research has primarily focused on adult
perception, with children perceptions so far only being extrapolated from 3T data. Expanding these questionnaires to children to gain
further necessary information is needed. Literature searches were carried out looking for both MRI transient effects and questionnaire
designs at all MRI field strengths. Content analysis of 32 articles was completed to identify common themes, directed the subject of
questions asked, following the patients’ MRI-scan journey. Answer formats included free comments and 5-point Likert scales.
Questionnaires were adapted to be age-appropriate for 5-8 year and 8-11 olds, and adults. Language level was verified using the
Flesch-Kincaid method. All questions were piloted (n=10) to gain feedback from intended users on content, design, and flow.
Results or Findings: Three comparable age-appropriate questionnaires were constructed, reflecting identified common themes
from literature. Pictures aided understanding for children aged 5-8 years old. Questionnaires were divided into six sections: initial
overview, positioning, entering, during the scan, exiting and post scan.
Conclusion: Newly designed questionnaires will allow a better understanding of how children or adults may experience 7T MRI and
enhance safety strategies. Evidence collected from future use will support change of current practice.
Limitations: 7T is self-limiting due to the narrow scope, as such there is the potential for a lack of diversity of participants.
Ethics committee - additional information: Ethical approval granted from Committee of the School of Health Sciences, City,
University of London under ETH2223-1703
Comparison between 0.3 Tesla and 1.5 Tesla MRI modalities in knee imaging (7 min)
Matej Jurjević; Kisovec / Slovenia
965
Author Block: M. Jurjević, N. Mekis, J. Izlakar; Ljubljana/SI
Purpose: This study compared knee imaging at 0.3T and 1.5T field densities. We examined differences in signal-to-noise ratio (SNR),
contrast-to-noise ratio (CNR), and image quality assessed by three experienced radiologists.
Methods or Background: A sample of 25 left knees was examined using both MR devices with a 3 mm slice thickness. The
volunteers were healthy and had no previous knee injuries. SNR and CNR measurements were performed on the medial meniscus, the
distal part of the femur, articular cartilage, and the background. In the second part of the study, three radiologists assessed the image
quality of the ACL, PCL, menisci, articular cartilage, and the overall image.
Results or Findings: In the SNR and CNR measurements, we recorded statistically significant differences in the area of the medial
meniscus in favor of the 1.5T modality for both SNR (p < 0.001) and CNR (p < 0.001). Similar results were observed in the area of the
distal part of the femur, with better values for the 0.3T modality for both SNR (p < 0.001) and CNR (p < 0.001). However, in the
measurements of the distal femur, we did not find statistically significant differences in the values of SNR (p = 0.677) and CNR (p =
0.861). For all selected structures, radiologists rated the 1.5T modality higher, and this is supported by the statistically significant
differences. The agreement between radiologists was moderate for the ratings of menisci, articular cartilage, and the overall image,
and poor for the quality of cruciate ligaments.
Conclusion: Our results show superior image quality on 1.5T MRI, while indicating the potential for 0.3T open-type scanners in knee
diagnostics.
Limitations: Smaller FOV in one plain on 0.3T to reduce wrap-around artifact. Surface array coil used on 1.5T scanner.
Funding for this study: Funds for this study were obtained from the Institute of Oncology Ljubljana, Medisken Trbovlje.
Ethics committee - additional information: This study received ethical approval from the National Medical ethics committee,
approval number: 0120-356/2022/3.
Effect of noise exposure during 7 Tesla MRI (7 min)

Linda Maria Viviann Wennberg; Bjärred / Sweden
Author Block: L. M. V. Wennberg, P. C. Maly Sundgren, S. Waechter, J. Brännström, A. Jönsson, B. Hansson, J. Mårtensson; Lund/SE
Purpose: This study aimed to assess the potential impact of noise exposure during a 7 Tesla (T) brain MRI in healthy adults.
Methods or Background: Excessive noise can harm the cochlear outer hair cells, leading to auditory damage. In this study, we used
otoacoustic emission (OAE) to evaluate the effects of noise exposure in 39 healthy adults after a 7T MRI scanning session utilizing
currently accepted hearing protection. The participants were enrolled in a research project involving two one-hour MRI scanning
sessions on the same day. OAE assessments were performed before and after each scan, with a follow-up performed one week later.
Results or Findings: Our analysis revealed no significant differences in outer hair cell function between the baseline measurements
and the first MRI scan. A significant difference was observed at 1.5 kHz and 2 kHz in the left ear, as well as at 4 kHz in the right ear
after the second MRI scan. However, the follow-up OAE measurement showed no significant difference compared to the baseline at
any of the frequencies in either ear.
Conclusion: Our study's findings suggest no lasting effects on outer hair cell function in adults who undergo two one-hour MRI
scanning sessions in a single day while using appropriate hearing protection.
Limitations: It should be noted that the participants in this study were restricted to healthy young adults who met specific criteria.
Therefore, further investigations involving a more diverse group of individuals would be beneficial to gain a more comprehensive
understanding of the impact of 7T MRI scanning sessions on different populations.
Funding for this study: This work was supported by grants from the Swedish Research Council (2017-00896) and the LMK
Foundation. Both were awarded to Johan Mårtensson.
Ethics committee - additional information: This study was approved by the local ethics committee (registration numbers
2019-05387, 2016/126, and 2020-06907).
Supplementary breast cancer screening in women with dense breasts: insights from European radiographers and
radiologists (7 min)
Deborah Mizzi; Msida / Malta
966
Author Block: D. Mizzi1, C. Allely2, F. Zarb1, C. Mercer2; 1Msida/MT, 2Salford/UK
Purpose: This study explored the understanding of challenges and requirements for implementing supplementary breast cancer
screening for women with dense breasts among clinical radiographers and radiologists in Europe.
Methods or Background: Fourteen semi-structured online interviews were conducted with European clinical radiologists (n=5) and
radiographers (n=9) specializing in breast cancer screening from eight different countries including United Kingdom, Malta, Italy, the
Netherlands, Greece, Finland, Denmark and Switzerland. The interview schedule comprised questions regarding professional
background and demographics and 13 key questions divided into six subgroups, namely supplementary imaging; training; resources
and guidelines; challenges; implementing supplementary screening and the women’s perspective of supplementary imaging. Data
analysis followed the six phases of reflexive thematic analysis.
Results or Findings: Six significant themes emerged from the data analysis: understanding and experiences of supplementary
imaging for women with dense breasts; challenges and requirements related to training among clinical radiographers and
radiologists; awareness among radiographers and radiologists of guidelines on imaging women with dense breasts; challenges to
implement supplementary screening; predictors of Implementing supplementary screening and Views of radiologists and
radiographers on women's perception towards supplementary screening.
Conclusion: The interviews with radiographers and radiologists provided valuable insights into the challenges and potential
strategies for implementing supplementary breast cancer screening. These challenges included cost and logistics problems and
patient and staff related challenges. Implementing multifaceted solutions such as Artificial Intelligence integration, specialised
training and resource investment can address these challenges and promote the successful implementation of supplementary
screening in women with dense breasts. Further research and collaboration are needed to refine and implement these strategies
effectively.
Limitations: The data collection period coincided with the reopening of screening units after COVID-19 closures. During this period,
participants were exceptionally busy, which limited their availability to partake in the study.
Funding for this study: This work is part of a PhD programme which is part-financed by the Tertiary Education Scholarship Scheme
(TESS), Government of Malta (TESS Contract MEDE 417/2018/61).
Ethics committee - additional information: Ethical permission was attained from the University of Salford School of Health and
Society Research Ethics Committee.
Mean glandular dose in mammography with breast implants (7 min)

Erna Alukić; Ljubljana / Slovenia
Author Block: E. Alukić, A. Simunovic, N. Mekis, M. Marolt Music; Ljubljana/SI

Purpose: The aim was to evaluate the mean glandular dose (MGD) in screening mammography in patients with breast implants and
to investigate the differences in MGD depending on the type of exposure setting - automatic exposure control (AEC) or manual
exposure technique.
Methods or Background: A retrospective study with secondary data analysis included 536 patients with breast implants who
completed screening mammography between 2008 and 2021. MGD, breast thickness, and type of exposure setting were assessed
and compared using standard mammography and the Eklund technique. Image quality was assessed by three radiologists
experienced in mammography assessment.
Results or Findings: MGD is statistically significantly lower for breast thicknesses of 6 < 8 cm, 8 < 10 cm, and > 10 cm for standard
images with manual exposure technique. For additional images with the AEC system, the mean MGD is statistically significantly lower
for breast thicknesses of < 4 cm and 4 < 6 cm. For additional images with breast thickness of 6 < 8 cm, the MGD value is significantly
lower with the manual exposure technique. There are statistically significant differences in MGD between images taken with standard
mammography and those taken with the Eklund technique for both types of exposure settings. The overall assessment of all quality
criteria images was statistically significantly higher for the manual exposure technique.
Conclusion: For additional images with breast thickness < 4 cm and 4 < 6 cm, the AEC system is better due to the lower MGD
values. The manual mode is more suitable for breast thicknesses of 6 > 8 cm.
Limitations: No limitations.
Ethics committee - additional information: This study was ethically approved by the Commission of the Republic of Slovenia for
Medical Ethics, Document number: ERIDEK-0047/2022.
Assessing the potential of digital breast tomosynthesis as a primary breast cancer screening tool in Malta: identifying
beneficiaries and advancements (7 min)
Maria Pule; Paola / Malta
967
Author Block: M. Pule1, D. Mizzi1, K. Borg Grima2; 1Msida/MT, 2Naxxar/MT
Purpose: The aim of the study was to prospectively identify who would benefit from having a digital breast tomosynthesis (DBT) as a
first line screening. This was done by identifying different characteristics of patients being referred for a DBT at the local screening
centre during further assessment clinics. The objectives of phase 1 were to audit the referrals for further assessment clinics within the
local screening centre. In phase 2, data on the reason of referral together with the different patient characteristics, and the imaging
results was collated to be able to reach set objectives.
Methods or Background: The study involved two sequential phases. Phase 1 consisted of a retrospective analysis of statistical
figures on the use of DBT locally from 2015 to 2021. Phase 2 was a prospective study composed of a self-designed patient
questionnaire distributed between March and April 2022, required to evaluate any link between the different patient characteristics
and the clinical outcome following a DBT.
Results or Findings: Phase 1 included 2,756 cases, where 35.9% had their first mammogram while 64.1% had a subsequent
mammogram. First time mammogram cases were statistically the most likely to be returned back to normal screening (37.8%
n=990). In both phases, asymmetric densities was the most common reason of referral. For phase 2, 53 participants were recruited.
Results indicated that the most common types of Mammographic Breast Densities were type BIRADS A (30.2 %) and B (50.9%).
Conclusion: From the results collected in both phases it was shown that women screened for the first-time would benefit from DBT
as a first line screening tool since they are more likely to be returned back to normal screening.
Limitations: Due to COVID-19 restrictions, the sample population for phase 2 was limited.
Ethics committee - additional information: Approval for this research was then obtained from the Faculty Research Ethics
Committee (FHS-2021-00023) and from the University Research Ethics Committee within the University of Malta.
968
RC 2217 - Vascular trauma in the abdomen and pelvis: a survival guide for the emergency
Categories: Emergency Imaging, Interventional Radiology, Vascular

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Mariano Scaglione; Napoli / Italy

Mariano Scaglione; Napoli / Italy
Abdominal vascular trauma: solid organs and mesentery (15 min)

Elizabeth Dick; London / United Kingdom
1. To develop a visual checklist for active bleeding in the solid organs and mesentery.
2. To understand the value of communicating subtle signs of mesenteric trauma to the clinical team so they can be alert to the
development of bowel injury.
3. To recognise different patterns of arterial and venous bleeding, including contained bleeding.
Pelvic vascular trauma (15 min)

Sameer Bhimjibhai Raniga; PO Box 38 PC 123, Al khoud / Oman
1. To understand and optimise different CT trauma protocols to detect, describe and discriminate different types of pelvic vascular
injuries on a polytrauma PAN-CT.
2. To utilise a systematic search pattern to identify different pelvic vascular injuries without overcalling or missing important injuries.
3. To recognise the role of CT in triaging and managing patients, determining whether they should be referred to interventional
radiology (IR) or the operating room (OR) pathway.
Role of IR in abdominopelvic vascular trauma (15 min)

1. To discuss the role of IR in the multidisciplinary trauma team.

2. To describe updated principles and techniques used in transcatheter embolisation in trauma.
3. To interpret expected and unexpected CT imaging findings after transcatheter embolisation.
Panel discussion: Plugging the leak: the ins and outs of the body interventionalist (10 min)
969
RC 2201 - Evaluation of liver metastases
Categories: Abdominal Viscera, Imaging Methods, Multidisciplinary, Oncologic Imaging

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Assessing response to systemic therapy in liver metastases (15 min)

Federica Vernuccio; Palermo / Italy
1. To summarise the role of CT and MRI for the assessment of response to systemic therapy in liver metastases.
2. To discuss established tumour response criteria and other morphological signs of response to systemic therapy in liver metastases.
3. To describe changes in the non-tumorous hepatic parenchyma after systemic therapy.
Imaging before and after locoregional treatment of liver metastases (15 min)
Monique Maas; Amsterdam / Netherlands
1. To know what to report in patients considered for locoregional treatment of liver metastases.
2. To know how to follow up with patients after liver ablations.
3. To identify normal and abnormal imaging findings after liver ablations.
Radiology before surgery: what do the surgeons need from us? (15 min)
Arturs Ozolins; Riga / Latvia
1. To explain different surgical techniques for liver metastases resection.

2. To report the imaging information necessary to plan hepatic metastases resection.
3. To know imaging findings that preclude a successful hepatic metastases resection.
Panel discussion: What are the main challenges of imaging in treatment planning and response assessment? (10 min)
970
E³ 2226 - Current applications in imaging-tailored therapy
Categories: Breast, Imaging Methods, Interventional Radiology, Oncologic Imaging

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Image adjusted post-NAC treatment and therapy (15 min)

1. To understand MRI's interpretation and clinical applications in post-NAC treatment.

2. To list causes of over- and underestimation of residual disease at MRI.
3. To discuss the advantages and limitations of different post-NAC therapy.
Update on image-guided therapy (15 min)

Alexandra Athanasiou; Athens / Greece
1. To understand the role of breast imaging in image-guided therapy.

2. To present updates on current techniques.
3. To understand patient selection criteria and eligibility for image-guided therapy.
4. To discuss future options.
Image-guided de-escalation of treatment: the PROSPECT trial paradigm shift (15 min)
Bruce Mann; Parkville / Australia
1. To consider the current treatment protocols for the omission of radiotherapy in early breast cancer (EBC).
2. To reflect on the nature of EBC, including the existence and potential importance of undiagnosed cancers occult on conventional
imaging.
3. To differentiate minimal/mild BPE from moderate/marked BPE on breast MRI and predict how it will affect the identification of
synchronous occult cancers, thereby impacting the ability to de-escalate treatment.
4. To adapt existing knowledge of local staging of breast cancers with MRI to enable identification of unifocal EBCs for which
radiotherapy may be safely omitted and multifocal/centric disease for treatment intensification.
Panel discussion: How does surgery look like in the near future? (10 min)
971
RC 2203 - Guidelines and patient management for stable coronary artery disease
Categories: Cardiac, Contrast Media, Imaging Methods, Professional Issues

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Guidelines for non-invasive imaging of stable coronary artery disease (15 min)
1. To know the most important guidelines for non-invasive imaging of stable coronary artery disease.
2. To discuss when non-invasive imaging (CT and MRI) is indicated in the diagnosis of stable coronary artery disease.
3. To become familiar with how the findings on non-invasive imaging dictate patient management.
CT for risk prediction and patient management in stable coronary artery disease (15 min)
1. To understand in which patients' CT has the highest diagnostic accuracy.

2. To get a systematic review overview of CT trial evidence.
3. To learn about the unmet needs in coronary CT imaging.
MRI for risk prediction and patient management in stable coronary artery disease (15 min)
Eike Nagel; Frankfurt am Main / Germany
1. To understand the strengths and limitations of MRI for stable coronary artery disease.
2. To learn about research evidence for MRI in stable coronary artery disease.
3. To elucidate the potential role of MRI in the management of the management of patients with stable coronary artery disease.
Panel discussion: For patients with stable chest pain, when should we use CT and when should we use MRI? (10 min)
972
E³ 24E - New horizons for hybrid imaging

ETC Level: LEVEL II
Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Total body PET/CT: a revolution for clinical routine? (15 min)

Axel Rominger; Bern / Switzerland
1. To know about the advantages of total body PET.

2. To carve out the central indications where total body PET contributes to optimized therapy guidance.
3. To discuss applications of PET multiparametric imaging.
The five future key applications of AI in hybrid imaging (15 min)

1. To know about the main fields where AI currently contributes to clinical routine.
2. To understand the benefits and risks of currently available algorithms.
3. To appreciate two future scenarios of how AI changes a radiologist’s job profile.
The magic forest: novel radiotracers on their way to clinical translation (15 min)
Lena Unterrainer; Munich / Germany
1. To know about the challenges of clinical tracer development.

2. To discuss a range of novel radiotracers with high translational potential and their potential clinical benefits.
3. To embrace an example of a once-promising radiotracer that got stuck in the pipeline, never made it and why?
Panel discussion: Innovations in hybrid imaging – which one will turn out as the most transformative? (10 min)
973
BS 22b - Genitourinary
Categories: General Radiology, Genitourinary, Imaging Methods

Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:

Imaging of the kidney (19 min)

Carlos Nicolau; Barcelona / Spain
1. To present the current imaging techniques for evaluating the kidney.

2. To demonstrate the most important findings of common kidney pathologies.
Imaging of the prostate (19 min)

Olivier Rouviere; Lyon / France
1. To present the current imaging techniques for evaluating the prostate.

2. To demonstrate the most important findings of prostate pathologies.
Imaging of the testis and penis (19 min)

Athina C Tsili; Ioannina / Greece
1. To present the current imaging techniques for evaluating the testis and penis.
2. To illustrate the imaging features in testicular and penile pathologies.
974
BS 22a - Thorax: all you need to know in daily clinical hospital practice
Categories: Chest, Imaging Methods

ETC Level: LEVEL II
Date: March 3, 2024 | 08:00 - 09:00 CET
CME Credits: 1
Moderator:
Jens Vogel-Claussen; Hannover / Germany
Chairperson’s introduction (3 min)

Jens Vogel-Claussen; Hannover / Germany
CT imaging of common infectious diseases (19 min)

Andreas Christe; Bern / Switzerland
1. To become familiar with the most common infectious aetiologies.

2. To demonstrate the most important imaging findings
Tips and tricks about pulmonary nodules (19 min)

1. To become familiar with pitfalls and the most common aetiologies.

2. To demonstrate the most important imaging findings.
Imaging recommendations and typical appearance of most frequent fibrosing interstitial lung diseases (19 min)
Sujal R. Desai; London / United Kingdom
1. To become familiar with the recommendations of pulmonary fibrosis and the most common aetiologies.
2. To demonstrate the most important imaging findings.
975
E³ 2323 - Cardiac and vascular
Categories: Cardiac, Vascular

Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

Cardiovascular imaging: coronary arteries (28 min)

Matthias Eberhard; Unterseen / Switzerland
1. To understand the anatomy, normal variants, and abnormalities of the coronary arteries.
2. To describe the technical aspects and methodology of cardiac CT.
3. To understand the clinical role of cardiac CT in the main clinical scenarios: coronary stenoses and imaging post-revascularization.
Cardiovascular imaging: myocardium (28 min)

1. To describe the diagnostic evaluation and imaging presentation of ischaemic heart disease.
2. To understand the diagnostic evaluation and imaging presentation of myocarditis.
3. To become familiar with the heterogeneity and corresponding main imaging findings of myocarditis.
Cardiovascular imaging: valves, endocardium, and aorta (28 min)

1. To recognise the imaging presentation of the different forms of valvular disease.

2. To understand the causes and imaging presentations of endocarditis.
3. To describe the diagnostic evaluation and imaging presentation of common diseases of the great vessels, including aortic
dissection and aneurysms.
976
E³ 2321 - Imaging of metastatic renal tumours

ETC Level: LEVEL II
Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Staging strategies (45 min)

1. To illustrate the TNM system for staging renal cancer.

2. To review the best imaging protocols for evaluating metastatic renal tumours at CT/MRI.
3. To learn how to integrate different imaging techniques.
Evaluation of medical treatment (45 min)

Sandy Van Nieuwenhove; Brussels / Belgium
1. To illustrate the challenges in assessing the response to medical treatment of metastatic renal cancer.
2. To learn how to apply different objective response criteria for categorising treatment responses.
3. To discuss the role of radiogenomics in predicting treatment response and prognosis.
977
RPS 2311 - Stroke: improving diagnosis through imaging
Categories: Artificial Intelligence & Machine Learning, Interventional Radiology, Neuro, Research, Vascular
Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Katarzyna Sklinda; Warsaw / Poland
Cerebral collaterals are associated with pre-treatment brain-blood barrier permeability in acute ischaemic stroke
patients (7 min)
Alexandre Bani Sadr; Lyon / France
Author Block: A. Bani Sadr, L. Mechtouff, M. Hermier, C. De Bourguignon, A. Martin, E. Tommasino, T-H. Cho, N. Nighoghossian, Y.
Berthezene; Lyon/FR
Purpose: The aim of this study was to investigate the relationship between cerebral collateral status and blood-brain barrier (BBB)
permeability at admission MRI in a cohort of acute ischemic stroke (AIS) patients treated with mechanical thrombectomy.
Methods or Background: The HIBISCUS-STROKE cohort is a single-centre observational study that prospectively included AIS
patients with anterior circulation occlusion treated with mechanical thrombectomy. Admission dynamic-susceptibility MRI were post-
processed to provide K2 maps with arrival-time correction as a marker of BBB permeability. After co-registration with ADC maps, 90th
percentiles of K2 were extracted in the infarct core and normalised to contralateral white matter. Cerebral collateral status was
assessed by the hypoperfusion intensity ratio (HIR). Good collaterals were defined by HIR<0.4. Multiple variable logistic regression
analysis was adjusted to investigate factors associated with poor collaterals.
Results or Findings: One hundred and fifty seven patients were included (67.916.6 years, 52.8% male) with a median HIR of 0.43
(interquartile range (IQR): [0.32; 0.65]) and a median K2 of 1.67 (IQR: [0.45; 65.94]). Patients with poor collaterals (n=81, 51.6%) had
worse NIHSS score (P=0.01), larger infarct core (P<0.0001) and higher K2 (median: 3.1, IQR: [0.5; 85.1] versus 1.2; IQR: [0.5; 3.4];
P=0.002). They were less likely to achieve successful recanalization (P=0.004) and had a higher rate of haemorrhagic transformation
(P=0.02). On multiple variable analysis, poor collaterals were independently associated with larger infarct core volume (odds ratio
(OR)=1.10; 95% confidence interval (CI): [1.06; 1.13]; P=0.002) and K2 (OR=1.46; 95% CI: [1.18; 1.84]; P=0.004).
Conclusion: Poor collaterals are independently associated with a larger infarct core and increased BBB permeability at admission
MRI.
Limitations: This study was a retrospective analysis of a single-centre cohort, limiting its scope.
Funding for this study: This study was funded by the RHU MARVELOUS (ANR-16-RHUS-0009) of Université de Lyon, within the
program “Investissements d'Avenir” operated by the French National Research Agency.
Ethics committee - additional information: This study was ethically approved by the IRB number: 00009118.
Accuracy of artificial intelligence for diagnosing intracranial haemorrhage: results of a year-long multicentre clinical
monitoring study (7 min)
Anna Nikolaevna Khoruzhaya; Moscow / Russia
978
Author Block: A. N. Khoruzhaya, K. M. Arzamasov, D. V. Burenchev, E. I. Kremneva, A. Vladzymyrskyy, Y. Vasilev; Moscow/RU
Purpose: The objective of this study was to evaluate the performance metrics of an artificial intelligence (AI) service aimed at
diagnosing intracranial haemorrhage on head CT during a year-long multicentre clinical follow-up.
Methods or Background: As part of the Moscow Computer Vision Experiment, an AI service with known characteristics obtained
from calibration testing (sensitivity - 0.89, specificity - 0.96, ROC AUC - 0.96) was connected on 28 April 2022 to the CT machines of
56 inpatient medical institutions. At the end of April 2023, it had analysed a total of 133,506 head CT. In order to assess accuracy
rates, monthly clinical monitoring was conducted throughout the year, during which 1112 randomly selected head CT scans were
independently assessed by three radiologists with more than 3 years' experience, assessing the actual presence of ICH (0/1) and the
result provided by the AI service (trigger threshold 0.75). According to the results of physician judgement (GT), 440 CT studies (39%)
contained signs of ICH.
Results or Findings: Full concordance of brain CT assessment by radiologists and AI service was achieved in 57% (633 CTs), partial
concordance in 22% (249). The number of false positive responses by the AI service was 19% (212) and false negative responses
were 1% (11). Thus, sensitivity was 97.5%, specificity was 68.4%, and ROC AUC was 0.94.
Conclusion: The AI service in a year-long multicentre clinical monitoring demonstrated a higher sensitivity than calibration testing,
but lost significantly in specificity. This suggests good potential for the service to be used in acute care and to perform triage.
Limitations: The decrease in AI performance metrics should be taken into account for long-term use in practice and feedback should
be given to developers for further retraining of AI services to reduce the number of FP results.
Funding for this study: This study was funded by the Program of the Moscow Healthcare Department “Scientific Support of the
Capital's Healthcare” for 2023–2025: АААА-А21-121012290079-3.
Ethics committee - additional information: The study was conducted in accordance with the Declaration of Helsinki, and
approved by the Independent Ethics Committee of MRO RORR (protocol code 2/2020, the date of approval February 20, 2020). Clinical
trial: NCT04489992.
Deep-learning augmented contrast enhancement improves the detection of cerebral vessel occlusions in CT-
angiography of acute stroke patients (7 min)
Sebastian Steinmetz; Mainz / Germany
Author Block: S. Steinmetz, M. A. Abello Mercado, A. Sanner, M. A. Brockmann, A. Othman; Mainz/DE

Purpose: The aim of this study was to examine the impact of deep-learning augmented contrast enhancement on diagnostic
performance of poorly contrasted computed tomography angiography (CTA) in acute stroke.
Methods or Background: This retrospective single centre study included patients with suspected ischaemic stroke who underwent
comprehensive CT imaging including cranial computed tomography (CCT), whole brain volume perfusion CT (VPCT) and computed
tomography angiography (CTA) and had poorly contrasted CTA (defined as <350HU in the proximal MCA) between 01/2021 and
12/2022. 58/102 patients had vascular occlusion with correlate in perfusion. Datasets were reconstructed both conventional iterative
methods (conventional CTA) and additionally using a pre-trained deep learning model allowing selective boosting of iodine-based
contrast agents (enhanced CTA). Subjective image analysis was conducted by three readers, who rated general and vessel specific
parameters on a 4-point Likert-scale. Furthermore, they evaluated both datasets for presence / absence of cerebral vessel occlusions.
VPCT served as reference standard for calculating sensitivity and specificity.
Results or Findings: Enhanced CTA revealed significantly higher subjective image quality parameters (p<.001). Readers
significantly improved sensitivity by enhanced CTA compared to conventional CTA (Reader 1 and 3: 55/58 [95%; 95% CI: 85.62% to
98.92%] vs. 48/58 [83%; 95% CI: 70.57% to 91.41%]; Reader 2: 53/58 [91%; 95% CI: 81.02% to 97.14%] vs. 46/58 [79%; 95% CI:
66.65% to 88.83%]. Reader 1 and 3 yielded no false positive findings on conventional CTA or enhanced CTA (specificity 44/44 [100%;
95% CI: 91.96% to 100%]), reader 2 yielded one false positive on enhanced CTA (specificity 43/44) [98%; 95% CI: 87.98% to 99.94%].
Conclusion: Deep-learning augmented contrast enhancements improves image quality and diagnostic performance in poorly
contrasted CT angiographies. This could contribute to improve diagnostic certainty in acute stroke.
Limitations: Retrospective study design limits this study.
Ethics committee - additional information: This was a retrospective study and was approved by the Ethics committee,
application number: 2022-16477.
Risk identification for the development of large-artery vasospasm after aneurysmatic subarachnoid haemorrhage: a
multivariate, risk- and location-adjusted prediction model (7 min)
Julian Schwarting; Munich / Germany
979
Author Block: J. Schwarting, D. Trost, C. Albrecht, C. Zimmer, M. Wostrack, B. Meyer, J. H. W. Bodden, T. Boeckh-Behrens; Munich/DE
Purpose: Large-artery vasospasm (CVS) after aneurysmatic subarachnoid hemorrhage (aSAH) can reduce cerebral perfusion and
cause severe neurological deficits. Delayed recognition of CVS risks the success of endovascular spasmolysis. An analysis of potential
risk factors could be used to enable risk stratification for early therapeutic interventions. Therefore, the aim of this study was to
confirm established and identify unknown risk factors for CVS at the time of aneurysm occlusion.
Methods or Background: In a single-center retrospective cohort study design, we compared 853 SAH patients (mean age 57; 67%
female) between 01/2006 and 03/2020. Patients with and without CVS were compared based on demographic, clinical, and
radiographic parameters at the time of aneurysm occlusion. CVS was defined as a blood flow velocity of >200 cm/s in large
intracranial arteries or the occurrence of secondary neurological deficits without competing causes. Cohort differences were included
as predictors in a multivariate analysis to address confounding. Logistic regression models were used to determine odds ratios (OR)
for the presence of CVS for each predictor.
Results or Findings: 32% of SAH patients developed CVS. CVS was associated with age, female sex, aneurysm location, modified
Fisher score, Barrow Neurological Institute (BNI) score, and surgical interventions in univariate analysis. Multivariable regression
analysis identified multiple risk factors (Table 1). BNI score (OR: 1.33, p = 0.002), de-compressive craniectomy (OR: 1.93, p = 0.005),
and aneurysm clipping (OR: 2.22, p < 0.001), were identified as only independent risk factors after correction for age, sex, aneurysm
site and clinical severity.
Conclusion: Patients undergoing surgical interventions or patients with thick layers of subarachnoid blood should be monitored most
intensively after aneurysm occlusion for early detection of CVS and endovascular intervention.
Limitations: The most important limitation was the retrospective, single-center study design.
Funding for this study: We did not receive any 3rd party funding for this study.
Ethics committee - additional information: The study was approved by the local Research Ethics committee of the Technical
University of Munich (186/20S).
Cost-effectiveness of endovascular thrombectomy in acute ischaemic stroke with large infarct (7 min)
Julian Schwarting; Munich / Germany
Author Block: J. Schwarting1, M. Froelich2, J. S. Kirschke1, J. H. W. Bodden1, K. Dimitriadis1, J. Ricke1, C. Zimmer1, T. Boeckh-Behrens1,
W. G. Kunz1; 1Munich/DE, 2Mannheim/DE
Purpose: Endovascular thrombectomy (EVT) is the standard of care for acute large vessel occlusion stroke. Recently, the ANGEL-
ASPECT and SELECT 2 trials showed improved outcomes in patients with acute ischaemic Stroke presenting with large infarcts. The
cost-effectiveness of EVT for this subpopulation of stroke patients has only been calculated with data from the previously published
RESCUE-Japan LIMIT trial. It is therefore limited in its generalizability to an international population. The aim of this study was,
therefore, to simulate patient-level costs, analyze the economic potential of EVT for patients with large ischaemic stroke from a public
health payer perspective based on the recently published data, and identify significant determinants of cost-effectiveness.
Methods or Background: A Markov model was developed to compare outcome and cost parameters for patients treated with EVT
or only with the best medical care based on the three recent prospective clinical trials ANGEL-ASPECT, SELECT2 and RESCUE-Japan
LIMIT. Treatment outcomes were derived from the most recent literature. Deterministic and probabilistic sensitivity analyses
addressed uncertainty. Willingness to pay was set at $100,000 per quality-adjusted life year (QALY).
Results or Findings: Endovascular treatment yielded an incremental gain of 1.32 QALYs per procedure with cost savings of $17,318
per patient. Lifetime costs were most sensitive to the costs of the endovascular procedure.
Conclusion: EVT is a cost-saving (i.e., dominant) strategy for patients presenting with large ischemic cores defined by inclusion
criteria of the recently published ANGEL-ASPECT, SELECT2, and RESCUE-Japan LIMIT trials in comparison to best medical care in our
simulation. Prospective data of individual patients need to be collected to validate these results.
Limitations: Simulation based on simplified linear pathways for diagnostics and therapy with limitations derived from the availability,
quality, and validity of input variables.
Funding for this study: No third party funding was used for this study.
Ethics committee - additional information: This was a Cost-effectiveness analysis without the use of individual patient data.
Comparative evaluation of 5.0T and 3.0T TOF MRA in assessing collateral circulation in moyamoya disease: a focus on
disease progression and diagnostic precision (7 min)
Yijun Zhou; Beijing / China
980
Author Block: Y. Zhou, Y. Zhai, K. Xue, Y. Yang, D. Zhang, J. Ni, F. Feng, Y. Wang; Beijing/CN
Purpose: The aim of this study was to compare the efficacy of 5.0T and 3.0T TOF MRA in evaluating collateral circulation and
visualising abnormal vascular networks in patients diagnosed with moyamoya disease (MMD).
Methods or Background: Using both 5.0T and 3.0T MRI systems, a retrospective analysis was conducted on 30 symptomatic
hemispheres from 21 ischemic-type MMD patients (10 males, age 7–39 years). A 0-5 grading system evaluated the visibility of
moyamoya vessels across 5 regions (basal ganglion, anterior communicating artery, MCA–ICA tip, posterior communicating artery-
PCA, basilar artery tip areas). Leptomeningeal anastomoses were evaluated with scores ranging from 0 to 6, comprising three parts of
the collateral networks (pPCA→ ACA, the anterior temporal branch of the PCA anastomoses to the temporal branch of the MCA, pPCA→
MCA ). High-signal-intensity regions in basal ganglia was manually counted. Comparative analysis of moyamoya vessels and
leptomeningeal system scores between 3.0 and 5.0T MRA was executed using the Wilcoxon matched-pair signed-rank test. A paired t-
test was employed to juxtapose the number of high-signal-intensity regions between 3.0T and 5.0T MRA.
Results or Findings: The 5.0T TOF MRA demonstrated enhanced detection capabilities, showing a more significant number of
moyamoya vessels and leptomeningeal anastomoses compared to 3.0T MRA. Furthermore, 5.0T MRA was more adept at detecting
high-signal-intensity regions in the basal ganglia. Overall, 5.0T MRA provided clearer visualisation of the abnormal vascular networks
associated with MMD.
Conclusion: The 5.0T TOF MRA presents a promising diagnostic tool for MMD, offering superior visualization of abnormal vascular
networks and potentially aiding in more accurate disease assessment and prognosis.
Limitations: The parameters for 3.0T and 5.0T MRI couldn’t be perfectly matched. The small sample size might lead to false-
negative findings.
Funding for this study: This study was funded by the Major International (Regional) Joint Research Project of National Natural
Science Foundation of China (Grant Nos. 82020108018, 2020), the Beijing Natural Science Foundation (Grant Nos. Z210013, 2021)
and National High Level Hospital Clinical Research Funding (2022-PUMCH-B-027).
Ethics committee - additional information: The study was approved by the Ethics Committee at Peking Union Medical College
Hospital (Reference number: K3147).
Association of vascular risk factors and stroke subtype with cerebral small-vessel disease score (7 min)
Valentina Opancina; Kragujevac / Serbia
Author Block: V. Opancina1, M. Muto2, B. Georgievski-Brkic3, E. Ciceri4; 1Kragujevac/RS, 2Naples/IT, 3Belgrade/RS, 4Milan/IT
Purpose: The objective of this study was to explore individual MRI features of cerebral small-vessel disease (CSVD) and examine
association between vascular risk factors and stroke subtype.
Methods or Background: Cerebral small vessel disease is a chronic cerebral disease mainly involving small blood vessels, Its
incidence is increasing day by day, due to the increasing aging high incidence of cerebrovascular risk factors globally.
The study was designed as a retrospective observational multicentric cross-sectional study. The patients included in the study were
hospitalised at the Hospital for cerebrovascular diseases “Sveti Sava” in Belgrade, Serbia and AORN Cardarelli in Naples, Italy
between 2018 and 2023. Study included patients with lacunar or cortical stroke who underwent brain MRI, which was rated for the
presence of lacunes, white matter hyperintensities, cerebral microbleeds, and perivascular spaces independently. CSVD score was
calculated (range 0–4). We tested associations with vascular risk factors and stroke subtype using multivariate regression
Results or Findings: In 350 patients, multivariable analysis shown that age ([OR] 1.25, 95% confidence [CI] 1.10–1.30), male sex
(OR 1.82, 95% CI 1.15–2.35), hypertension (OR 1.70, 95% CI 1.22–2.50), and lacunar stroke (OR 2.75, 95% CI 1.80–3.74) were
significantly and independently associated with the total CSVD score.
Conclusion: The use of cerebral small-vessel disease score may be pragmatic in every day clinical practice. It could stratify risk
factors and help prevent progression of cerebral small-vessel disease in a simple and elegant way.
Limitations: The limitations of the study is the fact that we did not have neurocognitive data on all patients so these variables
couldnt be included in the study.
Ethics committee - additional information: The study was approved by Institutional Ethics Committee: 01-09/2023.
Generalisation to ultra low-field MRI of an algorithm for detection of acute cerebral infarcts (7 min)
Silvia Ingala; Copenaghen / Denmark
981
Author Block: S. Ingala1, M. Perslev1, S. Yamazaki Jensen1, K. Dmitriev2, J. Schlemper2, C. Truwit2, M. B. Bachmann Nielsen1, M. Sofka2,
A. Pai1; 1Copenhagen/DK, 2Palo Alto, CA/US
Purpose: Cerebriu apollo brain is a software designed to detect acute cerebral infarcts on standard (1.5T and 3T) magnetic
resonance images (MRIs). In this work, we aim to investigate the generalisation of the out-of-the-box software in detecting acute
cerebral infarcts on ultra-low field MRI (ULF-MRI).
Methods or Background: Background: Apollo Brain was trained on 1300 standard MRI scans to detect acute cerebral infarcts,
intracranial tumors, and intracranial haemorrhages. In this work, only the cerebral infarct detection feature is evaluated. Apollo brain
accepts T2-FLAIR and DWI sequences with an option to add either T2* or susceptibility weighted images (SWI). No additional training
or fine-tuning was performed for generalisation to ULF-MRI sequences. Methods: MRI images using Hyperfine Swoop scanner
(software version > 8.4.0) collected between July 27, 2022, and June 13, 2023 were gathered. Inclusion criteria were a human non-
contrast MRI imaging study indicating ischaemic stroke or intracranial hemorrhage (intraparenchymal or subarachnoid haemorrhage).
We excluded MRI images displaying significant motion and other artifacts. All the scans were reviewed by an experienced radiologist.
Then, Apollo brain performance was tested.
Results or Findings: A total of 30 subjects were included. Of these, 19 were diagnosed with acute cerebral infarct, including one
with signs of haemorrhagic transformation of the infarct. Apollo brain software was able to detect the infarcts with a sensitivity of 94%
and specificity of 83%. The software had one false negative , and two false positives.
Conclusion: Out-of-the-box Apollo Brain software for detection of acute intracerebral infarcts exhibited excellent performance with
ULF-MRI images. These results demonstrate the possibility for an accelerated adoption of ULF-MRI in low resource settings for
detecting acute cerebral infarcts coupled with Apollo brain software.
Limitations: We expect to improve the study by increasing the sample size.
Funding for this study: No information was provided by the submitter.
Spectral imaging and analysis of monophasic CT angiography for assessment of penumbra and infarct core in acute
stroke (7 min)
Schekeb Aludin; Kiel / Germany
Author Block: S. Aludin, L-P. Schmill, P. Langguth, O. Jansen, N. Larsen, F. Wodarg, T. Klintz, S. Seehafer, A. Horr; Kiel/DE
Purpose: CT-angiography (CTA) and CT-perfusion (CTP) are part of acute stroke imaging. CTP thereby estimates the irreversibly
damaged parenchyma, the infarct core (IC), and the potentially salvageable parenchyma, the penumbra (PEN), but requires additional
contrast and radiation. Spectral-detector-CT (SDCT) enables spectral imaging like iodine-density imaging or virtual-monoenergetic
imaging and the aim of this study was to assess its potential in differentiating IC and PEN using only monophasic CTA.
Methods or Background: Twenty patients with stroke in the media territory were analysed. Areas of IC and PEN derived from CTP-
analysis as well as their healthy hemisphere’s counterparts were transferred to spectral maps of the CTA as regions of interest. The
areas were compared to each other and between the different spectral maps (measurements of Hounsfield-Unit in monoenergetic
images (MonoE) at 40 keV, 70 keV, and 120 keV, plus iodine-density (ID) and electron-density (ED) values). Unilateral absolute values
and ratios formed to the healthy hemisphere’s values were evaluated. Visual infarct delineation was also rated in each map.
Results or Findings: IC and PEN could be distinguished from their healthy counterpart by absolute values (p<0.01). IC could be
differentiated from PEN by absolute values (p<0.0001, except for ED) and by the ratio formed to the healthy counterpart (p<0.01).
Thereby, discrimination of IC and PEN by ratio was good in MonoE40 (AUC=0.74, p<0.0001) and best in ID (AUC=0.92, p<0.0001).
Visual delineation was best in ID and MonoE40.
Conclusion: SDCT-derived spectral maps from monophasic CTA, such as ID or MonoE40, allow detection and differentiation of IC and
PEN. As a result, additional CTP may be removed from acute stroke CT protocols in the future, potentially saving time, contrast and
radiation exposure.
Limitations: The limitations of the study are the low number of individuals, thus further studies are needed.
Funding for this study: No funding was sought for this study.
Ethics committee - additional information: The study was ethically approved by the institutional review board and written
informed consent was available from all participants.
Validation of the novel combined marker Haematoma Maturity Score for intracerebral haemorrhages (7 min)
Bárbara Teresa Catelani; Rosario / Argentina
982
Author Block: B. T. Catelani1, A. Lopez Rueda2, M. E. Santos Armentia3, G. B. Harvey1, A. Pérez Fernandez3; 1Rosario/AR,
2
Barcelona/ES, 3Vigo/ES
Purpose: This study aimed to validate the novel combined indicator Hematoma Maturity Score using non-contrast computed
tomography (NCCT) scans in patients with intracerebral haemorrhage (ICH) for predicting functional outcomes at discharge, in
comparison to individual radiological NCCT signs.
Methods or Background: A retrospective analysis of patients with ICH who had undergone baseline CT scans was performed. The
Black Hole Sign, Blend Sign, Island Sign, Swirl Sign, and the Hematoma Maturity Score were independently assessed by two
radiologists who were blinded to clinical information. Patients were dichotomised based on dependency and mortality at discharge
using the Modified Rankin Scale (mRS): no symptoms or no significant/mild disability (mRS 0–2); moderate/severe disability or
mortality (mRS 3-6).
Results or Findings: Sixty-four patients with confirmed spontaneous ICH identified on NCCT were included. The Swirl Sign, Island
Sign, and Hematoma Maturity Score exhibited statistically significant associations with the clinical outcomes of the patients (p <
0.01). There was almost perfect agreement between readers regarding the maturity score (kappa = 0.84), and substantial agreement
for the other signs.
Conclusion: The novel concept of the Hematoma Maturity Score demonstrated the highest impact on clinical outcomes when
compared to other evaluated radiological signs, with almost perfect agreement between readers.
Limitations: No limitations have been identified for this study.
Ethics committee - additional information: This study has been approved by the local hospital Ethics Committee.
Improved neurovascular imaging using Advanced intelligent Clear-IQ Engine (AiCE) (7 min)
Mario Alberto Abello Mercado; Mainz / Germany
Author Block: M. A. Abello Mercado, S. Steinmetz, A. Sanner, A. Kronfeld, M. A. Brockmann, A. Othman; Mainz/DE
Purpose: The aim of this study was to evaluate the effects of deep-learning image reconstruction on image quality and diagnostic
confidence of ultra-high-resolution computed tomography (UHRCT).
Methods or Background: In this single-center study, 100 consecutive patients with acute neurological symptoms underwent CT
imaging including cranial computed tomography (CCT) and computed tomography angiography (CTA) using an ultra-high resolution
CT scanner. CTA images were reconstructed with normal resolution mode and ultra-high resolution mode using iterative
reconstruction. A deep-learning reconstruction algorithm (advanced intelligent clear-IQ engine, AiCE); specifically trained for ultra-
high resolution CT-angiography of the brain was utilized to generate a further UHR-CTA datasets (DL-UHR-CTA, matrix 1024 x 1024,
slice thickness 0.25 mm). Image quality for all three reconstructions was evaluated visually by two blinded radiologists using a 4-point
Likert-scale. Therefore, general (overall image quality, contrast in general, artifacts, diagnostic confidence and image noise) and
vessel specific (assessability of proximal, intermediate and subcortical vessels as well as perforators) criteria were assessed. The
quantitative features including slope, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), noise, entropy and co-occurrence
matrix (COOC) were examined and compared using an in-house tool.
Results or Findings: Qualitative analysis revealed highest scores for DL-UHR-CTA, followed by UHR-CTA and NR-CTA, whereas DL-
UHR-CTA yielded excellent results for all qualitative parameters and was significantly superior to UHR-CTA and NR-CTA (all p<0.001).
The quantitative analysis was in line with the qualitative findings with significantly superior results for DL-UHR-CTA (slope: p<.01,
SNR/CNR: p=0.004, entropy p<.01, COOC: p<.01).
Conclusion: Deep-Learning image reconstruction significantly improves image quality of ultra-high resolution neurovascular CT-
angiography allowing for higher diagnostic confidence, potentially improving the detection of subtile but oftentimes-significant
pathologies.
Limitations: Deep-learning image reconstruction improves the quality of UHR-CTA images, leading to higher diagnostic confidence
and potentially aiding in the detection of subtile but clinically significant pathologies.
Ethics committee - additional information: Approval was received from the Ethics committee, application number:
2021-15948:1-retrospektiv.
Assessment of cerebral blood flow distribution at different phases between different types of the circle of Willis using
4D Ffow imaging (7 min)
Binbin Sui; Beijing / China
983
Author Block: B. Sui, B. B. XiaoYan; Beijing/CN
Purpose: Haemodynamic factors have been proved to play essential roles in the formation and development of vascular diseases.
This study aimed to investigate the blood flow distribution in the different types of circle of Willis and the changes of blood flow at
different phases of the cardiac cycle using 4D Ffow imaging.
Methods or Background: Four-dimensional flow magnetic resonance imaging (4D flow MRI) was performed in 30 healthy subjects
(mean 27.5±4.1 years; range 24-38 years old). CoW was classified into five types according to the vessel anatomical structures. The
cerebral blood flow distribution at different arteries and the total cerebral blood flow (tCBF) were analyzed and compared between
different types of CoWs. The changes of cerebral blood flow during different phases of the cardiac cycle were observed and compared
between different types of CoWs.
Results or Findings: Twelve subjects (40%) were found with complete CoW, and eighteen (60%) with incomplete CoW. No
significant difference was found in tCBF between different types of CoWs (P=0.787). In subjects with a fetal-type PCA (Type V), the
average flow rates (FRavg) at contralateral ICA was higher than that of the ipsilateral ICA (P=0.003). During the peak systolic and end
diastole, compared with Type I, in subjects with unilateral PCoA opening (Type IV), the Δflow rates (ΔFR) at bilateral MCA were
significantly higher than that of Type I (P=0.012 and P=0.006, respectively); in subjects with unilateral fetal-type PCA (Type V), the
ΔFR at contralateral MCA were significantly higher than that of Type I (P=0.007).
Conclusion: Blood flow distribution exists a difference in different types of CoWs during the different phases of the cardiac cycle,
especially in type IV and type V at peak systolic.
Limitations: small samples for type V and VI.
Funding for this study: This study was funded by the Beijing Municipal Natural Science Foundation (NO. 7162056, 7212028).
Ethics committee - additional information: This study was approved by the local ethics committee of Beijing Tiantan Hospital,
and all volunteers gave written informed consent.
984
RPS 2316 - Innovation in cancer research
Categories: Imaging Methods, Oncologic Imaging, Research

Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Rectal adenocarcinoma: ex vivo 9.4T MRI: correlation with histopathologic treatment response to neoadjuvant
chemoradiotherapy (7 min)
Minglu Liu; Shanghai / China
Author Block: F. Shen, M. Liu, Y. Yuan, Y. Lu, C. Shao; Shanghai/CN

Purpose: The aim of this study was to determine the imaging detail and diagnostic information that can be obtained at 9.4T
magnetic resonance imaging (MRI) of evaluating the treatment response to neoadjuvant chemoradiotherapy (nCRT) of rectal
adenocarcinoma by ex vivo.
Methods or Background: Fifteen cases with locally advanced rectal cancer (LARC) who underwent surgical resection after nCRT
between September 2022 and February 2023 were included. Excised rectal adenocarcinoma specimens were fixed in a
perfluoropolyether-filled test tube and scanned with a 3.0T and 9.4T MRI system ex vivo. The residual tumour depth and tumour
regression grade (TRG) based on T2-weighted imaging (T2WI) were subjectively assessed and then compared with the pathological
findings.
Results or Findings: The ex vivo 9.4T T2WI without fat suppression enabled clear differentiation between tumour tissue, fibrosis
and normal rectal wall, which clearly corresponded to the pathologic tissues of the rectal specimens. The TRG could be accurately
assessed on ex vivo 9.4T images in 13/15 specimens (86.7%), while in 11/15 specimens (73.3%) on ex vivo 3.0T images.
Conclusion: Ex vivo 9.4T MR imaging enables clear delineation of the rectal wall layers and demonstrated good performance for
evaluating the TRG of LARC after nCRT, which allow radiologists to understand and then assess more accurately the treatment
response to nCRT of rectal adenocarcinoma.
Limitations: Due to this relatively novel preclinical MR system, many sequences have not yet been optimised, such as DWI and T1
mapping. Thus, majorisations of the pulse sequences may be required to make its application more technically feasible in clinical
settings in the future.
Funding for this study: This research was financially supported by National Natural Science Foundation of China (82171891),
Scientific research program of Shanghai municipal science and technology commission (21ZR1439800), Shanghai International
Cooperation Project (22490713400), Foundation of Shanghai Municipal Health Commission (202240204), “Yi Yuan Xin Xing” young
medical talents funding project of Shanghai (project's number: N/A) and the Guhai Project of Changhai hospital (GH145-09).
Ethics committee - additional information: This study was approved by the ethics committee of the Shanghai Changhai Hospital,
Naval Medical University (IRB Approval No. B2022-013).
Clinical added value of MRI to CT in patients scheduled for local therapy of colorectal liver metastases (CAMINO): an
international multicentre prospective diagnostic accuracy trial (7 min)
Burak Görgec; Amsterdam / Netherlands
985
Author Block: B. Görgec1, I. Schrøder-Hansen2, G. Kemmerich2, T. Syversveen2, J. P. Sijberden1, Å. Fretland2, C. Verhoef3, M. G.
Besselink1, J. Stoker1; 1Amsterdam/NL, 2Oslo/NO, 3Rotterdam/NL
Purpose: Guidelines are inconclusive on whether contrast enhanced magnetic resonance imaging(ceMRI) should be added routinely
to computed tomography(ceCT) in the workup of patients with colorectal liver metastases(CRLM) scheduled for curative liver
resection and/or thermal ablation. Although ceMRI is reportedly superior in the detection and characterization of CRLM, its impact on
actual clinical patient management is unknown and through this project, we aim to understand that.
Methods or Background: This was an international multicentre prospective incremental diagnostic accuracy trial in patients with
primary or recurrent CRLM scheduled for local therapy based on ceCT. All patients had ceCT and liver ceMRI including gadoxetic acid
as contrast agent and diffusion-weighted imaging(DWI). Primary outcome was change in the local clinical treatment plan based on
liver ceMRI findings. A multidisciplinary blinded expert panel performed a post-hoc analysis.
Results or Findings: Between December 2019, and July 2021, we enrolled 298 patients with CRLM planned for local therapy based
on ceCT. A change in the local clinical treatment plan based on liver ceMRI findings was observed in 92/298 patients(31%;95% CI:
26% to 36%). This concerned 40 patients(13%) requiring more extensive local therapy, 11 patients(3.7%) requiring less extensive
local therapy, and 34 patients(11.4%) in whom the indication for curative-intent local therapy was revoked, including 26
patients(8.7%) with too extensive disease and 8 patients(2.7%) with benign lesions on liver ceMRI(confirmed by follow-up). Upon
assessment by the expert panel, liver ceMRI changed clinical management in 101/297 patients(34%;95% CI: 29% to 40%).
Conclusion: Liver ceMRI changed the local treatment plan in one third of patients scheduled for intervention for CRLM based on ceCT
imaging.
Limitations: First, primary outcome change in local treatment plan could not be blinded for physicians. Second, ceCT- and ceMRI-
scan protocols were not completely standardized between participating centres.
Funding for this study: This study was funded by the The Dutch Cancer Society and Bayer AG – Pharmaceuticals.
Ethics committee - additional information: The Medical Ethical Review Board (METc) of the Amsterdam UMC, location AMC, has
assessed primarily that the CAMINO study was NOT subject to the Medical Research Involving Human Subjects Act (WMO).
Diagnostic value of very early diffusion weighted changes at MRI after single-dose ablative radiation therapy (SDART)
for organ-confined prostate cancer (7 min)
Cammillo Roberto Giovanni Leopoldo Oreste Massimiliano Talei Franzesi; Milan / Italy
Author Block: C. R. G. L. O. M. Talei Franzesi, D. G. Gandola, P. N. Franco, C. Maino, D. Panizza, R. Lucchini, S. Arcangeli, R. Corso, D.
Ippolito; Milan/IT
Purpose: The objective of this study was to investigate diagnostic value of diffusion-weighted MRI early changes, an hour after
treatment, in patients with organ confined unfavorable prostate cancer(PCa) treated with single-dose ablative radiation
therapy(SDART), in comparison with biochemical markers(PSA and testosterone).
Methods or Background: Twenty-four patients treated with a single fraction of 24Gy to the whole prostate with urethra sparing in
association with androgen deprivation therapy(ADT) were enrolled. MRI was performed before SDART(time0), one-hour post-
SDART(time1), and 3-month after treatment(time2). All the patients were examined on a 3.0-T-MRI(Ingenia;Philips Healthcare) with a
phased-array external coil, with bowel preparations and 20 mg of butyl-scopolamine(Buscopan); the bladder was filled with 120 cc of
saline solution, to simulate the same conditions during irradiation. MRI was performed with axial T1-weighted TSE and high resolution
multi-planar T2-weighted TSE sequences. DWI was acquired with six b-values (0,50,100,150,800,1600mm2/s) and Apparent Diffusion
Coefficient(ADC) (0,800) maps were calculated. Finally, axial contrast-enhanced dynamic imaging was obtained during intravenous
injection of gadobutrol. ADC values were calculated at time 0,1,and 2 by placing region-of-interests(ROI) on ADC maps and the results
were compared with PSA and testosterone blood levels at time 0 and 2.
Results or Findings: Median patient’s age was 78 years. Median prostate volume was 36.2 cc. An increase of ADC value of tumour
lesion of 27%(range 7%-69%) and 54%(range 20%-83%) was registered at time 1 and time 2 respectively, compared to the baseline.
Median prostate volume was found unchanged at time 1, while decreased by about 25%(range 9%-59%) at time 2. At 3-months
follow-up,all patients were found bNED with PSA and testosterone levels of<0.01 ng/ml and<0.20 ng/ml, respectively, and nine of
them obtained a complete response.
Conclusion: Our findings demonstrated high diagnostic value of DWI imaging with good correlation between very early changes
(one-hour after treatment) in ADC values after SDART and later tumour response (biochemical and imaging) in patients with
unfavorable PCa.
Limitations: One limitation was that this was a single-centre study.
Ethics committee - additional information: This study utilised retrospective anonymous data analysis.
Comparison between MRI and conventional CT scans in the prediction of peritoneal cancer index (7 min)
Michela Polici; Rome / Italy
986
Author Block: M. Polici, B. Masci, F. Palmeri, P. Sammartino, A. Del Gaudio, M. Zerunian, D. De Santis, D. Caruso, A. Laghi; Rome/IT
Purpose: Peritoneal cancer index (PCI) has been recognised as an independent prognostic indicator for long-term outcomes, however
the preoperative assessment represents a clinical challenge. So, the aim of the study was to compare the performance of MRI to CT
having the surgery as reference standard.
Methods or Background: Twenty-three patients affected by peritoneal carcinomatosis with availability of pre-operative MRI and CT
scans of abdomen and pelvis, were prospectively enrolled, between July 2021 and May 2023. All CT and MRI scans were performed in
the same day, within one week from the surgery. Two expert abdominal radiologists blinded, with two different experience levels
(expert and inexperienced), evaluated PCI in both MRI and CT scans. The agreements between the radiologists' assessment with
surgical results, sensitivity, and specificity were evaluated in total and per site.
Results or Findings: In total PCI, for the expert radiologist the assessment was quite better for MRI (kappa=0.89) than CT (kappa =
0.73). Similarly, for the inexperienced radiologist MRI was better (kappa = 0.79) than CT (kappa = 0.69). In the sub analysis of per
site, both expert and inexperienced radiologists showed higher agreement for MRI (kappa = 0.83 and 0.75, respectively) than for CT
(kappa=0.68 and 0.63, respectively). Overall, MRI showed higher value of per-site sensitivity and specificity compared to CT (0.95 vs
0.65 and 0.79 vs 0.75, respectively).
Conclusion: MRI shows better results in terms of PCI assessment performance compared to CT, with quite higher agreement for the
expert radiologist.
Limitations: This study was performed on a small population; it was of retrospective nature; it lacks survival analysis.
approval was obtained
CT morphological phenotype of a tumour is influenced more by the host tissue environment than tumour type (7 min)
Sajjad Rostami; Amsterdam / Netherlands
Author Block: S. Rostami, C. Guerendel, S. H. Benson, Z. Bodalal, R. G. H. Beets-Tan; Amsterdam/NL

Purpose: Tumour morphology is influenced by numerous genetic and environmental factors. Tumour type and origin, oncogenic
mutational processes, and host tissue (micro)environment all impact how a tumour grows. In this study, we aimed to assess the
effects of the tumour type and the host tissue (micro)environment on tumour morphology, using radiomics and unsupervised
clustering techniques.
Methods or Background: Manually segmented baseline CT scans of 1659 cancer patients of various tumour types were used to
extract the radiomic features of 11,164 lesions. TSNE, an unsupervised clustering approach, was used to derive the components that
encapsulated the entire tumour morphology, as measured by radiomics. We visualised all the lesions on a TSNE plot to explore the
presence of morphological clustering. To assess the association between the observed morphological clusters and our endpoints, we
applied the labels of tumour type and host tissue (micro)environment (e.g., liver, lung, bone, lymphatics, etc.) to the plot.
Results or Findings: On the basis of morphology, TSNE identified several distinct clusters. Tumour type failed to explain the
clustering pattern observed, when all the lesions (primary and metastases) were included. Looking at only primary lesions, lung
cancer formed a distinct morphological cluster apart from non-lung cancers. We observed that the morphological clusters identified
could be largely explained by the host tissue (micro)environment.
Conclusion: In this study, we utilised routinely available clinical imaging data to explore the biological influence on tumour
morphology. Considering that tumour morphology is a multifactorial attribute, the host tissue (micro)environment might substantially
affect how a tumour appears, more than its very own origin.
Limitations: External validation of the results is not done yet, limiting this study.
Ethics committee - additional information: This study was ethically approved by the IRB Code: IRBd20-213.
Patient eligibility for trials with imaging response assessment at the time of molecular tumour board presentation (7
min)
Nabeel Mansour; Munich / Germany
987
Author Block: N. Mansour, B. Westphalen, M. Bergwelt-Baildon, W. G. Kunz; Munich/DE
Purpose: The goal of this study was to assess the eligibility of patients with advanced or recurrent solid malignancies presented in a
molecular tumour board (MTB) at a large precision oncology centre for inclusion in trials with the endpoints objective response rate
(ORR) or duration of response (DOR) based on Response Evaluation Criteria in Solid Tumours (RECIST 1.1).
Methods or Background: Prospective patients with available imaging at the time of presentation in the MTB were included. Imaging
data was reviewed for objectifiable measurable disease (MD) according to RECIST version 1.1. Additionally, we evaluated the patients
with MD for representativeness of the identified measurable lesion(s) in relation to the overall tumour burden.
Results or Findings: Two hundred and sixty-two patients with different solid malignancies were included. 177 patients (68%) had
MD and 85 (32%) had non-measurable disease (NMD) at the timepoint of MTB presentation in accordance with RECIST 1.1. MD was
not representative of the overall tumour burden in eleven patients (6%). The main reasons for NMD were lesions with longest
diameter shorter than 10 mm (22%) and non-measurable peritoneal carcinomatosis (18%). Colorectal cancer and malignant
melanoma displayed the highest rates of MD (>75%). In contrast, gastric cancer, head and neck malignancies, and ovarian carcinoma
had the lowest rates of MD (<55%).
Conclusion: Approximately a third of cancer patients with advanced solid malignancies are not eligible for treatment response
assessment in trials with endpoints ORR or DOR at the time of MTB presentation. The rate of patients eligible for trials with imaging
endpoints differs significantly based on the underlying malignancy and should be taken under consideration in the planning stages of
new basket trials.
Limitations: The data is limited as a single-center study with a limited sample size, hence the representation of tumour entities may
differ in larger cohorts.
Funding for this study: This study was not supported by any funding.
with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or
comparable ethical standards. The study was approved by the institutional review board. Informed consent was waived due to the
retrospective character of the study.
Iodine maps concentration and LI-RADS classification: to a quantitative evaluation of HCC (7 min)
Alessandro Barbaro; Concorezzo / Italy
Author Block: A. Barbaro, P. A. Bonaffini, A. Celestino, G. Muscogiuri, P. Marra, S. Sironi; 24127 Bergamo/IT
Purpose: The LI-RADS classification is universally recognized as a valid instrument for evaluating hepatocellular carcinoma (HCC),
but it is a qualitative evaluation. This study aims to investigate the possibility of having a quantitative assessment of HCC, by using
the Iodine maps of dual-energy CT (DECT).
Methods or Background: This retrospective study included 75 patients with HCC who underwent DECT between March 2022 and
April 2023. Some of them underwent more than one DECT in that period, for a total of 90 exams. We included all those lesions
classified as LI-RADS 5 and all those classified as LI-RADS 3 and 4 that were subsequently confirmed as HCC by other exams such as
MRI or CEUS.
We measured the Iodine Concentration in the arterial phase (ICa), in the venous phase (ICv) and normalized them for iodine
concentration in a non-pathologic part of liver parenchyma (NICa and NICv). We calculated the ratio between NICa and NICv, and
performed a Kruskal-Willis test.
Results or Findings: The median of ICa was 2.95 mg/ml for LI-RADS 3, 3.40 mg/ml for LI-RADS 4 and 3.30 mg/ml for LI-RADS 5. The
median of ICv was 2.55 mg/ml for LI-RADS 3, 2.90 mg/ml for LI-RADS 4, and 2.40 mg/ml for LI-RADS 5.
The median of the NICa/NICv ratio was 1.793 for LI-RADS 3, 3.495 for LI-RADS 4, and 2.611 for LI-RADS 5 (P = 0,000429),
demonstrating that the NICa/NICv ratio is significantly lower in the LI-RADS 3 sample.
Conclusion: Quantitative assessment of wash-in/was-out for HCC nodules on iodine concentration maps and, particularly, the
NICa/NICv ratio may increase radiologists’ confidence in LI-RADS class allocation.
Limitations: The main limitations are the small sample size and the selection bias due to the retrospective design of this study.
Stage IIIN2 NSCLC undergoing induction chemotherapy and subsequent surgery: preliminary results of CT morphologic
prognostic factors (7 min)
Nunzia Di Meglio; Siena / Italy
988
Author Block: N. Di Meglio, A. Perrella, G. Bagnacci, V. Di Martino, L. Volterrani, L. Luzzi, M. A. Mazzei; Siena/IT
Purpose: Patients diagnosed with stage IIIN2 Non-Small Cell Lung Carcinoma (NSCLC) have various treatment options, including
surgery following induction chemotherapy (CHT) or concurrent chemoradiotherapy followed by immunotherapy. This study aimed to
identify radiological prognostic factors in a cohort of stage IIIN2-NSCLC patients who received induction CHT and subsequent surgery.
Methods or Background: Among 93 patients with locally advanced NSCLC who underwent surgical resection following induction
CHT between 2013 and 2020, we retrospectively selected 72 patients. We assessed histological parameters (pTNM stage,
pathological regression grade, mutational profile) and radiological features (tumour size, margin characteristics, location, presence of
necrosis or carcinomatous lymphangitis for T-stage, and lymph node size, location, structure, and enhancement pattern for N-stage)
from staging and restaging CT scans. We correlated these parameters with overall survival (OS) to determine their prognostic
significance. We also evaluated the performance of different radiological analysis methods
Results or Findings: The mean OS was 65.54±6.170 months, with disease recurrence occurring in 32 out of 72 cases (44.5%).
Among clinical, histological, and radiological factors, a significant correlation with OS was observed only for the presence of the KRAS
mutation (p<0.001), regression of clinical nodal stage after CHT (cN2 to ypN0/1; p 0.02), and nodal stage regression based on
multiparametric criteria (p 0.001). Multiparametric criteria exhibited high sensitivity (95%), specificity (85%), positive predictive value
(PPV) (91%), negative predictive value (NPV) (92%), and an overall accuracy of 91%.
Conclusion: Nodal downstaging after induction CHT result to be a good prognostic factor. Radiological nodal involvement could be
established with high accuracy by the multiparametric criteria at CT examination.
Limitations: The main limitation of this study is that it is of retrospective design.
Dynamic contrast-enhanced ultrasound analysis in preoperative diagnosis between hepatocellular carcinoma and
intrahepatic cholangiocarcinoma in non-cirrhotic liver (7 min)
Yi Dong; Shanghai / China
Author Block: Y. Dong, S. Chen, Y. Huang, J. Chen, Y. Wang, W-p. Wang; Shanghai/CN
Purpose: The purpose of this study was to investigate the value of dynamic contrast-enhanced ultrasound (DCE-US) and quantitative
analysis in preoperative differential diagnosis of intrahepatic cholangiocarcinoma (ICC) and hepatocellular carcinoma (HCC) in non-
cirrhotic liver.
Methods or Background: In this retrospective study, patients with histopathologically proven ICC and HCC lesions in non-cirrhotic
liver were included. All patients underwent contrast enhanced ultrasound (CEUS) examinations within one week before surgery. B
mode ultrasound (BMUS) features and CEUS enhancement patterns were analysed. DCE-US analysis was performed by VueBox®
software (Bracco, Italy). Time intensity curves (TICs) were generated, and quantitative perfusion parameters were obtained and
compared between ICC and HCC groups using the Student t test or Mann-Whitney U test.
Results or Findings: From November 2020 to February 2022, patients with histopathologically confirmed ICC (n = 30) and HCC (n =
24) lesions in non-cirrhotic liver were included. Compared with HCC lesions, TICs of ICCs showed earlier and lower enhancement
during the AP, faster decline during the PVP, and reduced area under the curve (AUC). The combined area under the receiver
operating characteristic curve (AUROC) of all significant parameters was 0.946, with 86.7 % sensitivity, 95.8 % specificity, and 90.7 %
accuracy in differential diagnosis between ICC and HCC lesions in non-cirrhotic liver, which improved the diagnostic efficacy of CEUS
(58.3 % sensitivity, 90.0 % specificity, and 75.9 % accuracy).
Conclusion: ICC and HCC lesions in non-cirrhotic liver might show some overlap of CEUS features in diagnosis. DCE-US with
quantitative analysis is helpful for preoperative differential diagnosis.
Limitations: Single centre study with relatively small sample size and retrospective nature of the study were predominant
limitations.
Funding for this study: This study was supported by National Natural Science Foundation of China (Grant No. 82071942).
Ethics committee - additional information: This was a retrospective observational study approved by the institutional review
board of our university hospital (ID: B2020-424R). The informed consent was waived. All aspects of this study were performed in
accordance with the Declaration of Helsinki.
A comparative study of the diagnostic efficacy of spectral CT iodine content analysis and three chemotherapy
evaluation criteria for non-small cell lung cancer (7 min)
Ronghua Mu; Guilin / China
989
Author Block: R. Mu1, X. M. Liu2, X. Zhu1, Z. Song2; 1Guilin/CN, 2Guangzhou/CN
Purpose: The aim of this study was to compare three short-term therapeutic response evaluation criteria of RECIST 1.1, modified
RECIST, and Choi in advanced non-small cell lung cancer (NSCLC), and analyse the correlation between iodine content in dual-layer
spectral-detector CT (DLCT) and three clinical criteria.
Methods or Background: Sixty-three NSCLC patients who received platinum-based chemotherapy and targeted therapy as first-line
treatment were retrospectively included. All patients underwent DLCT scans to calculate the difference before and after treatment in
maximum diameter (∆MD), effective maximum diameter (∆EMD), CT number (∆CT) according RECIST 1.1, modified RECIST, and Choi
criteria to assess the patients' short-term therapeutic response. The parameters of tumour iodine content in arterial and venous
phase (∆AP-ID, ∆VP-ID), normalised iodine density (∆AP-NID, ∆VP-NID) were calculated.
Results or Findings: There were statistically significant differences of three criteria in the evaluation of therapeutic response
(RECIST 1.1 vs. mRECIST: χ^2=23.204, p<0.001; RECIST 1.1 vs. Choi: χ^2=6.514, p<0.001; mRECIST vs. Choi: χ^2=21.617,
p<0.001). Choi and mRECIST criteria identified a higher proportion of patients with effective treatment response compared to RECIST
1.1 (73.0%, 54.0% vs. 32.2%). No significant correlations were found between ∆MD in RECIST 1.1 and quantitative parameter in DLCT
(all p>0.05), while significant correlations were found between ∆EMD in mRECIST and ∆VP-ID, ∆VP-NID (p<0.05), ∆CT in Choi and ∆AP-
NID, ∆VP-IDand ∆VP-NID (p<0.05).
Conclusion: The difference of iodine content during the venous phase of DLCT was an important imaging biomarker for evaluating
therapeutic response in NSCLC. Among the three evaluation criteria, iodine content was correlated with Choi and mRECIST criteria,
hold greater clinical value for early assessment of chemotherapy efficacy in advanced NSCLC patients compared to RECIST 1.1.
Body CT scan for oncologic patients: the impact of subspecialty radiology in the clinical practice: a quality care study
(7 min)
Luca Bonomo; Massagno / Switzerland
Author Block: S. Rizzo1, L. Bellesi2, A. D'Ermo2, L. Bonomo1, O. D'Ecclesiis3, S. Presilla2, E. Rezzonico1, M. Del Grande2, F. Del Grande1;
1 2 3
Lugano/CH, Bellinzona/CH, Milan/IT
Purpose: The primary objective of this study was to assess if the CT dose delivered to oncological patients was different in a
subspecialty radiology department, compared to a general radiology department. Secondary explorative objective was to assess if the
objective image quality of CT examinations was different in the two settings.
Methods or Background: Chest and abdomen CT scans performed for oncological indications were selected from a general
radiology department and a subspecialty radiology department. By using a radiation dose management platform, we extracted and
compared CT dose index (CTDIvol) and dose length product (DLP) both for each phase and for the entire CT exams. For objective
image quality evaluation, we calculated the signal-to-noise ratio (SNR) and the contrast-to-noise ratio (CNR) at the level of the liver
and of the aorta. Appropriate statistical analysis was performed. P-value < 0.05 was considered significant. The statistical analyses
were performed with R software, version 4.2.3.
Results or Findings: Sven thousand and ninety-eight CT examinations were included. CTDIvol was evaluated in 12804 phases; DLP
in 10713 phases and in 6714 examinations. The CTDIvol and DLP overall as well as in the native, arterial and portal venous phases
were significantly lower in the subspecialty radiology department. The objective image quality showed no significant difference in the
general and subspecialty radiology departments.
Conclusion: In a subspecialty radiology department, CT protocols are optimized compared to general radiology department leading
to lower dose to oncologic patients without significant objective image quality degradation.
Limitations: We did not evaluate the patients’ exposure through the effective dose; we cannot exclude that other factors may have
influenced the difference in radiation dose; we did not evaluate the effects of the reduction of radiation dose and the effects of the
choice of a different imaging.
Ethics committee - additional information: The present study was considered a quality care control study by our Ethical
Committee and did not fall under the Swiss law of the human research. As such, specific approval and informed consent were waived.
Comparison of iRECIST and RECIST 1.1 for response assessment of immune checkpoint inhibitor therapy (7 min)
Christian Nelles; Cologne / Germany
990
Author Block: C. Nelles, P. Bernard, N. Große Hokamp, M. Gräf, T. Persigehl, P. J. Bröckelmann, S. Lennartz; Cologne/DE
Purpose: To compare response assessment of patients undergoing immune checkpoint inhibitor (ICI) therapy between iRECIST and
RECIST 1.1 in a clinical cohort of patients with non-small cell lung cancer (NSCLC) or melanoma.
Methods or Background: 216 patients with NSCLC or melanoma who received PD-1 inhibitors (Nivolumab or Pembrolizumab) or
CTLA-4 inhibitor Ipilimumab between January 2015 and May 2020 and who underwent oncologic staging and follow-up CT of the chest
and abdomen were included in this retrospective analysis. Treatment response assessment in compliance with the RECIST 1.1 and
iRECIST guidelines was performed for all patients for the time period of ICI therapy. Response patterns as well as overall response
rate (ORR) and progression-free survival time (PFS) were compared between iRECIST and RECIST 1.1.
Results or Findings: Out of 110 (50.9%) patients with progressive disease (PD) according to RECIST 1.1, 34 did not reach
confirmation of progression (iCPD) in iRECIST and for 9 patients, iCPD was reached at a later point in time compared to RECIST 1.1,
with a resulting difference of mean PFS (225.3 ± 92.8 days for RECIST 1.1 vs. 270.7 ± 91.9 days for iRECIST). In iRECIST, unconfirmed
progressive disease (iUPD) transitioned to stable disease (iSD) in the subsequent follow-up in 17 patients and to partial response (iPR)
in 12 patients, leading to a difference in ORR (31.0% for RECIST 1.1 and 35.2% for iRECIST).
Conclusion: Our analysis suggests that iRECIST is more suitable than RECIST 1.1 for correctly capturing atypical response to ICI
therapy in patients with melanoma and NSCLC outside of clinical trials, resulting in differences in clinical outcome evaluation.
Limitations: This was a retrospective and a monocentre study design, we did not assess response patterns separately for melanoma
and NSCLC.
Ethics committee - additional information: After reviewing the study design, the institutional review board waived the need for
informed patient consent and approved this retrospective monocentre study.
991
RPS 2303 - Pre- and periprocedural imaging and new techniques

Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Florian Wolf; Vienna / Austria
Early myocardial extracellular volume at CT imaging: prognostic value in patients undergoing transcatheter aortic
valve implantation (7 min)
Maria Luisa Lampus; Sassari / Italy
Author Block: M. L. Lampus1, F. Rizzetto2, C. B. Monti2, D. Artioli2, P. Pedrotti2, F. Musca2, A. Vanzulli2; 1Sassari/IT, 2Milan/IT
Purpose: The objective of this study was to investigate the prognostic value of early myocardial extracellular volume (eECV) at CT
imaging in patients undergoing transcatheter aortic valve implantation (TAVI).
Methods or Background: Consecutive patients who underwent angio-CT examinations for TAVI procedure planning between
January 2021 and December 2022 were retrospectively identified. Pre-TAVI CT scans, demographics, and death records, when
present, were retrieved for each patient. Region-of-interests in the septal midventricular wall and left ventricular blood pool were
delineated on axial images from unenhanced scan and venous phase (acquired 70-90 seconds after contrast injection). Myocardial
eECV was calculated as the ratio between myocardium and blood pool differential attenuations, adjusted for patient hematocrit.
Myocardial relative enhancement (MRE), computed as the ratio between myocardium and blood pool attenuation at venous phase,
and septal wall thickness were also calculated. The correlation between eECV and MRE was assessed with Spearman's rho. Kaplan-
Meier analysis and hazard ratios (HR) were used to evaluate the association of eECV and MRE with overall survival to all-cause
mortality during a 30-month follow-up period.
Results or Findings: The study included 244 patients, with median (1st-3rd quartile) eECV, MRE, and septal wall thickness of 32%
(29-36%), 68% (64-73%), and 15 mm (13-17 mm), respectively. A strong positive correlation was observed between eECV and MRE
(rho=0.67; p<0.001). Patients with eECV≥34% had an increased risk of mortality (HR=2.7, 95% confidence interval: 1.3-5.4,
p=0.006), as well as those with MRE≥74% (HR=3.3, 95% confidence interval: 1.4-7.5, p=0.005). No significant difference in
demographics and septal wall thickness was observed between high- and low-risk patients.
Conclusion: Myocardial eECV and MRE obtained from venous phase CT images are associated with all-cause mortality in patients
undergoing TAVI.
Limitations: The main study limitation is the retrospective single-centre design.
Ethics committee - additional information: Institutional Review Board approved the retrospective data collection in an
anonymous, aggregated form.
Interrater variability of ML-based CT-FFR during TAVR-planning: influence of image quality and coronary artery
calcifications (7 min)
Robin F. Gohmann; Leipzig / Germany
992
Author Block: R. F. Gohmann, A. Schug, C. Lücke, P. Seitz, M. Gutberlet; Leipzig/DE
Purpose: The purpose of this study was to compare machine learning (ML)-based CT-derived fractional flow reserve (CT-FFR) in
patients before transcatheter aortic valve replacement (TAVR) by observers with differing training and to assess the influence of
image quality and coronary artery calcium score (CAC).
Methods or Background: Patients considered for TAVR have a high prevalence of coronary artery disease (CAD). Coronary
computed tomography angiography (cCTA) can effectively exclude CAD, but remains limited by its specificity. CT-FFR may mitigate
this limitation also in patients prior to TAVR. While a high reliability of CT-FFR is presumed, little is known about the reproducibility of
ML-based CT-FFR.
Results or Findings: CT-FFR was successfully performed on 214/272 examinations by both observers. The median difference of CT-
FFR was -0.05(-0.12-0.02) (p<0.001). Differences were smaller with high values and larger with low values. Categorization into CAD
was different in 37/214 examinations, resulting in a net recategorizations of Δ13 (13/214) examinations and a difference in accuracy
of Δ6.1%. Categorization into CAD was independent of quantitative image quality or CAC.
Conclusion: CT-FFR values measured by two observers with different experience were different, with smaller differences of high and
much larger differences of low values. Categorization into CAD was different in several patients, but only led to a small net difference
and ultimately had a moderate influence on diagnostic accuracy, independently of image quality or CAC.
Limitations: Retrospective study. Assessment of coronary artery disease (CAD) is recommended prior to transcatheter aortic valve
replacement (TAVR). Invasive coronary angiography (ICA) may be omitted, if significant CAD can be excluded on coronary computed
tomography angiography (cCTA). Despite its high sensitivity, cCTA is limited by relatively low specificity and positive predictive value,
particularly in high-risk patients. CT-derived fractional flow reserve (CT-FFR) is a promising technique with the potential to increase
the diagnostic accuracy and specificity of cCTA for ruling-out CAD during pre-TAVR evaluation. However, the significance of the
segmentation process and reader experience on the reliability of CT-FFR is not well investigated.
Funding for this study: No information was provided for this section by the submitter.
Ethics committee - additional information: The local ethics committee approved the study and written informed consent was
waived.
CT fluoroscopy-guided percutaneous biopsy of cardiac and pericardial masses (7 min)

Patrik Rogalla; Toronto / Canada
Author Block: P. Rogalla, M. Pourafkari, B. Karasfi, S. Kandel, C. O'Brien, R. Cusimano; Toronto, ON/CA
Purpose: Multiple routes exist for obtaining tissue samples from cardio-pericardial masses (CPM), including percutaneous
transvenous catheter biopsy techniques, minimally invasive methods such as mediastinoscopy and open surgical approaches. This
study aims to assess the feasibility, success rate, and complications of CT-guided biopsy of CPM.
Methods or Background: Institutional Research Ethics Board approval was obtained. Twenty-four biopsies of cardio-pericardial
masses and pericardial thickening were performed in 23 patients under CT-fluoroscopy guidance with or without EKG gating on a 320-
slice scanner (Canon Medical Systems) under conscious sedation. 3D mapping of the coronary arteries prior to the intervention,
mediastinal widening and capnothorax techniques were applied when appropriate. The clinical presentation, details of the procedure,
the success rate of obtaining diagnostic tissue, and potential complications and mitigation strategies, and in-room time were
recorded.
Results or Findings: Twenty-three patients (12 women, 11 men; mean age, 52.3 years) were included in this study. The CT
fluoroscopy-guided procedure was successful in obtaining diagnostic tissue samples in all patients. The target lesions ranged from 1
cm of pericardial thickening to a 15 cm intrapericardial mass. Histology revealed 12 malignant and 12 benign diagnoses. One patient
developed a small hemothorax, and another patient developed a small pneumomediastinum immediately post-procedure, both of
which were successfully treated conservatively. Post-procedure hemopericardium or arrhythmia were not observed in any of the
patients. One patient with encasing sarcomatoid mesothelioma passed away from cardiac arrest in the ICU two hours after the
procedure. The mean in-room time was 34.2 min.
Conclusion: CT-guided percutaneous biopsy of cardio-pericardial masses is a safe procedure that provides an accurate diagnosis and
may be considered when a histological diagnosis is crucial for clinical management decisions.
Limitations: Single-centre evaluation, no alternative sampling methods tested in comparison.
Ethics committee - additional information: Institutional Research Ethics Board approval was obtained for this study.
Diagnostic accuracy of detecting significant mitral valve regurgitation on non-contrast thoracic CT (7 min)
Benedicta Mutiara Suwita; Uxbridge / United Kingdom
993
Author Block: B. M. Suwita1, R. Khattar2, K. Vakalis1, W. Banya2, T. Mittal1, A. Baltabaeva1, S. Mirsadraee2; 1Harefield/UK, 2London/UK
Purpose: Previous studies suggested that mitral regurgitation (MR) is an under-diagnosed disease. Earlier detection results in
appropriate management and a better prognosis. Non-contrast thoracic CT for non-cardiovascular indications may be used for mitral
annulus measurement. This study aimed to assess CT markers of MR on non-contrast thoracic CT and to determine cut-off values for
detecting moderate/severe MR on this modality.
Methods or Background: This is a retrospective study of adult subjects who underwent non-contrast thoracic CT and
echocardiography between 2013 and 2023. Exclusion criteria were poor image quality, prior intervention on mitral/aortic valve, and
>12 months period between CT and echocardiography. Using double-oblique multiplanar projects, 4 parameters were measured on
the thoracic CT: mitral valve anterior-posterior (MVAP), mitral valve septal-lateral (MVSL), left ventricle anterior-posterior (LVAP) and
left ventricle apical-basal diameters. The left ventricle sphericity index (LVSI) was measured by dividing LVAP and LV apical-basal.
Subjects were divided into “normal”, “mild MR”, “moderate MR”, and “severe MR” categories based on echocardiography.
Results or Findings: 220 subjects were included (120 normal, 44 mild, 27 moderate, 29 severe). Kruskal-Wallis test and Bonferroni
post-hoc analysis showed no statistically significant difference between the normal and mild MR. However, there is a significant
difference between normal and moderate/severe MR groups in all CT parameters. Cut-off points to detect moderate/severe MR were:
MVAP 41.5 mm, MVSL 37.5 mm, LVAP 72.5 mm, LV apical-basal 87.5 mm, and LVSI 0.83. The accuracy (with 95% confidence interval)
for each cut-off point was: 84% (78–89%), 82% (76–87%), 78% (72–84%), 63% (55–70%) and 72% (66–79%).
Conclusion: Simple measurements on non-contrast thoracic CT can predict the presence of moderate/severe MR with good
diagnostic accuracy. These results are important for early detection of undiagnosed significant MR.
Limitations: The limitations of the study are no sub-group analysis for different MR causes.
Funding for this study: No funding was provided for this study
Impact of postprocedural CT evaluation of lead to valve angle on the development of cardiac implantable electrical
devices related tricuspid regurgitation (7 min)
Alessandra Caracciolo; Milan / Italy
Author Block: A. Caracciolo, C. Lisi, F. Catapano, L. Monti, M. Francone; Milan/IT

Purpose: Research shows that direct injury from CIED leads is related to the development of tricuspid regurgitation (TR), with a
prevalence of up to 40% of patients. This study aimed to explore the role of CT in identifying predictors of the development or
worsening of TR secondary to CIED.
Methods or Background: 374 patients who underwent CIED implantation between January 2014 and December 2021 and had at
least one chest CT after implantation and repeated transthoracic echocardiograms (TTE) were selected. The angle between the
tricuspid valve plane and the lead was measured in a two-chamber view CT scan. Worsening TR was defined as an increased TR
severity compared to baseline while worsening right ventricular (RV) function was defined as the new occurrence of RV dysfunction,
as indicated by a tricuspid annular plane excursion (TAPSE) <18 mm. Univariate and multivariate regression analyses were performed
to identify the predictors of development or TR worsening.
Results or Findings: After implantation, 169 individuals (45.18%) had new-onset or worsening TR and 60 (16.04%) had a worse RV
function. Regression analysis reported that statistically significant predictors of development or worsening of TR were a lead-valve
angle >90° (HR 2.03, CI 1.09-3.76; p 0.025) and worsening right ventricular function (HR 2.22, CI 1.26-3.91; p 0.006); other strongly
related variables, although without a statistical significance were moderate-to-severe mitral regurgitation (HR 0.74, CI 0.47 - 1.16; p
0.184), baseline RV dysfunction (HR 1.36; CI 0.83-2.24; p 0.2).
Conclusion: Our study demonstrates that lead positioning has a significant impact on patient outcomes, with a two-fold increased
risk of worsening TR when the lead-valve angle exceeds 90°. Postimplantation CT scans may help identify patients at risk of
developing TR due to suboptimal lead positioning.
Ethics committee - additional information: Not applicable because the study is retrospective.
Myocardial tissue characterisation with late contrast enhancement (LCE-CT) in patients presenting with acute
symptoms and mild troponin rise (7 min)
Davide Vignale; Milan / Italy
994
Author Block: D. Vignale1, A. Palmisano1, M. Gatti2, R. Faletti2, A. Esposito1; 1Milan/IT, 2Turin/IT
Purpose: This study aimed to investigate the added diagnostic value of myocardial tissue characterisation with Late Contrast
Enhancement (LCE) scan in patients with acute symptoms, slight troponin increases but no acute myocardial infarction diagnosis.
Methods or Background: A total of 144 patients (M: F=80:64; median age=67, IQR 50-77) with elevated troponin but no ECG or lab
signs of acute myocardial infarction were enrolled. Initial CT angiography (CTA) screened for coronary artery disease (CAD),
pulmonary embolism, and thoracic aortic conditions. Those with negative CTA results underwent a low kVp LCE scan 7 min after an
iodinated contrast injection to detect myocardial damage. A subset of patients with obstructive CAD underwent LCE too. Diagnoses
were mostly confirmed through gold-standard methods.
Results or Findings: Median troponin was 69 ng/L (IQR 39-200; normal <14 ng/L). Out of 144 patients, 64 (44%) had CAD, two (1%)
aortic syndrome, and eight (6%) pulmonary embolism. LCE was performed in the remaining 70 (49%) with negative CTA, diagnosing
myocarditis in 35/70 (50%), Takotsubo in 8/70 (11%), MINOCA in 4/70 (6%), dilated cardiomyopathy in 4/70 (6%), and amyloidosis in
3/70 (4%). For 16/70 (23%), LCE was normal. Among the 27 with CAD also imaged with LCE, 12 (44%) had no LCE, four (15%) had
subendocardial LCE, and 11 (41%) had transmural LCE, with 3/11 (27%) showing microvascular obstruction.
Conclusion: In patients with acute symptoms and elevated troponin but negative CTA, supplementary LCE improved diagnostic
accuracy from 51% to 89% (p<0.001), mainly identifying myocarditis as the aetiology. In patients with CAD, LCE revealed different
extents of myocardial damage, offering valuable insights into the severity of acute coronary syndromes.
Limitations: This study had a relatively small sample size and not all patients underwent CMR. There was also a lack of follow-up
data on LCE capabilities of risk stratification.
Ethics committee - additional information: This study was approved by protocol CT-MyoC.
Feasibility of T2 mapping for the detection of myocardial injury in hypertrophic cardiomyopathy (7 min)
Jiaxin Wang; Beijing / China
Author Block: S. Yang, S. Zhao, J. Wang; Beijing/CN

Purpose: This study aimed to explore the association between T2 value and myocardial injury in patients with hypertrophic
cardiomyopathy (HCM).
Methods or Background: All patients underwent laboratory testing and CMR. Hs-cTnI was obtained as a marker of myocardial injury
(>0.016 ng/mL). Global T2 values were quantified for three LV short-axis slices. The maximum T2 value and ConSept T2 were
measured from the maximal value of 16 segments and the middle septum, respectively.
Results or Findings: In the HCM group, elevated hs-cTn I was found in 25 (20 males, 50±13 years) and normal hs-cTn I in 25 (17
males, 49±14 years). The hs-cTnI elevated group had higher T2 global, ConSept T2, T2 max, T1 global, ConSept T1 values, and LGE
extent (P<0.05) than the normal hs-cTnI group. The ConSept T2 value (r=0.52, P<0.001) and LGE extent (r=0.52, P<0.001) were all
moderately correlated with hs-cTnI. Among all parametric parameters, the ConSept T2 value showed the best performance in
identifying myocardial injury (AUC=0.83). In multivariate logistic regression analyses, after adjusting LV ejection fraction, LV end-
diastolic volume index, maximal ventricular wall thickness, LV myocardial mass index, LGE extent, T1 global, and ConSept T1 values,
respectively, ConSept T2 value was still significantly associated with elevated hs-cTnI (P<0.01).
Conclusion: We confirmed the feasibility of T2 mapping for detecting myocardial injury in HCM. T2 value was associated with an
elevated hs-cTnI, which provided in-vivo evidence by CMR for the ongoing myocardial injury in HCM.
Limitations: The study was conducted at a tertiary referral centre with a small sample size. Since the mapping values depend on the
sequence and vendor, the results observed in our study might not be generalised to the other centres.
Funding for this study: This study was supported by grants from the National Key Research and Development Program of China
(No. 2021YFF0501404, No. 2021YFF0501400) and the key projects of the National Natural Science Foundation of China (No.
81930044).
Ethics committee - additional information: The study complied with the Declaration of Helsinki and was approved by the
institutional review boards of Fuwai Hospital.
Fractal analysis on CMR: trabecular complexity as a new marker to predict outcomes in patients with dilated
cardiomyopathy (7 min)
Xiaorui Xiang; Beijing / China
995
Author Block: X. Xiang, S. Yu, K. Yang, C. Cui, X. Chen, M. Lu, S. Zhao; Beijing/CN
Purpose: Trabecular complexity is a unique biometric marker like a fingerprint, but its prognostic impact in dilated cardiomyopathy
(DCM) remains unclear. This study investigated the prognostic value of trabecular complexity by fractal analysis in patients with DCM.
Methods or Background: Consecutive 276 patients with DCM were enrolled in this study. Comprehensive clinical evaluation and
cardiovascular magnetic resonance (CMR) imaging investigation were obtained. Trabecular complexity was quantified with fractal
analysis of cine images to estimate the global, basal, and apical fractal dimensions (FD). All patients were followed up for major
adverse cardiac events (MACE) of all-cause mortality, aborted sudden cardiac death, and heart transplantation.
Results or Findings: Over a 5.37-year median follow-up, 103 (37.32%) patients experienced MACE. All left ventricular FD
parameters were higher in patients with events than those without events (all P<0.05). Max Basal FD emerged as the strongest MACE
prognosticator among FD parameters (AUC: 0.84 [95% CI, 0.78–0.88]), and the optimal cutoff value was 1.27. Furthermore, Cox
hazards analysis revealed that Max Basal FD was independently associated with MACE (HR: 1.07 per %, p=0.002) after adjustment for
clinical and imaging risk factors including NT-proBNP, LVEF, and LGE presence. By Kaplan-Meier analysis, the risk of MACE increased
significantly with increased Global FD, Mean Basal FD, and Max Basal FD (all log-rank p<0.001).
Conclusion: LV max basal FD was an independent predictor of adverse outcomes, and fractal analysis may contribute to improving
the risk stratification for patients with DCM.
Limitations: This was a single-centre study with no genetic characterization of the subjects. Despite the limitations, it is the first
study to apply Fractal Analysis to predict outcomes in DCM, demonstrating that myocardial trabecular complexity as a new marker
has the potential to improve DCM risk stratification algorithms.
Funding for this study: This study was funded by the National Key Research and Development Program of China (No.
2021YFF0501400) and the Key Project of the National Natural Science Foundation of China (No. 81930044).
Ethics committee - additional information: This study was approved by the hospital’s Institutional Review Board (IRB).
Deep learning reconstruction for transcatheter aortic valve implantation planning CT (7 min)
Yu Zhang; Chengdu / China
Author Block: Y. Zhang; Chengdu/CN

Purpose: This study aimed to assess the image quality, dose reduction potential and diagnostic performance of low tube voltage with
high-strength deep learning image reconstruction (DLIR-H) for transcatheter aortic valve implantation TAVI-planning CT in comparison
with conventional imaging protocols.
Methods or Background: We prospectively enrolled 128 patients who were referred to TAVI-planning CT. Patients were randomly
divided into two groups: the DLIR-H group (n=64) with a contrast injection rate of 28mgI/kg/s, and the conventional group (n=64) with
a contrast injection rate of 40mgI/kg/s. Radiation dose, contrast volume, contrast injection rate, and image quality were compared
between the two groups. The diagnostic performance of TAVI-planning CT for coronary artery stenosis was calculated using invasive
coronary angiography as the gold standard.
Results or Findings: There was a significant reduction in radiation dose (4.94±0.39 mSv vs. 7.93±1.20 mSv), contrast volume
(45.28±5.38 mL vs. 63.26±9.88 mL) and contrast injection rate (3.1±0.31 mL/s vs. 4.9±0.2 mL/s) in DLIR-H group compared to the
conventional group (all p<0.010). For diagnostic performance, TAVI-planning CT in the DLIR-H group provided 100% sensitivity, 92.1%
specificity, 100% negative predictive value (NPV), and 84.2% positive predictive value (PPV) on a per-patient basis for the detection of
>50% stenosis. In the conventional group, for the detection of >50% stenosis, TAVI-planning CT provided 94.7% sensitivity, 95.3%
specificity, 97.6% NPV, and 90.0% PPV on a per-patient basis.
Conclusion: The combination of low tube voltage and DLIR-H in TAVI-planning CT provides improved image quality while significantly
reducing radiation and contrast volume with no influence on the diagnostic performance of coronary artery stenosis in comparison
with the conventional protocol.
Limitations: We only compared a single DLIR level (high level) with a single ASIR-V blending factor (50%) and other levels of DLIR
and ASIR-V were not compared.
Ethics committee - additional information: This prospective study was approved by the ethics committee of our hospital and
written informed consent was obtained from all participants before final inclusion.
A new whole-heart motion correction algorithm enables diagnostic CT of the aortic valve and coronary arteries in the
systolic phase for transcatheter aortic valve implantation candidates (7 min)
Yu Zhang; Chengdu / China
996
Author Block: Y. Zhang; Chengdu/CN
Purpose: This study aimed to investigate the ability of the new generation snapshot freeze (NGSSF) algorithm to improve the
diagnostic image quality of both the aortic valve and coronary arteries for transcatheter aortic valve implantation (TAVI) candidates in
TAVI-planning CT.
Methods or Background: 64 TAVI candidates who underwent TAVI-planning CT were enrolled. Scans from coronary CT angiography
were reconstructed at 20%, 30%, 40%, and 75%; R-R cardiac phases with NGSSF and standard (STD) algorithm. In each phase, the
following parameters were compared: aortic valve measurements and their reproducibility, image quality of aortic valve and coronary
arteries. The diagnostic accuracies of TAVI-planning CT for coronary artery stenosis in 30% R-R phase with NGSSF and STD algorithms
were calculated in 47 out of 64 patients with invasive coronary angiography as reference standard.
Results or Findings: For subjective image quality evaluation, the excellent rate for the aortic valve improved from 25.0% to 93.8%,
and the interpretable rate for coronary arteries increased from 20.3% to 95.3% in the 30% phase images with NGSSF compared with
images with STD. For the detection of >50% coronary artery stenosis, the 30% phase images with NGSSF provided a sensitivity of
90%, specificity of 81.48%, negative predictive value of 91.7%, and positive predictive value of 78.3% on a per-patient basis, while
images with STD, had a corresponding result of 95.0%, 33.33%, 90.0%, and 51.4%, respectively.
Conclusion: NGSSF significantly improves image quality for both the aortic valve and coronary arteries compared with STD for TAVI
patients of all heart rates. NGSSF enables the accurate measurement of the aortic valve and satisfactory diagnostic performance for
coronary artery stenosis in the same systolic phase for TAVI planning.
Limitations: The cardiac phases reconstructed in our study were 20%-75%, while 0%, 10%, 50%, 60%, 70%, and 90% were not
reconstructed.
Ethics committee - additional information: The Institutional Review Board of our hospital approved our study. Written informed
consent was obtained from all patients.
Computed tomography angiography reconstructed with MBIR for TAVR planning as a comprehensive tool for the
assessment of coronary arteries (7 min)
Cammillo Roberto Giovanni Leopoldo Oreste Massimiliano Talei Franzesi; Milan / Italy
Author Block: C. R. G. L. O. M. Talei Franzesi, C. Maino, D. G. Gandola, M. Ragusi, T. P. Giandola, P. N. Franco, R. Corso, D. Ippolito;
Milan/IT
Purpose: This study aimed to evaluate the diagnostic performance of a model-based iterative algorithm (MBIR) in the assessment of
coronary arteries in an angiographic-CT (CTA) study during TAVR planning.
Methods or Background: A total of 82 patients who underwent CTA for TAVR planning were included in this study. All exams were
obtained using a 256-slice MDCT scanner with low-dose and low-contrast media volume CTA protocol (80 kV, 50 mL) with ECG-gated
technique and reconstructed with MBIR algorithm. All coronary arteries were divided into 3 segments and evaluated in terms of image
quality and contrast enhancement. Mean attenuation values (HU) of coronary arteries and thoracic aorta were calculated by drawing
a manual region of interest (ROI) in the lumen of vessels.
Results or Findings: The mean attenuation value in the thoracic aorta was 478 HU and in the abdominal aorta was 447 HU. The
proximal segment of the left anterior descending artery was identified and analysed in all (100%) cases, while the medium segment
was identified in 89% and the distal tract only in 65%.
The proximal segment of the left circumflex artery was recognisable in 89%, the medium segment in 70%, and only in 29% it was
possible to assess the distal segment. In 95% of CT scans, we were able to evaluate the proximal segment of the right coronary
artery, in 79% of the medium and 52% of the distal segment.
Conclusion: Low-dose and low-contrast media volume CTA performed for TAVR planning offers important insight information on
coronary status when combined with the IMR reconstruction algorithm.
Limitations: This was a retrospective and a single-centre study.
997
RPS 2305 - Cutting edge advances in AI for radiological imaging
Categories: Artificial Intelligence & Machine Learning, General Radiology, Imaging Methods, Research
Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:
Stefania Volpe; Milan / Italy
Automatic detection of bone fragility in radiographic images using deep-learning with multicentre cohort datasets (7
min)
Guillaume Gatineau; Lausanne / Switzerland
Author Block: G. Gatineau1, G. Nguyen2, M. De Gruttola2, K. Hind2, M. Kuzma3, J. Payer3, G. Guglielmi4, A. Fahrleitner-Pammer5, D.
Hans1; 1Lausanne/CH, 2Geneva/CH, 3Bratislava/SK, 4Foggia/IT, 5Graz/AT
Purpose: This study aimed to assess the accuracy and efficacy of an novel AI-driven radiographic processing tool designed to
opportunistically identify individuals predisposed to very high bone fragility risk, addressing a prevailing clinical challenge in the field.
Methods or Background: From four multinational cohorts, 4,764 paired lumbar-spine X-ray DICOM and DXA scans (GE and Hologic
systems) were acquired within 6 months. A total of 3,369 cases from three cohorts were allocated for training and validation of a new
AI-bone fragility detection tool (Medimaps Group, Switzerland). Three hundred cases were designated as internal test set. Two
hundred and seventy-one cases from the fourth cohort acted as an external test set. Very high fracture risk was defined using DXA
parameters as the ground truth: BMD T-score≤-2.5 and a trabecular bone score (TBS)<1.23.
Results or Findings: The mean age and BMI of the sample (5.2% male) were 66.1±10.8 y and 26.4±5.0 kg/m2 respectively. Using
the combination of DXA-derived BMD and TBS, 17.5% were identified at very high fracture risk. Uncertainties were obtained with a
95% confidence interval (CI) using binomial distribution approximations. The accuracy of the AI tool for the internal test set, was 0.85
(95% CI: 0.76-0.94), specificity 0.91 (0.8-0.99), and sensitivity 0.69 (0.53-0.84). For external validation, the accuracy, specificity, and
sensitivity were 0.80 (0.69-0.87), 0.88 (0.77-0.99), and 0.62 (0.47-0.77) respectively.
Conclusion: This AI-enhanced radiographic tool exhibits potential in accurately detecting individuals at very high risk of bone
fragility. Its robust specificity underscores its capacity to reduce false-positive rates, emphasising its clinical utility for efficient patient
screening.
Limitations: While this study demonstrates promise, further development and validation will be beneficial, using larger and more
diverse samples.
Funding for this study: This study was funded by the Fond National Suisse 32473B_156978 and 320030_188886.
Ethics committee - additional information: This study was ethically approved by the WMA declaration of Helsinki. Ethical
Principles For Medical Research Involving Human Subjects
Spectral CT fingerprinting: AI-based clustering of tissue signatures: proof of concept (7 min)

Nils Große Hokamp; Cologne / Germany
998
Author Block: N. Große Hokamp, M. M. Thakur, D. Maintz, M. Schöneck, L. Caldeira; Cologne/DE
Purpose: Spectral computed tomography (CT) allows for Material Decomposition by assessing attenuation caused by photoelectric
effect and compton scattering seperately. Followed by dedicated reconstruction algorithms, dedicated spectral results can be
generated that highlight different material characteristics. We claim that combining these quantitative characteristics will result in a
characteristic CT-signature or fingerprint of tissues and that this clustering can be automated using artificial intelligence. This study
served as proof-of-concept of this approach in a phantom setup.
Methods or Background: A 3D-printed phantom filled with different solutions containing iodine (2 different concentrations),
gadolinium, iron oxide and water was scanned using a dual layer spectral CT system. Conventional images, iodine maps and low
energy virtual monoenergetic images were reconstructed. An unsupervised, high dimensional, k-means clustering algorithm was
developed and used to automatically provide clusterings. These clusters were then forward-projected into the image and visually
checked for agreement.
Results or Findings: Aforementioned high-dimensional, AI-based clustering allowed for cluster formation (mean RMSE = 0.68, range
[0.06-1.63]). Forward projection revealed, that clusters match very closely to actual phantom compartments as indicated by DICE
coefficients (all >0.8).
Conclusion: CT Fingerprinting, i.e. the AI-based derivation of a tissue signature ist feasible from a technology standpoint. Future
research needs to focus on elevating the dimension of the clustering algorithm and on clinical translation.
Limitations: The study has been carried out in a phantom only, as it should serve as proof-of-concept. Only 3 dimensions have been
introduced in the clustering approach to date.
Funding for this study: This study was funded by the Advanced Clinician Scientist Program (AdCCSP), University of Cologne.
Ethics committee - additional information: This was a phantom study and hence, no ethical approval was sought.
Pre-operative prediction of axillary lymph node status using ML applied to breast US: a multicentre study (7 min)
Martina Caruso; Naples / Italy
Author Block: M. Caruso, A. De Giorgio, R. Cuocolo, L. R. La Rocca, M. Ferrante, A. Stanzione, S. Maurea, V. Romeo, A. Brunetti;
Naples/IT
Purpose: To assess whether a machine-learning (ML) algorithm could empower the ability of US to preoperatively define axillary
lymph node (ALN) status in breast cancer (BC) using a multicentric dataset.
Methods or Background: Patients with at least one histologically proven BC lesion, who underwent preoperative breast US, were
retrospectively selected in three different Institutions. BC lesions were segmented on US images by three different operators and
radiomics features (first, second higher order) were extracted. Multi-step feature selection was performed using Intraclass Correlation
Coefficient (ICC) analysis and principal component analysis (PCA). Thereafter, a Random Forest (RF) ML classifier was applied to the
dataset to predict the ALN status (positive/negative for metastasis) and its performance assessed through the Matthews Correlation
Coefficient (MCC).
Results or Findings: A total of 293 BCs (ALN negative: 176; ALN positive: 117) were included in the study. Three datasets were
identified as follows: 1) training set, composed of 235 BCs (ALN-: 141; ALN+: 94); 2) validation set including 30 BCs (ALN-: 17; ALN+:
13); and 3) test set made of 30 BCs (ALN-: 18; ALN+: 12). 549 radiomics features were extracted from US images; of these, 280 were
discarded according to ICC analysis, with a total of 5 features finally selected by PCA. RF classifier showed a MCC of 0.97, 0.11 and
0.08 in the training, validation, and test set, respectively.
Conclusion: ML applied to a multicentric dataset showed promises in the preoperative assessment of ALN status in BC. However,
further efforts are necessary to improve the generalisability of the model when applied to external datasets.
Limitations: Limited sample size
Ethics committee - additional information: The study was approved by local ethics committee.
Prediction of patient survival with residual convolutional neural network (ResNet) in unresectable synchronous liver-
only metastatic colorectal cancer treated with bevacizumab-based chemotherapy (7 min)
Sung-Hua Chiu; Taipei / Taiwan, Chinese Taipei
999
Author Block: S-H. Chiu, C-C. Wu, W-C. Chang, P-Y. Chang; Taipei/TW
Purpose: To verify the prediction of survival with residual convolutional neural network (ResNet)-determined morphological response
(ResNet-MR) in patients with unresectable synchronous liver-only metastatic colorectal cancer (mCRC) treated with bevacizumab-
based chemotherapy (BBC).
Methods or Background: A retrospective review of liver-only mCRC patients treated with BBC from December 2011 to April 2021
was performed. Patients with metachronous liver metastases or receiving locoregional treatment before initiation of BBC were
excluded. Downstaging to curative treatment and overall survival (OS) were recorded. Two abdominal radiologists evaluated CT
images based on the morphological criteria and divided images into group 1, 2, and 3. These divided images established the
radiologists-determined morphological response (RD-MR), which classified patients into responders and non-responders based on the
morphological change 3 months after initiation of BBC. Then, group 1 and 3 images divided by radiologists were inputted into ResNet
as training dataset. The trained ResNet redivided group 2 images into group 1 and 3. ResNet-MR was determined by these redivided
images and initial group 1 and 3 images determined by radiologists.
Results or Findings: Eighty-four patients were enrolled (53 male with median age 60.0 years old). The follow-up time ranged from
10-86 months. There were 407 group 1 and 3 images imputed into ResNet as training dataset. Both RD-MR and ResNet-MR correlated
with OS (p-value = 0.0167 and 0.0225, respectively). RD-MR classified 28 patients (33.3%) as responders, and ResNet-MR classified
additional 16 patients (19.0%) as responders; these 16 patients showed longer OS than the rest of non-responders in RD-MR (27.49
versus 21.20 months, p-value=0.043), and had higher percentage to reach downstaging (37.5% versus 17.5%, p-value=0.1610).
Conclusion: ResNet showed it ability to predict therapeutic effect of BBC in mCRC patients, which will optimise individualised cancer
treatment.
Limitations: A single-centre retrospective study .
Funding for this study: Funding was obtained for this study with the funding number: Nsc 109-2314-B-016-012.
Ethics committee - additional information: The study was approved with the approval protocol number: TSGHIRB No.
2-108-05-153, under the protocol title "Development of artificial intelligence approaches for Detection and Characterization of Hepatic
Lesion and Evaluation of HCC treatment".
Predicting histological response in pancreatic adenocarcinoma treated with neoadjuvant chemotherapy: conventional
clinicoradiological variables and added value of radiomics (7 min)
Junaid Mushtaq; Milan / Italy
Author Block: J. Mushtaq, D. Palumbo, F. Prato, M. Mori, S. Crippa, M. Falconi, C. Fiorino, F. De Cobelli; Milan/IT
Purpose: Pancreatic cancer prognosis remains abysmal due to late-stage diagnosis and limited treatment options. Neoadjuvant
chemotherapy (NAT) is increasingly used to improve outcomes. However, it complicates radiological assessment, making it difficult to
accurately determine tumor status. This study aims to explore radiomics as a non-invasive tool for predicting important prognostic
factors and compares its performance with traditional radiological assessments.
Methods or Background: This retrospective single-centre study included patients with pancreatic adenocarcinoma who underwent
NAT and pancreaticoduodenectomy between January 2015 and December 2021. Clinical, radiological and pathological data were
collected, and endpoints identified: disease recurrence, N2, Tumor Regression Grade (TRG). Radiomic features were extracted from
pre-NAT and post-NAT CT images. Machine learning approaches, including bootstrapping, were used to develop predictive models.
Two models were built: a purely radiomic model and a combined model including clinical and radiological data. The population was
divided in a training and a validation cohorts. Balanced groups ensured robust analysis following established guidelines.
Results or Findings: The study included 156 patients (training n=103, validation n=53). Radiological and clinical variables, except
for delta Ca19.9 levels, failed to predict the endpoints. Radiomics, using delta values to capture changes in the tumor
microenvironment, showed promise. Delta radiomic models were successfully validated for the N2 and TRG endpoints and performed
moderately well with AUCs of 0.749 (p=0.031) and 0.710 (p=0.046) respectively. Excellent negative predictive values, 82% and 79%
respectively for N2 and TRG, indicate the model's ability to identify low-risk patients. A combined model with delta CA19.9 did not
significantly improve performance.
Conclusion: Radiomics may hold potential for improved patient selection in pancreatic cancer treatment.
Limitations: The study is limited by its retrospective nature and single-center design. External validation is necessary for confirming
the results.
Ethics committee - additional information: This retrospective study was conducted at the San Raffaele Scientific Institute (Milan,
Italy) and was carried out within the framework of an approved Ethics Committee study (28/INT/2015).
Monitoring over time of pathological complete response to neoadjuvant chemotherapy in breast cancer patients
through an ensemble vision transformers-based model (7 min)
Maria Colomba Comes; Bari / Italy
1000
Author Block: M. C. Comes, S. Bove, A. Fanizzi, R. Massafra; Bari/IT
Purpose: Morphological and vascular peculiarities of breast cancer can change during neoadjuvant chemotherapy (NAC). Dynamic
contrast-enhanced magnetic resonance imaging (DCE-MRI) acquired pre- and mid-treatment quantitatively capture information about
tumour heterogeneity as potential earlier indicators of pathological complete response (pCR) to NAC in breast cancer. This study
aimed to develop an ensemble deep learning-based model, exploiting a Vision Transformer (ViT) architecture, which merges features
automatically extracted from five segmented slices of both pre- and mid-treatment exams containing the maximum tumour area, to
predict and monitor pCR to NAC.
Methods or Background: Imaging data analysed in this study referred to a cohort of 86 breast cancer patients, randomly split into
training and test cohorts at a ratio of 8:2, who underwent NAC and for which information regarding the pCR achievement was
available (37.2% of patients achieved pCR). As far as we know, our research is the first proposal using ViTs on DCE-MRI exams to
monitor pCR over time during NAC.
Results or Findings: The performances of the proposed model were assessed using standard evaluation metrics and promising
results were achieved: AUC value of 91.4%, accuracy value of 82.4%, a specificity value of 80.0%, a sensitivity value of 85.7%,
precision value of 75.0%, F-score value of 80.0%, G-mean value of 82.8%.
Conclusion: Finally, the heterogeneity changes in DCE-MRI at pre- and mid-treatment could affect the accuracy of pCR prediction to
NAC.
Limitations: The study needs to be validated in a larger cohort.
Funding for this study: The funding for this study was obtained by the Ministery of Health, Italy.
Ethics committee - additional information: The study was approved by the Ethic Commitee of Istituto Tumori ‘Giovanni Paolo II’,
Bari, Italy.
Improving the ability of junior radiologists in distinguishing the morphology of carotid plaque fibrous cap using
artificial intelligence (7 min)
Junni Shi; Shanghai / China
Author Block: J. Shi, Y. Huang, M. Chen; Shanghai/CN

Purpose: We proposed a deep learning model which can estimate the morphology of carotid plaque fibrous cap. The aim of this
study was to assess the diagnostic performance of the AI model and to evaluate the AI model in clinical application.
Methods or Background: Two hundred and forty-five carotid plaques from 245 patients with complete clinical information and
imaging information (ultrasound and CEUS) finally recruited in our study. We develop a target boundary and perfusion feature guided
video analysis network(BP-Net) and build an AI diagnostic model of plaque fibrous cap morphology. To evaluate the AI model in
clinical application, we collected additional 62 patients information for morphology analysis of carotid plaque fibrous cap by a junior
radiologist and the new-built AI diagnostic model.
Results or Findings: Two hundred and forty-five patients (70 women and 175 men) were evaluated. The AI model achieves an
average of 92.35%, 0.935, 90.58%, 93.60%, 91.16%, 93.36% and 90.79% for accuracy, AUC, sensitivity, specificity, positive predictive
value, negative predictive value and F1-score in the integrity of carotid plaque fibrous cap. The sensitivity, specificity, accuracy and
AUC of carotid plaque fibrous cap diagnosis in 62 plaques of junior radiologist achieved 71.43%, 60.97%, 64.51%, 0.662. With the
assist of AI diagnostic model, the sensitivity, specificity, accuracy and AUC reached 90.47%, 90.24%, 90.32%, 0.904.
Conclusion: The AI diagnostic model we proposed showed excellent performance in distinguishing the morphology of carotid plaque
fibrous cap. The AI model also could assist junior radiologists in evaluating the fibrous cap appearance which was an important
characteristic of vulnerable plaque .
Limitations: Firstly, in our study the ground truth in deep learning were from a panel of medical experts instead of histopathologic
identification. Second, our network was limited by certain tracking conditions.
Ethics committee - additional information: The study was approved by the ethics committee of our institution and informed
consent was obtained from all participants.
Retrospective evaluation of interval breast cancer: can the number of interval carcinomas be reduced utilising AI
diagnostic software? (7 min)
Jonas Subelack; St. Gallen / Switzerland
1001
Author Block: J. Subelack1, A. Geissler1, J. Vogel1, M. Blum1, A. Eichenberger1, R. Morant1, A. Gräwingholt2, D. Kuklinski1; 1St.
Gallen/CH, 2Paderborn/DE
Purpose: We investigate whether an artificial intelligence (AI) powered mammography screening software can support radiologists
and screening programs to reduce interval carcinomas (ICs).
Methods or Background: Combining data from the cancer registry of eastern Switzerland and the “donna” screening program, we
include data from 151,245 screening mammograms between 2010 and 2019. Hereby, 264 ICs were identified when a carcinoma was
detected opportunistically within 24 months after screening. Mammograms of the 264 ICs which in the screening round prior to
detection were considered as normal and 90 randomly selected true normals were reviewed retrospectively by three independent
radiologists and an AI-diagnostic software (ProFound-AI, iCAD). The software calculates three measures: (1) lesion-score: probability of
a marked lesion in a mammogram to be a cancer; (2) case-score: probability whether the set of mammograms of one woman contains
a cancer; (3) risk-category: woman’s risk to be detected with breast cancer within two years. We evaluate which measure
(combination) finds best signs of later detected ICs in the prior mammograms that were read as normal and also compare it to the
review of the three radiologists. We further estimate the accuracy of the software depending on thresholds.
Results or Findings: We expect that both retrospective assessments have an improved detection rate of ICs compared to the initial
screening. Detection probability of ICs and the number of false-positives is highly dependent on the thresholds set for the measure(s).
Using predefined thresholds, we expect that the software is superior to the retrospective assessments of radiologists in detecting ICs.
Conclusion: If the software detects significantly more cancer signs than the radiologists, it needs to be discussed how best to
integrate the software into the mammography screening process to reduce ICs.
Limitations: Only ICs are analysed in depth, limiting the scope of this study.
Funding for this study: The cancer league of eastern Switzerland funds this cooperative research project.
Ethics committee - additional information: A declaration of non-responsibility is at hand.
Exploring the potential of ChatGPT and a context-aware ChatGPT in gastrointestinal radiology: a differential diagnostic
accuracy assessment (7 min)
Stephan Rau; Freiburg / Germany
Author Block: S. Rau, A. Rau, J. Nattenmüller, A. Fink, F. Bamberg, M. Russe; Freiburg im Breisgau/DE
Purpose: The growing volume of imaging studies necessitates tools to assist radiologists in delivering accurate and effective
diagnoses. We aimed to investigate the potential of ChatGPT by OpenAI and a context-aware chatbot based on ChatGPT in providing
differential diagnoses.
Methods or Background: Utilizing the LlamaIndex framework, which integrates datasets into large language models, ChatGPT 4
was enhanced using the 96 documents from the Radiographics Top 10 Reading list on gastrointestinal imaging, creating a
gastrointestinal imaging-aware chatbot (giaGPT). To assess the differential diagnostic capability, a set of fifty case-files on abdominal
pathologies was created, comprising of radiological findings in fluoroscopy, MRI and CT. We compared the giaGPT to the generic
ChatGPT 4 (gGPT) in terms of offering the top three differential diagnoses using interpretations from senior level radiologists as
ground truth. Additionally, the trustworthiness of the giaGPT was evaluated by investigating the utilized source document as provided
by the embedded context-retrieval mechanism.
Results or Findings: Within the evaluated dataset, the giaGPT demonstrated a high capability in identifying the most appropriate
differential diagnosis in a vast majority of cases with a sensitivity of 78%, significantly surpassing the gGPT with 54%. Notably, giaGPT
offered the primary differential in the top 3 differential diagnoses in 90% of the cases, gGPT in only 74%.
Conclusion: Context-aware ChatGPT-based algorithms may provide a promising tool for supporting radiologists in the task of
differential diagnostics. It offers accurate differentials and direct access to the employed source documents, providing insight into the
decision-making process, providing trustworthy and evidence-based clinical decision-support.
Limitations: Although the findings are promising, they call for a more extensive evaluation using datasets from clinical routine
covering a more diverse spectrum of pathologies. Additionally, future research should focus on the impacts of diagnostic confidence
and time efficiencies.
Funding for this study: This study was financially supported by internal funds.
Ethics committee - additional information: No ethics committee decision was needed in the study because no patient data were
used.
CT effective dose in intensive care unit patients: comparison between deep learning image reconstruction, filtered
back projection and iterative algorithms (7 min)
Elena Agostini; Padua / Italy
1002
Author Block: E. Agostini, E. K. Lanza De Cristoforis, C. Zanon, E. Quaia; Padua/IT
Purpose: The aim of this study was to assess the effective dose reduction and image quality improvement provided by deep learning
image reconstruction (DLIR) in comparison to filtered back projection (FBP) and iterative reconstruction (IR) algorithms.
Methods or Background: Eighty-three critical care patients who underwent CT imaging of the chest, abdomen and trunk (chest +
abdomen) within a period of 30 days using both DLIR (TrueFidelity) and FBP or IR hybrid (AIDR3D) and model-based IR algorithm
(ADMIRE) were included. All examinations were performed using automatic exposure control (AEC) which modulates the tube current
and hence radiation exposure according to the algorithm applied. Radiation dose was assessed using CT dose index volume (CTDI
volume), dose-length product (DLP) and Effective Dose. For the quantification of image quality, noise and signal to noise ratio (SNR)
were used. All parameters were compared across the different reconstruction methods for each patient using both parametric (t test)
and non-parametric (Wilcoxon) tests. In cases of contrast-enhanced CT (CECT), all parameters were retrieved for every acquisition
phase (direct, arterial, venous or delayed) as well as for their total value as stated in the patient protocol.
Results or Findings: DLIR vs FBP improved both CTDIvol (9.56 ± 5.86 vs 24.67 ± 61.01), DLP (1085.33 ± 626.30 vs 1350.62 ±
1191.68), and effective dose (16.13 ± 9.55 vs 20.19 ± 17.9), while DLIR vs FBP and vs IR improved both image noise (8.45 ± 3.24 vs
28.85 ± 32.77 vs 14.85 ± 2.73 HU) and SNR (3.99 ± 1.23 HU vs 11.53 ± 9.28 vs 4.84 ± 2.74 HU).
Conclusion: DLIR provides benefits in terms of effective dose and image quality over the traditional FBP. It also outperforms IR
methods for image quality, but not for effective dose.
Ethics committee - additional information: The study was approved by the institutional review board.
Insights from a commercial AI tool’s performance in detecting common acute and chronic chest radiograph
abnormalities in a multi-ethnic Singapore dataset using AimSG’s automated auditing tools (7 min)
Timothy Shao Ern Tan ; Singapore / Singapore
Author Block: T. S. E. Tan , G. Han Leong, J. Zou, J. Y. C. J. Liew; Singapore/SG

Purpose: To clinically validate the performance of a machine learning-based algorithm tool in identifying abnormal findings on chest
radiographs (CXRs).
Methods or Background: Two consultant radiologists with between 5 and 15 years of experience read 300 adult CXRs performed in
2021 at a high-volume tertiary Singapore acute hospital to establish ground truth. The Lunit INSIGHT CXR (South Korea) artificial
intelligence (AI) algorithm was integrated into the hospital RISPACS on a national vendor-neutral AI imaging platform (AimSG,
Singapore), which contains an integrated data stack and performance monitoring data tools (Carpl.AI, New Delhi). Audit of the AI
algorithm's performance on the CXRs was compared with the radiologists' analyses on AimSG. The area under the receiver operating
characteristic curve (AUROC) was calculated for ten specific abnormalities.
Results or Findings: Three hundred adults (median age 75.7, range +/- 15.6 ; 182 males, 61%) comprising 183 (61% Chinese), 96
(32% Malays), 16 (5% Indians), and 5 (2%) other ethnicities had CXRs performed for acute respiratory and thoracic symptoms. A total
of 281 (93.7%) of 300 CXRs demonstrated at least one acute or chronic abnormality. The algorithm achieved an AUROC of between
0.95 and 0.99 on this dataset at a threshold of 50%. The F1 score ranged from 0.3-0.76 for chronic findings such as lung nodules,
calcifications, and fibrosis to between 0.86-0.93 for acute/life-threatening findings such as pleural effusion, pneumothorax, and
consolidation, with the highest positive predictive value of 0.97 for consolidation.
Conclusion: The AI algorithm performed better in identifying acute or life-threatening findings compared to chronic findings on CXRs
and can be a useful triage tool in acute clinical settings in multi-ethnic South-East Asian cohorts. This ongoing study also
demonstrates the utility of algorithmic monitoring of AI model performance through AimSG.
Limitations: Small sample size.
Ethics committee - additional information: This study was ethically approved by the Singhealth CIRB Ref: 2021/2337.
MammoGPT: a large language model for mammography-related dialogues (7 min)

Ma Jie; Shenzhen / China
1003
Author Block: Z. Cao, M. Jie, T. Liao, Y. Yang, X. Lin, J. Yuan, L. Ma; Shenzhen/CN
Purpose: The objective of this study was to develop MammoGPT, a Large language model (LLM), to assist patients with questions
related to Mammography’s diagnostic reports.
Methods or Background: A large-scale dataset containing 33,630 mammogram studies with diagnostic reports from 30,495
patients was collected from three collaborative hospitals at distinct geographical locations using Siemens and Giotto equipment
following the ACR standard from 2011 to 2021. We set radiologists into two groups, junior and senior, separated by their years of
expertise. We recorded 3,753 questions from relevant patients. We used ChatGLM-6B as the initial language model (ILM) to provide
three answers to each question with the corresponding diagnostic report. Junior radiologists corrected any common sense errors
inside those answers. Then, senior radiologists picked the best answer based on its correctness and professional level. Each training
sample consisted of a patient's question, a diagnostic report, and the best answer. These samples were used to train ILM into
MammoGPT. Eight Nvidia A100 were used for training and two for inference.
Results or Findings: We compare the performance of MammoGPT with three open-source LLMs: Vicuna-13B, ChatGLM-6B, and
Llama 2-Chat. Twenty patients were invited to provide their questions and diagnostic reports. We requested six senior radiologists to
evaluate answers generated by each LLM from correctness, rationality, helpfulness, and professionalism. MammoGPT scored highest
on all but rationality, with scores of 90.2, 67.6, 88.0, and 94.6. In particular, MammoGPT showed strong professionalism compared to
other LLMs, with a noticeable leading gap of over 15 points.
Conclusion: We present MammoGPT, an LLM that assists patients with questions related to Mammography reports. This LLM shows a
distinctive professional level and proves helpful along with correct answers based on our experiments.
Limitations: This model cannot deal with visual inputs for now.
Ethics committee - additional information: This project was approved by the IRB number LL-XJS-2020011. The ethics review and
institute approved this retrospective case-control study- national review board, which waived individual informed consent
requirements.
1004
CB - Case-Based Diagnosis Training
Categories: Education, General Radiology, Radiographers

ETC Level: LEVEL I
Date: March 3, 2024 | 09:30 - 12:00 CET
CME Credits: 2.5
Moderators:
Klaus M. Friedrich; Vienna / Austria

Cardiac (10 min)

Dietrich Beitzke; Vienna / Austria
Neuro (10 min)

Daniela Prayer; Vienna / Austria
Paediatrics (10 min)

Maxillofacial (10 min)

Genitourinary (10 min)

Michael Toepker; Wien / Austria
Interlude: The road less travelled: tracing clinical clues to a diagnosis (30 min)
Burce Özgen Mocan; Chicago / United States
Head and neck (10 min)

Ursula Schwarz-Nemec; Vienna / Austria
Chest (10 min)

1005
Spine (10 min)

Abdominal (10 min)

Breast (10 min)

1006
RC 2310 - Bone marrow imaging: new techniques for old clinical problems
Date: March 3, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Apostolos H. Karantanas; Heraklion / Greece

Apostolos H. Karantanas; Heraklion / Greece
How to image bone marrow? (15 min)

Bruno Vande Berg; Kessel-Lo / Belgium
1. To review the appearance of normal bone marrow by using currently available MR sequences.
2. To emphasise the merit of MR imaging technique in bone marrow disease.
3. To discuss the diagnostic strategy in bone metastases.
Bone marrow lesions: benign or malignant? (15 min)

Patrick Omoumi; Lausanne / Switzerland
1. To describe and reflect on the importance of the imaging of fat in distinguishing benign and malignant bone marrow lesions.
2. To list the sequences that should be part of all MRI protocols aiming at the assessment of bone marrow.
3. To diagnose the most common benign bone marrow lesions.
Bone marrow necrosis (15 min)

Charbel Mourad; Achrafieh, Beyrouth / Lebanon
1. To describe the typical MR imaging features of yellow (fatty) marrow necrosis.

2. To describe MR imaging features of red (hematopoietic) marrow necrosis.
3. To identify signs of early epiphyseal collapse of femoral head osteonecrosis on radiographs and MRI.
4. To discuss the role of CT as a problem-solving technique in early collapse of the femoral head osteonecrosis.
Panel discussion: Which protocol for which indication? (10 min)
1007
RC 2314 - Recent developments in MR safety
Categories: Education, Imaging Methods, Professional Issues, Radiographers, Students

Date: March 3, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderators:
Andrea Cradock; 04 / Ireland

Andrea Cradock; 04 / Ireland
Making the case for formal MR Safety Officer training (15 min)
Kurt Van Belle; Sint Andries / Belgium
1. To understand the need for basic MR safety training for all MR radiographers and advanced training for MR safety officers.
2. To describe why formal MR safety officer training is needed across Europe.
3. To learn to apply the European qualification framework to MR safety officer training.
Current safety considerations and practices for patients with implantable devices (15 min)
1. To learn to distinguish the different types of implantable devices that a patient undergoing an MR examination may have.
2. To recognise the risks and hazards of MR in patients with implantable devices.
3. To be aware of the current MRI safety considerations, practices and risks to promote best practices for patients with implantable
devices, enhancing MRI safety as a major element in the organisation of an MR department.
What do patients need to understand about MR safety (15 min)

Christos Tsiotsios; Limassol / Cyprus
1. To summarise the basic principles of MRI.

2. To identify the possible risks associated with the MRI environment and MRI examinations.
3. To understand what information needs to be communicated to patients regarding MR safety.
Panel discussion: What are the key steps to improve the safety profile of MRI (10 min)
1008
US 23 - Contrast-enhanced ultrasound (CEUS) in the intervention
Categories: Education, EuroSafe Imaging/Radiation Protection, Evidence-Based Imaging, Imaging Methods, Interventional Radiology
Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderators:
Markus Herbert Lerchbaumer; Berlin / Germany

CEUS for biopsy and ablation (16 min)

Ole Graumann; Odense / Denmark
1. To understand the role of CEUS in enhancing the accuracy and safety of biopsy and ablation procedures.
2. To acquire knowledge of the specific CEUS techniques used for biopsy and ablation guidance.
3. To evaluate clinical scenarios where CEUS-guided biopsy and ablation are particularly beneficial.
CEUS for drainage and endocavitary CEUS (16 min)

Timm Kleffel; Augsburg / Germany
1. To comprehend the utility of CEUS in drainage procedures and its impact on clinical outcomes.
2. To understand the principles and techniques of endocavitary CEUS.
3. To recognise the diverse clinical applications of endocavitary CEUS.
CEUS with fusion for intervention (16 min)

1. To understand the fundamentals of fusion imaging and its integration with CEUS in interventional procedures.
2. To identify strategies for overcoming common challenges associated with CEUS-guided fusion imaging.
3. To recognise the range of clinical applications and potential benefits of CEUS-guided fusion imaging in interventional practice.
CEUS in a paediatric intervention (16 min)

Abhay Srinivasan; Philadelphia / United States
1. To understand the unique considerations and safety profile of CEUS in paediatric interventional procedures.
2. To develop proficiency in the techniques and best practices for performing CEUS-guided interventions in children.
3. To recognise the wide range of clinical applications and benefits of CEUS in paediatric interventional practice.
Novel applications for CEUS in an intervention (16 min)

1009
1. To explore emerging and innovative applications of CEUS in interventional procedures.
2. To assess the clinical relevance and potential benefits of emerging CEUS applications in various interventional scenarios.
3. To identify ongoing research and future directions of novel CEUS applications in interventional practice.
1010
RC 2308 - State-of-the-art post-operative temporal bone
Categories: Education, Head and Neck, Imaging Methods, Neuro

Date: March 3, 2024 | 09:30 - 10:30 CET
CME Credits: 1
Moderator:
Salman Qureshi; Sale / United Arab Emirates

Salman Qureshi; Sale / United Arab Emirates
Imaging of the post-operative middle ear and mastoid (15 min)

Sabrina Kösling; Halle A.D. Saale / Germany
1. To review the indications and most common types of surgical procedures in the middle ear.
2. To apprehend the typical appearance of the reconstructed ossicular chain.
3. To discuss the imaging findings of possible complications.
Post-implant imaging for cochlear and brainstem implants (15 min)

1. To present indications and functions of cochlear and brainstem implants.

2. To review expected imaging findings after implantation.
3. To discuss possible complications and the role of different imaging modalities for post-implant assessment.
Post-treatment imaging for vestibular schwannomas (15 min)

Ayça Karaosmanoğlu; Ankara / Turkey
1. To discuss different treatment options for vestibular schwannomas.

2. To present recommendations for surveillance imaging and to depict the expected post-operative imaging appearances.
3. To discuss tools for the detection of residual or recurrent disease and for assessment of complications.
Panel discussion: Post-treatment imaging - how to overcome the artefacts and things I wished someone had told me
(10 min)
1011
RC 2302 - Current applications of artificial intelligence (AI) in breast cancer imaging
Categories: Artificial Intelligence & Machine Learning, Breast

ETC Level: LEVEL II
Date: March 3, 2024 | 09:30 - 10:30 CET
CME Credits: 1

Artificial intelligence (AI) in breast imaging: where we stand? (15 min)

Fredrik Strand; Stockholm / Sweden
1. To understand alternative implementations of AI in screening.

2. To know the results from prospective trials on AI in screening.
3. To have an overview of potential AI use outside of screening.
AI in screening (15 min)

Tamar Sella; Jerusalem / Israel
1. To present the applications of artificial intelligence in screening methods (mammography/ DBT/ US and MRI).
2. To summarise AI's advantages and limitations in mammography, DBT, US and MRI.
3. To identify the challenges of the integration of AI into screening.
The role of radiomics and radio-genomics for breast cancer prognosis (15 min)
1. To understand how AI can be used in predicting breast cancer.

2. To describe the role of AI-enhanced breast imaging in providing information on tumour prognostic factors.
3. To present the current limitations and emerging clinical applications of AI-enhanced breast imaging in clinical practice.
Panel discussion: Current challenges in the application of AI in screening (10 min)
1012
EU 23 - Communicating benefits and risks of medical radiation to patients undergoing radiological

procedures
Categories: EuroSafe Imaging/Radiation Protection, Interventional Radiology, Professional Issues, Radiographers

ETC Level: LEVEL II
Date: March 3, 2024 | 09:30 - 11:00 CET
CME Credits: 1.5
Moderator:

What radiology professionals should know about what patients, parents and carers would like to know (18 min)
1. To appreciate that patients, parents, and carers seek answers on a healthcare issue.
2. To learn that they seek a personalised approach to them and their medical problem.
3. To understand that they have very different problems but also very different capacities to understand and cope, and, therefore,
need personalised communication.
Communicating with patients undergoing interventional radiology procedures (18 min)

Philipp Wiggermann; Regensburg / Germany
1. To appreciate the importance of clear risk and benefit communication towards patients for IR procedures not neglecting radiation
exposure.
2. To learn the marked variations of patients' radiation exposure depending on intervention type.
3. To understand the usefulness of longitudinal radiation exposure observation and increase efficiency in IR procedures.
What and how to communicate with pregnant patients (18 min)

Ilze Apine; Riga / Latvia
1. To understand the potential detrimental effects and risks of radiation exposure to pregnant women and foetuses.
2. To become familiar with the ethical concerns related to exposing pregnant women and foetuses to ionizing radiation.
3. To learn how to effectively communicate the risks and benefits of the radiological procedure to a pregnant patient.
How to communicate that patient contact shielding, in most cases, is no longer needed (18 min)
1. To appreciate the current understanding of the benefits and risks of contact shielding application during imaging examinations.
2. To understand current guidance on contact shielding use.
3. To be able to communicate the reasons why shielding is not routinely required in radiology.
Panel discussion: How to train radiology professionals to effectively communicate the benefits and risks of medical
radiation with patients? (13 min)
1013
BS 24a - Neuroradiology in daily clinical practice
Categories: General Radiology, Neuro

ETC Level: LEVEL II
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Philippe Demaerel; Leuven / Belgium

Philippe Demaerel; Leuven / Belgium
Imaging of vascular malformations of the brain (19 min)

Jose Luis Munuera del Cerro; Barcelona / Spain
1. To learn about the typical imaging features of vascular malformations.

2. To know the advantages and disadvantages of different modalities.
Imaging of intracranial infectious and inflammatory diseases (19 min)

Alexandre Krainik; Grenoble / France
1. To learn about imaging features of the most frequent infectious and inflammatory diseases.
2. To demonstrate the most important findings.
Imaging of ischaemic diseases of the brain (19 min)

Karl-Olof Loevblad; Geneva / Switzerland
1. To learn about different modalities to diagnose ischaemia.

2. To demonstrate the most important findings.
1014
RPS 2403 - Spectral CT: new developments and clinical applications (part 2)

Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Moritz Christian Halfmann; Mainz / Germany
Coronary artery calcium scoring using virtual vs true non-contrast images from photon-counting coronary CT
angiography (7 min)
Jan Robert Kröger; Minden / Germany
Author Block: N. Haag1, A. E. Michael1, S. Lennartz2, C. Panknin3, I. Shahzadi3, J. H. Niehoff1, J. Borggrefe1, J. R. Kröger1; 1Minden/DE,
2
Cologne/DE, 3Forchheim/DE
Purpose: Coronary artery calcium scoring (CACS) requires true non-contrast (TNC) CT. Photon-counting CT provides an algorithm
(PureCalcium) for reconstructing virtual non-contrast images from contrast-enhanced coronary CT angiography (CCTA) specifically for
CACS. The aim of this study was to assess the CACS differences based on PureCalcium images derived from contrast-enhanced CCTA
compared with TNC images and evaluate the impact of these differences on the classification of patients into plaque burden
categories.
Methods or Background: Images from patients who underwent photon-counting CCTA between August 2022 and May 2023 were
identified and retrospectively analysed. Agatston scores were derived from TNC and PureCalcium images and tested for agreement
using the intra-class correlation coefficient. Patients were categorised into different plaque burden groups based on Agatston scores
and agreement was evaluated using weighted Cohen’s kappa. Wilcoxon signed-rank test and Bland-Altman analysis were used to
evaluate differences, the DLP for the entire examination and TNC were analysed.
Results or Findings: Median Agatston scores were comparable between TNC and PureCalcium (4.8; IQR= 83.6 [0.0 – 2151.8] vs 2.7;
IQR =89.2 [0.0-2377.1], P=.99) and showed a strong agreement and relationship (ICC = 0.98, 95% CI [0.97, 0.99]. On Bland-Altman
analysis only two patients showed a bias outside of the limits of agreement. With PureCalcium CACS, 74% of patients were correctly
classified into plaque burden groups agreement excellent (κ = 0.88). TNC contributed a mean of 19.7 ± 8.8% SD of the radiation dose
of the entire examination.
Conclusion: Agatston scores for coronary artery disease derived from PureCalcium and TNC images from contrast-enhanced photon
counting CCTA high agreement and did not differ significantly. PureCalcium CACS correctly classified most patients into plaque
burden groups. Omitting TNC scans reduced the radiation dose.
Limitations: Distribution of Agatston scores was skewed and lower than expected in our study cohort.
Ethics committee - additional information: This study was approved by the local ethics committee as a retrospective study
In vitro study of the quality of multi-parameter coronary stent images on second-generation dual-layer detector
spectral CT (7 min)
Lihong Chen; Xi An / China
1015
Author Block: L. Chen, Z. Jian; Xi An/CN
Purpose: The aim of this study was to evaluate the image quality of the multi-parameter coronary stent images on second-
generation dual-layer detector spectral CT.
Methods or Background: Eighteen coronary stents of different materials and diameters were placed in simulated coronary vessels
containing a concentration of 11 mg/L iodine contrast agent. Scanning was performed using second generation dual-layer detector
spectral CT. 50~190 keV monoenergetic (20 keV interval), conventional, and iodine map images were reconstructed. Stent lumen
score, stent structure score, and degree of blooming artefact were subjectively evaluated (Likert 4-score). Stent lumen contrast noise
ratios (CNRs), non-stent lumen contrast noise ratios (CNRc), inner diameter differences (IDD), and blooming artifact index (BAI) were
used for objective evaluation. The stents were divided into small diameter subgroups (<3 mm) and large diameter subgroups (≥3
mm) according to their diameters. Repeated-measure ANOVA and Tukey's test were used for multi-comparison.
Results or Findings: For all stents and the large diameter subgroup, the optimal monoenergetic images had the highest stent
lumen score, which was significantly greater than those of iodine map images. The iodine map images had the highest stent structure
score, which was significantly greater than those of iodine maps (for all stents and small diameter subgroup) and conventional images
(for all stents and both subgroups). CNRs and CNRc decreased with increasing mono-energy level. For all stents and both subgroups,
the CNRs and CNRc of optimal monoenergetic images were greater than for conventional and iodine map images. For all stents,
iodine map images had the smallest IDD and BAI.
Conclusion: Monoenergetic imaging on the second-generation dual-layer detector spectral CT has advantages in assessing stent
lumen and improving CNR. Meanwhile, iodine map imaging has advantages in assessing stent structure and internal diameter.
Keywords: Stent; Computed tomography; Coronary vessels
Limitations: The main limitation was that this was a phantom study.
Ethics committee - additional information: No ethical clearance was sought for this study.
Ultra-high resolution coronary CT angiography on photon-counting detector CT: impact of quantum iterative
reconstruction (7 min)
Milán Vecsey-Nagy; Budapest / Hungary
Author Block: M. Vecsey-Nagy1, G. Tremamunno1, U. J. Schoepf1, N. Fink1, E. Zsarnóczay1, P. Maurovich-Horvat2, B. Szilveszter2, A.

Varga-Szemes1, T. S. Emrich1; 1Charleston, SC/US, 2Budapest/HU
Purpose: The objective of this study was to assess the impact of different quantum iterative reconstruction (QIR) levels on objective
and subjective image quality of ultra-high resolution (UHR) coronary CT angiography (CCTA) images and to determine the accuracy of
stenosis quantification using photon-counting detector CT (PCD-CT) .
Methods or Background: A dynamic vessel phantom containing two calcified lesions (25% and 50% stenosis) was scanned at 60,
80 and 100 beats per minute with a PCD-CT system. In vivo examinations were performed in 102 CCTA patients. All scans were
acquired in UHR mode (slice thickness: 0.2 mm, kernel: bv64) and reconstructed with four different QIR levels (1-4). Image noise,
signal-to-noise ratio (SNR), sharpness, and percent diameter stenosis (PDS) were quantified on the phantom, while subjective quality
metrics (graininess, sharpness, overall image quality) were assessed on clinical scans.
Results or Findings: Increasing QIR levels resulted in significantly lower objective image noise and higher SNR (p<0.001), while
sharpness and PDS values did not differ significantly between QIRs (all pairwise p>0.013). Subjective graininess of in vivo images
significantly decreased with increasing QIR levels, with significantly higher image quality scores for each stepwise increase (all
pairwise p<0.001). Qualitative sharpness, on the other hand, did not differ across different levels of QIR (p=0.15).
Conclusion: The novel QIR algorithm may enhance image quality of CCTA datasets without compromising image sharpness or
accurate stenosis measurements, with the most prominent benefits at the highest strength level.
Limitations: The subjective aspect of the qualitative analysis introduces the possibility of bias into the study. Moreover, although
potential usefulness of QIR in reducing radiation dose may be proposed, we have not yet examined methods of radiation dose
reduction with QIR.
Ethics committee - additional information: The protocol for this prospective, HIPAA-compliant study was approved by the local
Institutional Review Board (#6030).
Feasibility of coronary CT angiography-based fractional flow reserve using a clinical photon-counting detector CT
system (7 min)
Emese Zsarnóczay; Budapest / Hungary
1016
Author Block: E. Zsarnóczay1, D. Pinos2, U. J. Schoepf2, N. Fink2, J. O'Doherty2, M. Vecsey-Nagy2, P. Maurovich-Horvat1, T. S. Emrich2,
A. Varga-Szemes2; 1Budapest/HU, 2Charleston, SC/US
Purpose: The aim of this study was to intra-individually compare CT-FFR between photon-counting detector (PCD) and conventional
energy-integrating detector (EID) CT systems.
Methods or Background: In this single-centre prospective study, adults who underwent CCTA on an EID-CT system were recruited
for research PCD-CT scans between July 2021 to March 2022. EID-CT images were reconstructed (Bv36 kernel, iterative reconstruction
level 3, 0.5 mm slice thickness). PCD-CT images were post-processed with settings matched to EID-CT as close as possible (Bv36
kernel, quantum iterative reconstruction level 3, virtual mono-energetic level 55 keV, 0.6 mm slice thickness). CT-FFR was obtained
semi-automatically using a prototype on-site machine learning algorithm by two readers. The lowest CT-FFR value was used for the
per-patient analysis and a CT-FFR ≤0.75 was considered haemodynamically significant. Correlation and reliability between CT-FFREID-
CT and CT-FFRPCD-CT were assessed with Spearman (r) and intraclass correlation coefficients (ICC).
Results or Findings: A total of 22 patients (63.3±9.2 years; 7 women) were included. The median time between EID-CT and PCD-CT
was 5.5 days. Comparison of CT-FFR values showed no significant difference and strong agreement between EID-CT and PCD-CT in
the per-vessel analysis (0.88 [0.74-0.94] vs. 0.87 [0.76-0.93], P=0.096, mean bias 0.02, limits of agreement [LoA] -0.14/0.19, r=0.83,
ICC=0.92), and in the per-patient analysis (0.81 [0.60-0.86] vs. 0.76 [0.64-0.86], P=0.768, mean bias 0.02, LoA -0.15/0.19, r=0.90,
ICC=0.93). All included patients were classified into the same category (CT-FFR>0.75 vs ≤0.75) with both CT systems.
Conclusion: PCD-CT-based CT-FFR evaluation is feasible and correlates well with EID-CT-based CT-FFR.
Limitations: CT-FFR values from EID- and PCD-CT were not compared with invasively measured FFR. Only one reconstruction setting
was used for image postprocessing with EID- and PCD-CT.
Funding for this study: This research was partially funded by a grant from Siemens Healthineers.
Ethics committee - additional information: The protocol of this prospective, Health Insurance Portability and Accountability
Act–compliant, single-centre study was approved by the local Institutional Review Board, and written informed consent was obtained
from each participant.
Impact of reconstruction parameters on the accuracy of myocardial extracellular volume quantification on a first-
generation photon-counting detector CT (7 min)
Chiara Gnasso; Milan / Italy
1 2 2 2 2 2 3 2
Author Block: C. Gnasso , D. Pinos , U. J. Schoepf , M. Vecsey-Nagy , E. Zsarnóczay , N. Fink , R. J. Holtackers , A. Varga-Szemes , T.
2 1 2 3
S. Emrich ; Milan/IT, Charleston, SC/US, Maastricht/NL
Purpose: The potential role of cardiac CT has increasingly been demonstrated for the assessment of diffuse myocardial fibrosis
through the quantification of extracellular volume (ECV). The recently introduced photon-counting detector (PCD)-CT technology may
deliver more accurate ECV quantification compared to energy-integrating detector CT. This study aimed to establish the image
reconstruction settings for PCD-CT delayed acquisitions yielding the highest agreement in ECV quantification compared to MRI-ECV.
Methods or Background: Patients (n=27, 53.1±17.2 years; 14 women) prospectively underwent same-day cardiac MRI and PCD-
CT. Delayed CT scans were reconstructed with different quantum iterative reconstruction levels (QIR 1-4), slice thicknesses (s.t. 0.4-8
mm), and virtual monoenergetic imaging levels (VMI, 40-90 keV), as shown in Figure 1. ECV was quantified for each reconstruction
setting and compared using ANOVA for repeated measures and t-test for pairwise comparisons. The agreement with MRI was
assessed with Bland-Altman plots and Lin’s concordance correlation coefficient (CCC).
Results or Findings: ECV values didn’t differ significantly among QIR levels (p=1.00), whereas a significant difference was observed
throughout different slice thicknesses, with 0.4 mm yielding the highest agreement (CCC 0.944). 45 keV VMI reconstructions had the
lowest mean bias (0.6, 95%CI 0.1–1.4) compared to MRI. Using the most optimal reconstruction settings (QIR4; s.t. 0.4 mm; VMI 45
keV), a 63% reduction in mean bias and a 6% increase in concordance with MRI were achieved compared to the standard settings
used in previous publications (QIR3; s.t. 1.5 mm; VMI 65 keV).
Conclusion: The appropriate selection of CT reconstruction parameters improves the agreement between PCD-CT and MRI ECV
values.
Limitations: Our study has the following limitations: first, it is a single-centre study with a relatively small sample size; second, the
lack of a histological reference, and third, the use of MRI-derived synthetic Hct.
Funding for this study: This study was supported by an institutional research grant from Siemens Healthcare.
Ethics committee - additional information: The study is HIPAA-compliant.
Impact of photon-counting CT-based virtual mono-energetic imaging on detecting myocardial late iodine enhancement:
comparison with MRI reference (7 min)
Chiara Gnasso; Milan / Italy
1017
Author Block: C. Gnasso1, M. Vecsey-Nagy2, U. J. Schoepf2, D. Pinos2, E. Zsarnóczay2, G. Tremamunno2, R. J. Holtackers3, T. S.
Emrich4, A. Varga-Szemes2; 1Milan/IT, 2Charleston, SC/US, 3Maastricht/NL, 4Frankfurt/DE
Purpose: Cardiac CT has the ability to detect myocardial scar, which can potentially be improved by the use of the recently-
introduced photon-counting detector (PCD)-CT technology. This study aims to investigate the image quality and diagnostic
performance of late iodine enhancement (LIE)-CT scans at different virtual monoenergetic image (VMI) levels in detecting and
characterising myocardial scars, using late gadolinium enhancement (LGE)-MRI as the reference standard.
Methods or Background: This is a post-hoc analysis of a prospective research study including patients with various
cardiomyopathies who underwent clinical LGE-MRI and same-day research LIE-CT scan between July 2021 and January 2022. LIE PCD-
CT scans were reconstructed at different VMI levels (40, 45, 50, 60, 70, and 90 keV). Two blinded readers evaluated subjective and
objective image quality, presence of scar and scar pattern on a per-segment level. The diagnostic performance was evaluated as
sensitivity, specificity, and accuracy. Agreement with MRI in scar detection and pattern identification were evaluated with Cohen’s κ
statistic.
Results or Findings: The LIE-CT scans of 27 patients (48% male, mean age 52.9 ± 17.2 years) were analyzed. VMI at 50 keV
demonstrated an adequate objective and subjective image quality and the best trade-off between sensitivity (87.4%), specificity
(97.8%), and accuracy (95.9%), in scar detection, with an excellent agreement with MRI (κ = 0.86). 50 keV also showed the highest
concordance in discriminating different scar patterns, with a 100% rate for detecting subepicardial scars and patchy fibrosis.
Conclusion: PCD-CT VMI reconstructions at 50 keV are a valuable tool for myocardial scar detection and characterisation,
demonstrating excellent diagnostic performance and agreement with cardiac MRI.
Limitations: Due to the sub-analysis nature of this study, selection bias needs to be considered. Due to the limited number of
patients, per-patient analysis was not performed.
Ethics committee - additional information: The study is HIPAA-compliant.
Dual-layer spectral CT-based quantification of myocardial extracellular volume fraction using CMR as a reference in
aortic valve disease (7 min)
Gang Wang; Lanzhou / China
1 1 1 1 1 1 2 1 1 1 2
Author Block: Y. Li , X. He , J. Nan , W. Tian , X. Lu , L. Cao , Y. Wang , G. Wang , Z. Chen ; Lanzhou/CN, Shanghai/CN
Purpose: The objective of this study was to assess the agreement of myocardial extracellular volume fraction (ECV) measurements
in patients with aortic valve disease (AVD) between dual-layer spectral detector CT (DLCT) and cardiovascular magnetic resonance
(CMR) measurements.
Methods or Background: Twenty-three patients diagnosed with AVD who underwent both simultaneous DLCT late iodine
enhancement and CMR examinations were retrospectively enrolled. These examinations performed within a week. ECV values were
calculated from the iodine map (CT-ECV-Iodine) and the conventional CT images (CT-ECV-HU (Hounsfield units)) derived from DLCT.
CT-ECV-Iodine, CT-ECV-HU, and CMR-ECV were measured at three short axes (base, middle, and apex), according to the American
Heart Association 16-segment model. Global and segmental ECVSs of cardiac is recorded. The Pearson correlation coefficient and
Bland-Altman plot were used to calculate the correlation and agreements among these measurements.
Results or Findings: The correlation analysis revealed a strong correlation between CT-ECV-Iodine and CT-ECV-HU (r =0.954, 95%
confidence interval (CI) [0.89-0.98], p <0.001), a relatively high correlation between CT-ECV-Iodine and CMR-ECV (r =0.862, 95% CI
[0.70-0.94], p <0.001), and a significant correlation between CT-ECV-HU and CMR-ECV (r =0.812, 95% CI [0.60-0.92], p <0.001). The
mean differences between these measurements were as follows: CT-ECV-Iodine and CT-ECV-HU exhibited a mean difference of
-0.40% (with a 95% limit of agreement (LOA) ranging from -3.2% to 2.4%), CT-ECV-Iodine and CMR-ECV showed a mean difference of
0.76% (95% LOA: -3.5-5.0%), and CT-ECV-HU and CMR-ECV had a mean difference of 1.17 (95% LOA: -4.22-6.55%).
Conclusion: DLCT-based ECV measurements showed a strong correlation with CMR in patients with AVD. Furthermore, it is a
potential alternative way for assessment of myocardial interstitial fibrosis in AVD.
Limitations: Small sample size is the limitation of this study.
and approved by the Ethical Committee of the first hospital of Lanzhou University.
Quantification of myocardial extracellular volume and aortic valve calcification scores using dual-layer spectral CT to
identify severe aortic regurgitation in patients with aortic valve disease (7 min)
ZiXian Chen; Lanzhou / China
1018
Author Block: Y. Li1, X. He1, L. Cao1, W. Tian1, X. Lu1, J. Xu1, Y. Wang2, G. Wang1, Z. Chen1; 1Lan Zhou city/CN, 2Shanghai/CN
Purpose: To investigate the diagnostic performance of a combination of myocardial extracellular volume fraction (ECV) and aortic
valve calcification scores (AVCS) measured using dual-layer spectral detector CT (DLCT) in the identifying of severe aortic
regurgitation (sAR) in patients with aortic valve disease (AVD).
Methods or Background: A total of 54 patients with AVD (37 aortic regurgitation (AR), including 16 severe AR (sAR) and 21 with
mild to moderate AR (mAR); 17 aortic stenosis (AS)) who underwent non-contrast and late iodine enhancement DLCT examinations
were retrospectively included. The ECV map was reconstructed using a dedicated workstation, and AVCS was evaluated semi-
automatically using multiplanar reconstruction by Agatston algorithm from the non-contrast phase of DLCT. AVCS scores were
categorized into four grades. Student’s t-test and Mann-Whitney U test were performed. The feasibility of CT-ECV combined with AVCS
grades in identifying sAR was evaluated using the receiver operating characteristic curve.
Results or Findings: In the sAR group, the ECV was significantly higher (32.29 [30.26-35.57]% vs 29.30 [27.08, 32.71]% , p=0.03)
than non-sAR group, while the AVCS grades were significantly lower (1.88±0.48 vs 2.5±0.94, p=0.02). Using an ECV cutoff of 29.13%,
the AUC for discriminating sAR was 0.75. The combined model of ECV and AVCS grades demonstrated notable diagnostic performance
in distinguishing sAR from non-sAR, with an AUC of 0.82, sensitivity of 94% and specificity of 55%. To differentiate between AR
severity levels, there was a significant difference in ECV between the sAR group and the mAR group (p =0.02). The AUC was 0.79,
with a sensitivity of 75% and specificity of 72%.
Conclusion: The combination of DLCT-ECV and AVCS grades, assessed via spectral CT, shows promise for identifying severe aortic
regurgitation in patients with aortic valve disease.
Limitations: Small sample size was a limitation of this study.
and approved by the Ethical Committee of the first hospital of Lanzhou University.
1019
RPS 2401 - Artificial intelligence in abdominal imaging: current directions
Categories: Abdominal Viscera, Artificial Intelligence & Machine Learning, Oncologic Imaging
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Risk stratification of gallbladder masses by ML‑based ultrasound radiomics models: a prospective and
multi‑institutional study (7 min)
Chong-Ke Zhao; Shanghai / China
Author Block: C-K. Zhao; Shanghai/CN

Purpose: This study aimed to evaluate the diagnostic performance of machine learning (ML)–based ultrasound (US) radiomics
models for risk stratification of gallbladder (GB) masses.
Methods or Background: We prospectively examined 640 pathologically confirmed GB masses obtained from 640 patients between
August 2019 and October 2022 at four institutions. Radiomics features were extracted from greyscale US images and germane
features were selected. Subsequently, 11 ML algorithms were separately used with the selected features to construct optimum US
radiomics models for risk stratification of the GB masses. Furthermore, we compared the diagnostic performance of these models with
the conventional US and contrast-enhanced US (CEUS) models.
Results or Findings: The optimal XGBoost-based US radiomics model for discriminating neoplastic from non-neoplastic GB lesions
showed higher diagnostic performance in terms of areas under the curves (AUCs) than the conventional US model (0.822–0.853 vs.
0.642–0.706, p < 0.05) and potentially decreased unnecessary cholecystectomy rate in a speculative comparison with performing
cholecystectomy for lesions sized over 10 mm (2.7–13.8% vs. 53.6–64.9%, p < 0.05) in the validation and test sets. The AUCs of the
XGBoost-based US radiomics model for discriminating carcinomas from benign GB lesions were higher than the conventional US
model (0.904–0.979 vs. 0.706–0.766, p < 0.05). The XGBoost-US radiomics model performed better than the CEUS model in
discriminating GB carcinomas (AUC: 0.995 vs. 0.902, p = 0.011).
Conclusion: The proposed ML-based US radiomics models possess the potential capacity for risk stratification of GB masses and may
reduce the unnecessary cholecystectomy rate and use of CEUS.
Limitations: First, the sample size was relatively limited. Second, since surgery-confirmed pathology was used as the gold standard,
GB lesions sized less than 6 mm were not included, which are commonly not recommended for cholecystectomy.
Funding for this study: This work was supported in part by the National Natural Science Foundation of China (Grant 82202174).
Ethics committee - additional information: This study was approved by the ethics committee of the institution (No:
SHSYIEC–4.1/21–263/01; 2022-187R).
Development and validation of comprehensive nomogram based on imaging features and MRI radiomics to predict
microvascular invasion and overall survival in patients with intrahepatic cholangiocarcinoma (7 min)
Gengyun Miao; Shanghai / China
1020
Author Block: G. Miao, X. Qian, Y. Zhang, C. Yang, M. Zeng; Shanghai/CN
Purpose: Microvascular invasion (MVI) is a predictor of poor prognosis in intrahepatic cholangiocarcinoma (ICC). The aim of this study
was to establish a comprehensive model based on MR radiomics for MVI status stratification and overall survival prediction in ICC
patients preoperatively.
Methods or Background: A total of 249 ICC patients were randomized into training and validation cohorts (174:75), and a time-
independent test cohort with 47 ICC patients was enrolled. Independent clinical and imaging predictors were identified by univariate
and multivariate logistic regression analyses. The radiomic model was based on the robust radiomic features extracted by a logistic
regression classifier and the least absolute shrinkage and selection operator algorithm. The imaging-radiomics (IR) model integrated
the independent predictors and robust radiomics features. The predictive efficacy of the models was evaluated by receiver operating
characteristic curves, calibration curves and decision curves. Multivariate Cox analysis identified the independent risk factors for
overall survival, Kaplan‒Meier curves were plotted, and a nomogram was used to visualize the predictive model.
Results or Findings: The imaging model comprised tumour size and intrahepatic duct dilatation. The radiomics model comprises 25
stable radiomics features. The IR model shows desirable performance (AUCtraining= 0.890, AUCvalidation= 0.885 and AUCtest=
0.815). The calibration curve and decision curve validate the clinical utility. Overall survival predicted by histological and IR model-
predicted MVI groups exhibited similar predictive efficacy.
Conclusion: The IR model and nomogram based on IR model-predicted MVI status may be a potential tool in MVI status stratification
and overall survival prediction of ICC patients preoperatively.
Limitations: The models are based on retrospectively collected data from a single institution.
Funding for this study: This work was supported by 1. Shanghai Municipal Health Commission (Grant number 202240152); 2.
National Natural Science Foundation of China (Grant number 82171897); 3. Shanghai Municipal Key Clinical Specialty (Grant number
shslczdzk03202); 4. Clinical Research Plan of SHDC (Grant number SHDC2020CR1029B).
Ethics committee - additional information: Approval for this retrospective study was granted by the Ethics Committee of our
Hospital.
Automatic detection and segmentation of IPMN pancreatic cysts in MRI with a multi-sequence cascaded deep learning
pipeline (7 min)
Leo Joskowicz; Jerusalem / Israel
Author Block: N. Mazor, N. Lev-Cohain, R. Lederman, J. Sosna, L. Joskowicz; Jerusalem/IL

Purpose: Radiological detection and follow-up of IPMN pancreatic cysts in multi-sequence MRI studies is required to assess their
malignancy potential. The evaluation requires expertise and is not automated. This study evaluates a novel multi-sequence cascaded
deep learning pipeline for the detection and segmentation of IPMN pancreatic cysts in abdominal MRI.
Methods or Background: The pipeline consists of three steps: (1) pancreas Region of Interest (ROI) segmentation in the axial MRI
TSE; (2) transfer and masking of the computed pancreas ROI to the coronal MRI MRCP; (3) detection and segmentation of cysts in the
masked MRCP. Both steps 1 and 3 use 3D U-Net models with Hard Negative Patch Mining, a new technique for class imbalance
correction and reduction of false positives. The pipeline was evaluated on 158 MRI patient studies of patients with pancreatic cysts
undergoing follow-up. The training/validation/testing sets split was 118/17/23. Ground truth segmentations of a total of 840 cysts
were manually obtained by an expert radiologist: Six hundred and nineteen cysts were >5 mm, 221 cysts were >10 mm with a mean
number of cysts/scan of 5.3±2.6 and mean cyst diameter (volume) of 7.4 mm (0.91cc). The computed test set results were then
compared to their respective manual ground truth delineations.
Results or Findings: The pipeline achieved mean recall of 0.80±0.19 and 0.99±0.05, precision of 0.75±0.26, and 0.95±0.16, and
dice score of 0.80±0.19 and and 0.81±0.11 for pancreatic cysts of diameter > 5 mm and > 10mm respectively, which is the clinically
relevant endpoint. The cyst inclusion in the pancreas ROI Recall is 0.94±0.22 and 0.98±0.07, respectively.
Conclusion: Automatic pancreatic cyst detection and segmentation in multi-sequence MRI may provide an accurate and reliable
method for precise disease evaluation and save time.
Limitations: The major limitation of this study was this was of single observer annotation, one centre.
Ethics committee - additional information: No ethical approval was obtained for this study.
Delta CT-radiomics derived response prediction in advanced pancreatic ductal adenocarcinoma (7 min)
Felix N. Harder; Munich / Germany
1021
Author Block: F. N. Harder1, G. O'Kane2, E. Salinas-Miranda2, K. Lajkosz2, A. Farooq2, S. Gallinger2, J. Knox2, M. Haider2; 1Munich/DE,
2
Toronto, ON/CA
Purpose: The aim of this study was to evaluate multi-time point kinetic delta radiomics for response and overall survival (OS)
prediction in advanced PDAC.
Methods or Background: One hundred and fifty-seven patients with advanced PDAC (108/157 with synchronous liver metastases)
were retrospectively enrolled serving as the training cohort. Twenty-eight patients with metastatic PDAC from a second prospective
study served as an external validation cohort. All patients received mFOLFIRINOX or gemcitabine/nab-paclitaxel as first-line
chemotherapy. Baseline CT-models and delta radiomics models reflecting the kinetic between baseline and first follow-up CT were
build based on size-related, non-size and combined size and non-size features from the primary tumour and largest liver metastasis to
predict progression under chemotherapy based on RECIST1.1 and OS at 9 months in the overall cohort and liver metastases
subgroup. Baseline and delta radiomics models were compared against each other and established Moffitt RNA-signature.
Results or Findings: Non-size and combined delta-radiomics models significantly discriminated between responders
(complete/partial response and stable disease) and non-responders (progressive disease) in the training cohort and external
validation cohort (AUC 0.714-0.873, p = < 0.001-0.01) outperforming baseline-only models (AUC 0.55-0.645) and Moffitt RNA
signature (0.551-0.675). Radiomics models represented an independent survival predictor at 9 months in the training and the
external validation cohort, with non-size models yielding the highest AUC in the training cohort, yet not significantly outperforming
Moffitt RNA signature (training: 0.726 vs 0.588, p = 0.16; validation: 0.713 vs. 0.567, p = 0.35).
Conclusion: Delta-radiomics models outperformed baseline-models and Moffitt RNA signature as predictive biomarkers for response
and OS prediction in advanced PDAC. In particular non-size feature models from the primary tumour and the largest liver metastasis
provided additive value.
Limitations: Although including the largest cohort for radiomics-based response prediction in advanced PDAC so far, further studies
need to validate the herein-found results.
Funding for this study: Funding for this study was provided by the Ontario Institute of Cancer Research, The Sinai Health
Foundation and University Medical Imaging Toronto. F.N.H. received funding from Deutsche Forschungsgemeinschaft, HA 9949/1-1.
Ethics committee - additional information: This study was approved by our institutional research ethics board and informed
consent was obtained.
Radiomic feature profiles to define treatment response in rectal cancer: not just a tumour matter (7 min)
Ana Marhuenda; Valencia / Spain
1 1 1 1 2 1 1
Author Block: A. Marhuenda , A. Nogue , M. Domingo Pomar , I. Machado , R. Garcia Figueiras , F. Bellvis , A. Fuster Matanzo , A.
1 1 1 2
Jimenez-Pastor , A. Alberich-Bayarri ; Valencia/ES, Santiago de Compostela/ES
Purpose: In rectal cancer (RC), predicting response to specific treatments is essential for defining appropriate therapeutic strategies.
Moreover, the involvement of peritumoral regions and/or the presence of certain histopathological conditions are associated with poor
outcomes. The project aimed to assess whether radiomics may help stratify patients at baseline based on treatment response and
involvement of peritumoural regions.
Methods or Background: A retrospective, single-centre study was conducted. Baseline T2W MRIs of RC patients receiving
neoadjuvant treatment (NT) and surgery were included. Manual delineation of seven labels was performed: tumour, extramural
venous invasion (EMVI), tumoural deposits (TD), lymph nodes (LN) including intra- and extra- mesorectum nodes, peritumoural wall
(PW), mesorectum fat, and presacral space (PS). Radiomic features were extracted for each segmentation, and four models were
evaluated: tumour [model 1], tumour surroundings (EMVI, TD, LN, and mesorectum) [model 2], model 1 plus model 2 [model 3], and
model 3 with PW and PS [model 4]. Univariate and multivariate (logistic regression) analyses were performed.
Results or Findings: A total of 50 RC patients who received any type of NT and underwent surgery were included. In the univariate
analysis, the greatest differences between responders and non-responders were found in model 4. Statistical differences (p < 0.05)
were noted in four radiomic features—Kurtosis, GLRLM Run Entropy, NGTDM_Busyness and NGTDM_Strength. In the multivariate
analysis, model 4 outperformed the other models, with an AUC of 0.787.
Conclusion: Radiomic features could assist oncologists in therapeutic decision-making by predicting treatment responses.
Segmentations including tumour and peritumoral regions provide more solid results. This highlights the relevance of a more holistic
approach that would simplify segmentation´s tasks. Further studies with larger sample sizes are required.
Limitations: The limitations of the study are basically focused on the reduced number of patients.
Ethics committee - additional information: The study was approved by 2018-27.
Effect of artificial intelligence-aided differentiation of benign and premalignant colorectal polyps as a second reader at
CT colonography (7 min)
Sergio Grosu; Munich / Germany
1022
Author Block: S. Grosu, M. P. Fabritius, M. Winkelmann, S. Maurus, A. Graser, J. Ricke, P. M. Kazmierczak, M. Ingrisch, P. Wesp;
Munich/DE
Purpose: Premalignant adenomatous colorectal polyps require endoscopic resection, as opposed to benign hyperplastic colorectal
polyps. The aim of this study was to evaluate the effect of artificial intelligence (AI)-assisted differentiation of benign and
premalignant colorectal polyps as a second reader for general radiologists at CT colonography.
Methods or Background: CT colonography images with colorectal polyps of all sizes and morphologies were retrospectively
evaluated by three independent board-certified radiologists with moderate experience in CT colonography. The readers’ task was to
decide whether the depicted polyps required endoscopic resection. After a primary unassisted read, a second read with access to the
classification of a radiomics-based random forest AI model labelling each polyp as “adenomatous” or “hyperplastic” was performed.
No polyp used for training the AI model was included in this study. The performance of the unassisted and AI-assisted reading was
evaluated using polyp histopathology as the reference standard.
Results or Findings: Seventy-seven polyps in 59 patients comprising 118 polyp image series (47% supine position, 53% prone
position) were evaluated unassisted and AI-assisted by three radiologists, resulting in a total of 708 readings (subsequent
polypectomy: yes or no). Compared with unassisted reading, the AI-assisted reading had a significantly higher accuracy (76% +/- 1%
vs. 84% +/- 1%, p < 0.001), sensitivity (76% +/- 2% vs. 85% +/- 0%, p < 0.001), and specificity (75% +/- 1% vs. 81% +/- 2%, p <
0.001) in selecting polyps eligible for polypectomy.
Conclusion: In this proof-of-concept study, AI-based characterisation of colorectal polyps at CT colonography as a second reader
enabled a more precise selection of polyps eligible for subsequent endoscopic resection.
Limitations: The limitation of this study is the rather small sample size.
Ethics committee - additional information: This study was approved by the ethics committe of the Ludwig-Maximilians
University, Munich (18-401).
A radiomics model for the preoperative prediction of lymph node metastasis in colorectal carcinoma (7 min)
Alba Salgado-Parente; Madrid / Spain
Author Block: A. Salgado-Parente1, L. González Campo1, A. M. VERA CARMONA1, J. Soto1, B. Rodriguez-Vila2, I. De Vicente Bernal1, P.
Abadía Barnó1, A. Torrado-Carvajal2, J. Blazquez Sanchez1; 1Madrid/ES, 2Mostoles/ES
Purpose: The aim of this study was to develop and validate a radiomics nomogram for the preoperative prediction of lymph node
(LN) metastasis in colorectal carcinoma (CRC).
Methods or Background: The prediction model was developed in a primary cohort that consisted of 110 patients with
clinicopathologically confirmed CRC with data collected from January 2013 to September 2017 in a single institution. The patients
were divided into a training set (n = 88) and a validation set (n = 22) with statistically comparable demographic features. Radiomics
features of the primary tumor and lymph node were extracted from portal venous phase CT images of each patient. Mutual
information was used on the whole dataset for feature selection. The following models were trained: Random Forest, Logistic
Regression, Naive Bayes, Gaussian Process, Support Vector Machine, MultiLayer Perceptron, K-Nearest Neighbors, Gradient Boosting,
Neural Network. A crossvalidated fine-tuning of the hyperparameters was performed on each model to enhance the overall
performance. A majority voting approach was followed assessing the different combinations of the individual classifiers. The
combinations with the best F1 score performance were then selected to present the results.
Results or Findings: The radiomics signature demonstrated favorable discriminatory ability in predicting lymph node involvement
based on both tumor segmentation (ROC AUC 0.88, sensitivity 1, specificity 0.73) and lymph node segmentation (ROC AUC 0.94,
sensitivity 1, specificity 0.83). Accuracy was highly discriminative with values of 0.86 for tumor-based segmentations and 0.91 for
lymph node-based segmentations.
Conclusion: The CT-based radiomics nomogram has the potential to be used as a non-invasive tool for individualised preoperative
prediction of LN metastasis in CRC. External validation is further required prior to clinical implementation.
Limitations: The major limitations were that this was a single-centre and retrospective analysis.
Development and validation of CT-based radiomics deep-learning signatures to preoperatively predict lymph node
metastasis in non-functional pancreatic neuroendocrine tumour: a multi-cohort study (7 min)
Wei Tang; Shanghai / China
1023
Author Block: W. Tang; Shanghai/CN
Purpose: Lymph node status is an important factor for the patients with non-functional pancreatic neuroendocrine tumours (NF-
PanNETs) with respect to surgical methods, prognosis, recurrence. Our aim is to develop and validate a combination model based on
contrast-enhanced CT images to preoperatively predict the lymph node metastasis (LNM) in NF-PanNETs.
Methods or Background: Retrospective data were gathered for 320 patients with NF-PanNETs who underwent curative pancreatic
resection and CT imaging at two institutions (Centre 1, n= 236 and Centre 2, n=84) between January 2010 and March 2022. RDPs
(Radiomics deep learning signature) were developed based on ten machine-learning techniques. These signatures were integrated
with the clinicopathological factors into a nomogram for clinical applications. The evaluation of the model's performance was
conducted through the metrics of the area under the curve (AUC).
Results or Findings: The RDP signature showed excellent performance in both centres with a high AUC for predicting LNM and DFS
in centre 1 (AUC, 0.88; 95% CI: 0.84, 0.92; DFS, p <.05) and centre 2 (AUC, 0.91; 95% CI: 0.85, 0.97; DFS, p <.05). The clinical factors
of vascular invasion, perineural invasion, and tumour grade were associated with LNM (p <.05). The combination nomogram showed
better prediction capability for LNM (AUC, 0.93; 95% CI: 0.89, 0.96). Notably, our model maintained a satisfactory predictive ability for
tumours at the 2-cm threshold, demonstrating its effectiveness across different tumour sizes in centre 1 (≤2 cm: AUC, 0.90 and >2
cm: AUC, 0.86) and centre 2 (≤2 cm: AUC, 0.93 and >2 cm: AUC, 0.91).
Conclusion: Our RDPs may have the potential to preoperatively predict LNM in NF-PanNETs, address the insufficiency of clinical
guidelines concerning the 2-cm threshold for tumour lymph node dissection, and provide precise therapeutic strategies.
Limitations: Small sample size was a limitation of this study.
Funding for this study: This work was supported by Project of Shanghai Municipal Health Commission (202340123) and The Rare
Tumour Research Special Project of the National Natural Science Foundation of China (82141104).
Ethics committee - additional information: This two-centre retrospective study received approval from the Research Ethics
Committee of the Institutional Review Boards from all participating hospitals, and the need for informed consent was exempted.
1024
RPS 2416 - Imaging advances in oncology: from diagnosis to therapy
Categories: Imaging Methods, Oncologic Imaging

Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Ravikanth Balaji; Chennai / India
An external validation study of the Birmingham atypical cartilage tumour imaging protocol for the management of
solitary central cartilage tumours of the proximal humerus and around the knee (7 min)
Thomas Van Den Berghe; Lochristi / Belgium
Author Block: T. Van Den Berghe, F. Delbare, E. Candries, M. Lejoly, C. Algoet, F. Laloo, W. Huysse, D. Creytens, K. Verstraete;
Ghent/BE
Purpose: This study aimed to externally validate the Birmingham atypical cartilage tumour imaging protocol (BACTIP)
recommendations for differentiation and follow-up of central cartilage tumours (CCTs) of the proximal humerus, distal femur and
proximal tibia and to recommend BACTIP adaptations to reduce diagnostic delay in high-risk CCTs.
Methods or Background: MRIs of 123 patients (45±11 years, 37 men) with an untreated CCT with MRI follow-up (n=62) or
histopathological confirmation (n=61) were retrospectively and consecutively included and categorised following the BACTIP
(2003-2020/Ghent University Hospital/Belgium). Tumour length and endosteal scalloping differences between enchondroma, atypical
cartilaginous tumour (ACT) and high-grade chondrosarcoma (CS II/III/dedifferentiated) were evaluated. ROC-curve analysis for
differentiating benign from malignant CCTs and for evaluating the BACTIP was performed.
Results or Findings: For lesion length and endosteal scalloping, ROC-AUCs were poor and fair-excellent, respectively, for
differentiating between different CCT groups (0.59-0.69 versus 0.73-0.91). The diagnostic performance of endosteal scalloping and
the BACTIP was higher than that of lesion length. A 1° endosteal scalloping angle cut-off differentiated the enchondroma group from
the ACT/high-grade CS group with a sensitivity of 90%, reducing diagnostic delay. However, the specificity was only 29%, inducing
overmedicalisation with excessive follow-up. ROC-AUC of the BACTIP was poor for differentiating enchondroma from ACT (ROC-
AUC=0.69; 95%CI=0.51-0.87; p=0.041) and fair-good for differentiation between other CCT groups (ROC-AUC=0.72-0.81). BACTIP
recommendations were incorrect/unsafe in five ACTs and one CS II which were all discharged from follow-up, inducing diagnostic
delay. Eleven enchondromas received unnecessary referral and/or follow-up.
Conclusion: Although promising as a useful tool for the management and follow-up of CCTs of the proximal humerus, distal femur
and proximal tibia, five ACTs and one CS grade II were discharged, inducing diagnostic delay, which could be reduced by adapting
BACTIP cut-off values.
Ethics committee - additional information: This study was approved by the IRB (BC-08631).
Evaluation of the value of (dynamic) contrast-enhanced MRI for the diagnosis and follow-up of central cartilaginous
tumours (7 min)
Thomas Van Den Berghe; Lochristi / Belgium
1025
Author Block: T. Van Den Berghe, M. Lejoly, F. Delbare, W. Huysse, D. Creytens, G. M. Sys, K. Verstraete; Ghent/BE
Purpose: The objective of this study was to evaluate the value of (dynamic) contrast-enhanced MRI for the diagnosis and follow-up of
central cartilaginous tumours of the proximal humerus, distal femur and proximal tibia.
Methods or Background: Ninety-seven patients (44±11 years, 31 men) with a central cartilaginous tumour (histopathologically
proven or more than two years follow-up (5±3 years)) were retrospectively and consecutively included. Thickness of the enhancing
rim and dynamic contrast-enhanced MRI parameters were calculated. Tumour volumes were calculated for the assessment of tumour
growth. Significant differences between enchondromas, atypical cartilaginous tumours, high-grade chondrosarcomas, tumours with
and without growth at follow-up were searched for and ROC-curves were analysed.
Results or Findings: There was a significant difference (p=0.015) in enhancing rim thickness between high-grade chondrosarcomas
and other diagnostic central cartilaginous tumour groups. A ROC-curve with an AUC of 0.89 and cut-off value of 1.2 mm had a
sensitivity of 100% and a specificity of 64% to identify high-grade chondrosarcomas. A significant correlation was found between the
relative maximal enhancement of the whole tumour on dynamic contrast-enhanced MRI and the absolute growth rate (ml/year)
(ρ=0.75; p<0.001). Lesions with a relative maximal enhancement <1 compared to muscle remained stable or showed regression
during follow-up (mean -0.1±0.3 ml/year). Lesions with a relative maximal enhancement between one and two showed little growth
(mean +0.2±0.2 ml/year), and lesions with a relative maximal enhancement >2 had the highest growth rate (mean +0.4±0.2
ml/year).
Conclusion: A thick enhancing rim of more than 1.2 mm has a high sensitivity to detect high-grade chondrosarcomas. The higher the
relative maximal enhancement of the whole tumour on dynamic contrast-enhanced MRI, the higher the growth rate of a central
cartilaginous tumour.
Ethics committee - additional information: This study was approved by the IRB (BC-08631).
Changes in Deauville scores on delayed imaging with total-body 18F-FDG PET/CT scans: any association with outcome?
(7 min)
Yasser Gaber Abdelhafez; Sacramento / United States
Author Block: Y. G. Abdelhafez, C. Mingels, H. Nalbant, M. Rokni, F. Sen, R. Badawi, N. S. Esteghamat, J. Tuscano, L. Nardo;
Sacramento, CA/US
Purpose: Deauville score (DS) relies on comparing lesion uptake to reference tissues (liver and blood pool) that demonstrate tracer
washout over time. We aimed to evaluate DS changes for 18F-FDG total-body PET/CT scans acquired at 2-hr compared to 1-hr post-
injection and the association of these changes with clinical outcome data.
Methods or Background: Forty-five patients (54.3±19.1 years; 25 females) with lymphoma (Hodgkin's lymphoma=12; non-
Hodgkin's=33 [11 DLBCL, 13 follicular, and 9 marginal-zone]) underwent a total of 64 18F-FDG total-body PET/CT studies (36
staging/restaging, 22 interim, and 6 end-of-treatment [EOT] scans). All studies were acquired as dual-timepoint total-body scans at 1-
and 2-hr post-injection. Changes in scan-based DSs between the two timepoints were recorded by a single radiologist blinded to
clinical notes. Interim and EOT scans were evaluated for therapy response according to Lugano criteria. Follow-up data (progression
and/or all-cause mortality) was collected.
Results or Findings: The median follow-up from diagnosis was 32.6 months. A total of 15 out of 45 patients progressed, seven of
them died. One patient (2%) was upstaged from DS3 on 1-hr to DS4 on 2-hr interim PET scan. This upstaging would have changed the
response category from complete to partial metabolic response. Subsequent follow-up revealed complete response.
Conclusion: DSs assessed on delayed 2-hr 18F-FDG total-body PET/CT scans may potentially change the response category of a
small fraction (~2%) of lymphoma patients. Further validation of these initial findings in a larger cohort, with multiple readers, and
stratification according to the pathologic subtypes is ongoing.
Limitations: Small cohort limits this study.
Funding for this study: US National Institutes of Health R01-CA249422; R01-CA206187 funded with this study.
Ethics committee - additional information: This single-centre study was approved by the Institutional Review Board at University
of California Davis (IRB#1470016) and all included subjects signed an informed consent form.
Relaxation-compensated CEST imaging of the APT has greater predictive value in patients with glioma at baseline
before radiotherapy, compared to imaging of the APTw and ssMT at 3T (7 min)
Nikolaus von Knebel Doeberitz; Heidelberg / Germany
1026
Author Block: N. von Knebel Doeberitz, F. Kroh, S. Graß, L. König, C. Bauspieß, M. E. Ladd, H-P. Schlemmer, A. Korzowski, D. Paech;
Heidelberg/DE
Purpose: Chemical exchange saturation transfer (CEST) MRI can be applied to use endogenous compounds as peptides and semi-
solid macromolecules as imaging contrasts. However, CEST-contrasts are heavily dependent on the field strength and the applied
metrics for contrast reconstruction from the Z-spectrum. Therefore, for the purpose of this study was to compare the predictive value
of different CEST contrast of the amid proton transfer (APT) and semi-solid magnetization transfer (ssMT) in a larger clinical cohort of
patients with glioma before radiotherapy at 3T.
Methods or Background: Seventy-nine patients who had received biopsy or surgery for diffuse glioma prospectively underwent
CEST imaging of the APT and ssMT at the baseline MRI before radiotherapy, applying Lorentzian-fit- and asymmetry-based
reconstruction metrics first described by Zou et al. (APTw_asym), Mehrabian et al. (MTconst) and Zaiß et al. (APT_MTRRex and
MT_MTRRex). Contrast-enhancing (CE) and whole tumour (WT) volumes were segmented on contrast-enhanced T1w-CE and T2w-
FLAIR images. Therapy response at the first follow-up four to six weeks after the completion of radiotherapy and progression-free
survival (PFS) were assessed by longitudinal follow-up according to the response assessment in neuro-oncology (RANO) criteria.
Statistical testing included receiver-operator-characteristic (ROC) and Kaplan-Meier analyses.
Results or Findings: Imaging of the APT_MTRRex (CE: AUC=0.73, p=0.01) and MT_MTRRex (CE: AUC=0.67, p=0.05) was associated
with therapy response. The APT_MTRRex (WT: HR=2.75, p<0.01) was furthermore associated with PFS. Imaging of the APTw_asym
and MTconst were not associated with any outcome.
Conclusion: Relaxation-compensated CEST imaging of the APT has great predictive value in patients with glioma at the baseline MRI
at 3T, and might help to inform clinical decision making in the future.
Limitations: Single centre study. Median overall survival was not reached at the time of data analysis (05/07/2023), leading to
limitations.
Funding for this study: German Cancer Research Foundation (Grant No. 445704496) funded this study.
Ethics committee - additional information: The local institutional review board committee approved this prospective study.
Written informed consent was obtained from each participant prior to study inclusion.
Therapeutic application of single-atom iron nanozymes in magnetic resonance imaging, synergistic chemodynamic
therapy, and photothermal therapy in colorectal carcinoma (7 min)
Jun Zhang; Shanghai / China
Author Block: J. Zhang, L. Wang; Shanghai/CN

Purpose: This study was aimed at designing a nanotherapeutic strategy that incorporates nanoscale chemodynamic therapy while
integrating efficient photothermal conversion performance in colorectal cancer treatment.
Methods or Background: We constructed a novel and simple Fe single-atom coordinated N-doped porous carbon nanozyme (FeCA
SAzyme) at the atomic level for catalytic anti-tumour treatment coupled with photothermal therapy. The catalytic performance of
FeCA SAzyme was verified by cell experiment and animal experiment.
Results or Findings: The FeCA SAzymes exhibited a superior catalytic kinetic Km value (4.62 mM) compared with natural catalase
and other reported nanozymes. In addition, the dual-mode strategy of chemodynamic therapy in combination with photothermal
therapy further increased the tumour treatment effect. Moreover, the FeCA SAzymes exhibited favourable performance in magnetic
resonance imaging. The r2 value of FeCA SAzymes was approximately 149.8 mM-1S-1, which highlighting their potential as negative-
contrast media for imaging purposes.
Conclusion: Our study shows that single-atom nanozymes can increase the Fenton reaction at the atomic level, which provides a
new research direction for exploring multi-modal biological applications in tumour therapy and diagnosis.
Limitations: No limitations identified for this study.
Ethics committee - additional information: Fudan University Ethics committee approved this study.
The value of DLSCT-based multi-parameters for assessment of VETC and MVI in hepatocellular carcinoma (7 min)
Anqi Li; Guangzhou / China
1027
Author Block: A. Li1, M. Luo1, M. Li1, H. You1, J. Zhou1, W. Deng2, X. Yu2, Z. Zhang2, J. Wang1; 1Guangzhou/CN, 2Shanghai/CN
Purpose: The objective of this study was to explore the diagnostic performance of multi-parameters derived from dual-layer spectral
detector CT (DLSCT) to preoperatively identify vessels encapsulating tumour clusters (VETC) and microvascular invasion (MVI) of
hepatocellular carcinoma (HCC).
Methods or Background: Patients with histopathology-confirmed HCC who preoperatively underwent dual-phase contrast-enhanced
DLSCT were enrolled between May 2020 and June 2022 in this retrospective single-center study. Multi-parameters, including effective
atomic number (Zeff), electron density (ED), iodine density (ID), and the slope of the spectral attenuation curve (λ) were obtained
from DLSCT in arterial-phase (AP) and portal-venous-phase (PVP). The parameters of Zeff, ID and ED were then normalised with three
different normalization approaches, N1=tumor-to-aorta ratio; N2=tumor-to-liver parenchyma; N3=[tumourPVP-tumourAP]-to-
tumourAP ratio. The diagnostic performance of multi-parameters was analyzed by the area under the receiver operating characteristic
curve (AUC), and was compared using the Delong test.
Results or Findings: There were 122 HCC patients (positive/negative VETC, n=57/65; positive/negative MVI, n=54/68). Compared
with ZeffAP (AUC:0.64) and IDAP (AUC:0.64), the performance of N1-ZeffAP (AUC:0.77) and N1-IDAP (AUC:0.70) for VETC diagnosis of
HCC had been significantly improved, respectively (both p<0.05), and N1-ZeffAP had very high sensitivity (98.2%). The AUCs of N1-
ZeffPVP (0.68) and N1-IDPVP (0.70), N2-ZeffPVP (0.71) and N2-IDPVP (0.71) for MVI diagnosis of HCC non-significantly increased
compared with ZeffPVP (0.65) and IDPVP (0.65) (all p>0.05). N1-EDPVP showed high sensitivity (88.2%) and N1-IDPVP showed high
specificity (88.9%) for identifying MVI. λAP was used to diagnose VETC with good sensitivity (84.2%), and λPVP was used to diagnose
MVI with good specificity (79.6%).
Conclusion: Multi-parameters derived from dual-phase contrast-enhanced DLSCT could be the promising biomarkers for
noninvasively identifying VETC and MVI of HCC, and the tumour-to-aorta ratio is recommended for the normalization of parameters to
improve the diagnostic performance.
Funding for this study: This work was supported by the National Natural Science Foundation of China grant (82271973, JW;
91959118, Jin Wang), Guangdong Basic and Applied Research Foundation (2021A1515010582, Jin Wang), Key Research and
Development Program of Guangdong Province (2019B020235002, Jin Wang), China International Medical Foundation SKY Research
Fund for Medical Imaging (Z-2014-07-2101 and Z-2014-07-1912-15, Jin Wang) and Clinical Research Foundation of the 3rd Affiliated
Hospital of Sun Yat-Sen University (YHJH201901, Jin Wang)
Node reporting and data system 1.0 (Node-RADS): validation on abdominal lymph nodes (7 min)
Elena Grimaldi; Milan / Italy
Author Block: E. Grimaldi1, F. Rigiroli2, O. Hamam2, B. Aslan2, A. Brook2, B. Siewert2, S. Nougaret2, O. R. Brook2; 1Milan/IT, 2Boston,
MA/US
Purpose: The aim of the study is to validate the Node Reporting and Data System 1.0 (Node-RADS) by assigning biopsied abdominal
lymph nodes to one of the Node-RADS categories and calculating malignancy rates for each.
Methods or Background: This IRB-approved, HIPAA-compliant study included 505 consecutive patients who underwent CT-guided
biopsy of abdominal lymph nodes between May 2016 and July 2023 at a single tertiary institution. A third-year radiology resident,
blinded to pathology results, reviewed the lymph node’s size and configuration on a pre-procedure contrast-enhanced CT or MRI study
acquired within 30 days prior to biopsy. This resulted in a malignancy suspicion score summarized in five Node-RADS categories: “1-
very low”; “2-low”; “3-equivocal”; “4-high”; “5-very high”. This score was then compared to the pathology result. Clinical or imaging
follow-up was used as the reference standard for inconclusive pathology results.
Results or Findings: Five hundred and five consecutive CT-guided abdominal lymph node biopsies were included (median age 66
years, IQR 58-75 years), of which 326/505 (65%) were retroperitoneal, 123/505 (24%) pelvic and 56/505 (11%) mesenteric lymph
nodes. 10/505 (2%) lymph nodes were classified as Node-RADS 2, 61/505 (12%) Node-RADS 3, 66/505 (13%) Node-RADS 4 and
367/505 (73%) Node-RADS 5. Pathology was the reference standard in 476/505 (94%) patients, imaging follow-up in 25/505 (5%), and
clinical follow-up in 4/505 (1%). Malignancy rates were 1/10 (10%, 95% CI: 0%-31%) of Node-RADS 2, 26/61 (43%, 95%CI: 30%-55%)
of Node-RADS 3, 53/66 (80%, 95%CI: 71%-90%) of Node-RADS 4, and 332/367 (90%, 95% CI: 87%-93%) of Node-RADS 5.
Conclusion: Node-RADS categories were validated in the largest cohort of abdominal lymphadenopathies with pathology. This
system can provide a numeric risk score for malignant involvement of abdominal lymph nodes.
Limitations: The limitation is that the inter-reader agreement is still under evaluation.
Ethics committee - additional information: The study was approved by IRB (reference number: 2018P000099).
Evaluation of sarcopenia as predicting factor of perioperative chemotherapy toxicity and response to treatment in
patients with locally advanced gastric cancer (7 min)
Stefano Nardacci; Rome / Italy
1028
Author Block: S. Nardacci, M. Zerunian, G. Arrivi, C. Santangeli, F. Mazzuca, D. De Santis, D. Caruso, A. Laghi; Rome/IT
Purpose: In the management of patients with locally advanced gastric cancer, treatment is gastrectomy with perioperative
chemotherapy (p-ChT). Not all patients tolerate chemotherapy, therefore it would be useful to identify a non-invasive biomarker to
predict the outcome of p-CHT. Thus, we aimed to test the role of sarcopenia in predicting p-ChT toxicity and treatment response in
patients with gastric cancer.
Methods or Background: Patients with advanced gastric cancer who underwent contrast-enhanced CT both before and after the p-
ChT and had images available were retrospectively enrolled. Twenty-nine patients were enrolled, six excluded for CT unavailability.
Skeletal muscle mass (SMM) was assessed by manually segmented specific abdominal body region on unenhanced CT at L3 lumbar
vertebra level with a dedicated software (ImageJ). Skeletal muscle index was obtained by normalizing SMM for patient height in
metres squared. SMI<52.4 and 38.5 cm2/m2 was considered as sarcopenic for male and female respectively. Data on haematological,
gastrointestinal and neurological toxicity and response to treatment were recorded. Rank correlation and receiver operating
characteristics (ROC) curve were obtained to assess sarcopenia status performance for p-Cht toxicity and response to treatment
prediction; P<0.05 considered significant.
Results or Findings: Among the toxicities of p-Cht, sarcopenia showed significant ability to predict both haematologic and
neurologic toxicity (AUC=0.929 and 0.972 respectively, all P<0.001). In addition, sarcopenia was shown to be a significant predictor
of postoperative complications in the sarcopenic population (p=0.033). Gastrointestinal toxicity and response therapy did not show
significant results (P>0.05).
Conclusion: Evaluation of sarcopenia might represent an important non-invasive imaging biomarker to predict haematologic and
neurologic toxicity in patients with gastric cancer before the beginning of chemotherapy to allow a personalised treatment
management.
Limitations: The limitations of the study are the poor population and the manual segmentation method.
Ethics committee - additional information: The study is retrospective hence, no ethical approval was required.
1029
EIBIR 24 - Artificial intelligence (AI), real impact: redefining clinical decision making based on medical
images
Categories: Artificial Intelligence & Machine Learning, Breast, Oncologic Imaging, Research, Vascular
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Oliver Diaz; Barcelona / Spain

RadioVal: international clinical validation of radiomics AI for breast cancer treatment planning (15 min)
1. To learn about the FUTURE-AI guideline for trustworthy AI in healthcare.

2. To appreciate its advantages in guiding AI teams for building deployable AI tools in healthcare.
3. To understand its potential for building and validating trustworthy AI tools in breast cancer imaging.
ODELIA: open-source swarm learning for decentralised medical AI for breast cancer detection (15 min)
Daniel Truhn; Aachen / Germany
1. To learn about collaborative training of AI algorithms.

2. To get up to date about the current progress of AI in MR breast imaging.
3. To learn about possibilities to contribute to collaborative AI development.
AI-POD: trustworthy AI tools for the prediction of obesity-related vascular diseases (15 min)
Georg Langs; Vienna / Austria
1. To learn about the potential of AI tools in obesity-related vascular disease.

2. To show how to build and validate AI tools in the prediction of obesity-related vascular disease.
1030
RPS 2405 - AI in breast cancer screening
Categories: Artificial Intelligence & Machine Learning, Breast, Oncologic Imaging, Physics in Medical Imaging
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Susan Mary Astley; Altrincham / United Kingdom
Bimodal CADx assessment in mammography and tomosynthesis: a standalone study for breast screening (7 min)
Hubert Beaumont; Nice / France
Author Block: H. Beaumont, A. Iannessi, S. Pacilè, T. Louis; Nice/FR

Purpose: Digital mammography (FFDM) is the standard for breast cancer screening. Digital breast tomosynthesis (DBT), compared to
FFDM, enhances cancer detection and reduces unnecessary biopsies. Despite DBT's adoption, critical questions remain—higher
radiation, time, cost, and clinical benefits, particularly for systematic breast screening. In the era of AI CADx (Computer-Aided
Diagnosis) for breast screening, one unresolved question is the role of bimodal algorithms in predicting cancer risk and offering
guidance when opinions differ, and we aim to understand this.
Methods or Background: We analysed a cohort of 1071 screened patients for breast cancer who underwent both mammography
and tomography. A CADx software assigned a score of malignancy to each tumour, allowing to compute the joint distribution of the
paired mammography/tomography scorings. From the joint distribution, we defined areas of “perpendicular diagnosis” (PD) as the
areas of highly discordant scoring. We evaluated the potential of systematic reclassifications of perpendicular scoring through
sensitivity and specificity both for tumoural mass and calcifications.
Results or Findings: We observed a modest inter-modality agreement, indicated by a kappa of 0.19 (95% CI: 0.18; 0.22). PD scoring
was present in 32.7% (95% CI: 29.7; 35.8) of mass cases and 38.6% (95% CI: 30.1; 47.6) of calcification cases. Specific
reclassification rules significantly increased mass sensitivity from 0.74 (95% CI: 0.71; 0.77) to 0.80 (95% CI: 0.77; 0.83), resolving
29.1% of PD cases. For calcifications, reclassification addressed 12% of discrepancies, improving specificity from 0.85 (95% CI: 0.76;
0.91) to 0.86 (95% CI: 0.78; 0.92).
Conclusion: Our analysis revealed a significant proportion of CADx algorithmic discrepancies. In cases where FFDM classified masses
as benign while DBT CADx suggested malignancy, DBT CADx aided decision-making. Conversely, for calcifications initially assessed as
negative with tomosynthesis, FFDM CADx proved valuable. Exploring alternative reclassification methods is essential.
Limitations: No limitations were applicable for this study.
Are AI-detected interval cancers actionable for recall in a real screening setting? An informed review of 120 interval
cancer cases with high AI scores in breast screen Norway (7 min)
Henrik Wethe Koch; Stavanger / Norway
1031
Author Block: H. W. Koch1, M. Larsen2, S. Hofvind2; 1Stavanger/NO, 2Oslo/NO
Purpose: Retrospective studies have suggested that using artificial intelligence (AI) systems in breast cancer screening might help
us detect 30-40% of interval cancers. However, it is uncertain whether the AI-markings match the location of the tumour on diagnostic
mammograms, and if the findings are actionable for recall in a real screening setting, which, is the aim of this study.
Methods or Background: In 2022, we conducted a retrospective study comparing the performance of an AI-system with
independent double reading by radiologists according to cancer detection. The AI-system (Transpara v.1.7.0) scored mammograms
from 1-10 based on risk of malignancy. 42% (120/289) of the interval cancers had an AI-score of 10. In this study, four radiologists did
a consensus review of the interval cancers with AI-score 10 and compared AI-markings with cancer location on diagnostic
mammograms. Interval cancers were classified as false negative, minimal sign (actionable or non-actionable) or true negative.
Mammographic breast density was classified as BI-RADS a-d.
Results or Findings: Of 120 interval cancers with AI-score 10 (group1), 77.5% (93/120) had AI-markings matching the cancer
location (group2). 20.8% (25/120) had AI-markings matching cancer location and were considered actionable for recall (false
negative/minimal sign actionable) (group3). Density distribution as percentage of all 289 interval cancers:
Group1: a: 17% (1/6), b: 42% (46/110), c: 41% (56/138), d: 49% (17/35), Group2: a: 17% (1/6), b: 33% (36/110), c: 38% (53/138), d:
9% (3/35), Group3: a: 0% (0/6), b: 10% (11/110), c: 10% (14/138), d: 0% (0/35).
Conclusion: Our results indicate that the true effect of AI in screen reading regarding earlier detection of interval cancers is still
uncertain. Although 49% of interval cancers in extremely dense breasts had AI-score 10, none were considered actionable for recall in
an informed consensus review.
Limitations: Retrospective study design and informed consensus review was the limitation of this study.
Ethics committee - additional information: The study was approved by the Regional Committee for Medical and Health Research
Ethics (#13294).
Integration of artificial intelligence (AI) in double-read population-based mammography screening: simulated

replacement of one reader and beyond (7 min)
Mohammad Talal Elhakim; Odense / Denmark
Author Block: M. T. Elhakim1, S. Wordenskjold Stougaard1, O. Graumann2, M. Nielsen3, O. Gerke1, L. B. Larsen1, B. Schnack Brandt
1 1 2 3
Rasmussen ; Odense/DK, Aarhus/DK, Copenhagen/DK
Purpose: The aim of this study was to compare the accuracy and feasibility of three AI-integrated screening scenarios compared to
double reading with arbitration (combined reading).
Methods or Background: A study sample of 249,402 consecutive screening mammograms representative of an entire screening
population was obtained from the Region of Southern Denmark. The AI system Lunit INSIGHT MMG v.1.1.7.1 (Lunit Inc.) processed all
mammograms. In Scenario 1, AI replaced first reader. In Scenario 2, AI replaced second reader when it agreed with the decision of the
original first reader. In Scenario 3, AI was applied as a standalone triage tool, with AI replacing both readers for assessing low- and
high-risk screenings, while moderate-risk screenings were assessed by the original combined reading. AI cut-offs were chosen partly
based on a previous validated threshold and partly based on maintaining a workload reduction at around 50% for comparability
across scenarios. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), recall rate (RR), and
arbitration rate (AR) were calculated.
Results or Findings: AI cut-off scores were 80.99% (Scenarios 1 and 2), and <3.36% and ≥95.29% for low-risk and high-risk
screenings, respectively (Scenario 3). Compared to combined reading, AI-integrated screening showed no statistically significant
difference in any outcome measures other than a higher AR by +1.0% (p<0.0001) in Scenario 1, and a higher sensitivity by +1.5%
(p=0.001) and lower AR by -0.8% (p<0.0001) in Scenario 3. In Scenario 2, AI-integrated screening had statistically significantly lower
sensitivity (-4.5%; p<0.0001), NPV (-0.1; p=0.001), RR (-0.6%; p<0.001), and AR (-1.5; p<0.0001), and higher specificity (+0.6%;
p<0.0001) and PPV (+4.7; p<0.0001).
Conclusion: Partial or full replacement of one or both readers in double reading with AI seems feasible without markedly affecting
accuracy in screening.
Limitations: Retrospective design and correlated radiologist readings with reference standard were the limitations of this study.
Funding for this study: Region of Southern Denmark funded this study.
Ethics committee - additional information: This study was approved by the Danish National Committee on Health Research
Ethics (identifier D1763009).
RSNA 2023 screening mammography breast cancer detection AI challenge results (7 min)
George Partridge; Nottingham / United Kingdom
1032
Author Block: G. Partridge1, M. Vazirabad2, R. Ball3, H. Trivedi4, F. C. Kitamura5, H. Frazer6, R. M. Mann7, L. Moy8, Y. Chen1;
1
Nottingham/UK, 2Chicago, IL/US, 3Bar Harbor, ME/US, 4Atlanta, GA/US, 5São Paulo/BR, 6South Yarra/AU, 7Nijmegen/NL, 8New York,
NY/US
Purpose: Artificial intelligence (AI), used alongside human readers, in breast cancer screening could revolutionise the screening
workflow. The RSNA hosted the 2023 screening mammography breast cancer detection AI challenge, where participants were invited
to develop AI algorithms to interpret mammograms. The purpose was to assess the performance of submitted algorithms and explore
the potential for improving performance by combining high-performing algorithms.
Methods or Background: Teams were provided a training-set of 11,913 2-view 2D digital mammogram (2DDM) screening cases,
from two institutions (US and Australia) for AI training. AI performance was evaluated using an independent test-set of 5,415 2DDM
cases from the same source. Cancer cases were pathology proven and non-cancer cases had at least 1-year of normal follow-up.
Algorithms were ranked in the challenge using the pF1 accuracy score (incorporating sensitivity and PPV). In the current study, all
algorithms were assessed independently. In addition, combined models were constructed from top-ranked algorithms.
Results or Findings: One thousand six hundred and eighty seven teams participated in the challenge, each submitting their own
algorithm. Median specificity and NPV were high across algorithms (98.7% and 98.5%, respectively), yet median cancer detection was
low (sensitivity: 27.6%, PPV: 36.9%), with a median recall rate (RR) of 1.7%. The highest ranked algorithm (as per pF1) had a RR of
1.5%, specificity of 99.5%, NPV of 99.0%, sensitivity of 48.6%, and a PPV of 64.6%. Combining the top 3 and top 10 ranked algorithms
demonstrated an increased RR (2.4% and 3.5%), while achieving a marked improvement in sensitivity (60.7% and 67.8%), while
specificity remained more stable (98.7% and 97.8%, respectively).
Conclusion: Variation in performance of submitted AI algorithms to the RSNA challenge is substantial. Combining the highest-
performing algorithms demonstrated improvement in performance.
Limitations: Relatively small size of evaluation test-set, low cancer prevalence (but screening setting) limit this study.
Improving breast cancer recurrence forecasts: combining multi-time-point mammography and medical reports (7 min)
Chunyao Lu; Amsterdam / Netherlands
1 1 1 1 1 1 2 1 2
Author Block: C. Lu , X. Wang , L. Han , T. Zhang , Y. Gao , T. Tan , R. M. Mann ; Amsterdam/NL, Nijmegen/NL
Purpose: Predicting the risk of recurrence post-breast cancer surgery continues to be a challenging task, despite having access to
complete medical records. The purpose of this study was to develop a deep learning model based on multi-time-point mammogram
breast images and medical reports to predict the risk of postoperative recurrence of breast cancer.
Methods or Background: At a large academic medical center, we collected consecutive digital screening mammograms and
medical reports in 3188 patients between January 1, 2000, and December 31, 2020. Our model synergistically integrates risk factors
derived from multi-time-point mammograms with patients’ preoperative clinical data. We compared our method with commonly used
machine learning methods based on clinical data and image-based deep learning models.
Results or Findings: We conducted a comprehensive comparison between our model and common machine learning models as well
as deep learning methods, demonstrating that our model attained the highest AUCs in three datasets of patients who relapsed at
different times, with scores of 0.72, 0.76, and 0.83 within 5, 10, and 20 years respectively. We discovered that while traditional risk
factors are significant contributors, our model enhances the accuracy of predicting cancer recurrence risk by deducing potential risk
factors from multi-time-point mammography images.
Conclusion: Our model underscores the advantages of incorporating complete and consecutive medical data into predictive
algorithms, enhancing accuracy in forecasting recurrence and informing health policies for post-surgical treatment of breast cancer
patients.
Limitations: This study needs to be validated in more external datasets.
Funding for this study: This study is supported by the Chinese Scholarship Council Studentship and the Guangzhou Elite Program
(TZ-JY201948).
Microwave breast cancer screening and early detection: SAFE clinical study (7 min)
Aleksandar Janjic; Istanbul / Turkey
1033
Author Block: A. Janjic, I. Akduman, M. Cayoren, A. Yurtseven, O. Bugdayci, M. E. Aribal; Istanbul/TR
Purpose: The SAFE (Scan and find early) system is a microwave breast cancer imaging (MBI) device designed for non-invasive and
non-ionising breast cancer screening and early detection. This technology relies on distinguishing dielectric properties between
cancerous and healthy tissue to provide valuable clinical insights. This study aims to evaluate SAFE's capability to precisely identify
lesions within the patient's breast.
Methods or Background: This study exclusively enrolled patients scheduled for biopsy, following approval from the ethics
committee of Marmara University School of Medicine. The approach utilised to identify breast lesions was based on analysing the
difference in backscattered signals between healthy and cancerous breast tissue. Furthermore, we employed a machine learning
approach, specifically utilising extreme gradient boosting (XGBOOST), to discern the presence of cancerous tissue within the breast.
In tandem, a qualitative microwave imaging method was employed to precisely pinpoint the location of the tumour.
Results or Findings: Our dataset comprised 394 samples, with 284 originating from healthy tissue and 110 from cancerous tissue.
Among the 110 cancerous cases, 69 were identified as benign and 41 as malignant findings. The devised detection model exhibited
commendable performance, with a sensitivity, specificity, and accuracy of 91%, 92%, and 92%, respectively.
Conclusion: The findings from our study demonstrate the capability of our MBI system in detecting a significant majority of breast
lesions. This suggests that SAFE holds promise in positively influencing breast cancer screening and early detection, given its non-
invasive and safe characteristics. We are in the process of planning further clinical studies to validate the results obtained.
Limitations: The study's limitations encompass a modest sample size, potential selection bias from exclusive enrollment, reliance on
a single machine learning algorithm, and the suggestion for external validation to fortify the findings.
Funding for this study: This research was funded by the Scientific and Technology Research Council of Turkey (TUBITAK) grant
number 120N388 and by the European Union’s Horizon 2020 research and in-novation program under the Marie Sklodowska Curie
grant agreement No. 764479.
Ethics committee - additional information: The study was approved by the Ethics committee of Marmara University School of
Medicine (Protocol number: 70737436-050.06.04; Date of approval: Jun. 09, 2014). All protocols and procedures were in accordance
with both institutional and national ethical standards in research and with the World Medical Association Declaration of Helsinki.
Clinical and operational benefits of artificial intelligence (AI) in prospective UK breast screening service evaluation (7
min)
Gerald Lip; Aberdeen / United Kingdom
Author Block: G. Lip1, A. Ng2, C. F. De Vries1, L. A. Anderson1, R. Staff1, G. Fox2, C. Oberije2, P. Kecskemethy2; 1Aberdeen/UK,
2
London/UK
Purpose: The objective of this study was to assess the clinical and operational benefits of AI workflows in a live breast screening
service.
Methods or Background: An AI system has been evaluated in a prospective paired design at NHS Grampian to assess the impact of
using AI in a combination of workflows that use AI as an: (1) Independent reader when it agrees on ‘no recall’ with Reader 1 (i.e.
Double Reader Triage (DRT) workflow) or when it agrees on ‘recall’/’no recall’ with Reader 1 (i.e. Supporting Independent Reader
(sIR)) to provide workload savings, and (2) Extra Reader (XR) to triage positives for additional arbitration (not recalled by standard
double reading (DR)) to provide an opportunity for increased cancer detection. Over 10,000 non-opt-out women who had a four-view
FFDM processed by the AI were included. All screens were human double-read, maintaining the standard of care. DRT and sIR
performance outcomes were simulated, while XR outcomes were measured from live use. Planned analyses included assessment of
recall rate, cancer detection rate, arbitration rate, positive predictive value, sensitivity, specificity, and workload savings for DR, DRT,
sIR, and XR and combination workflows DRT+XR and sIR+XR, as well as non-inferiority and superiority tests for the combination
workflows against DR.
Results or Findings: Interim analyses have identified that cancers from at least six women have been found through the XR
process, resulting in a relative 13% (6/47) increase in cancer detection (0.9/1000 absolute) compared to DR. The sIR and DRT
workflows are expected to provide significant workload savings that offset the additional arbitration workload that XR requires by 6x.
Conclusion: Interim analyses indicate clinical and operational benefits when using AI in breast cancer screening, which will be
confirmed and presented in the final evaluation analysis (expected Q1 2024).
Limitations: Single-site evaluation limits this study.
Funding for this study: NHSE/AAC/NIHR AI in Health and Care Award funded this study.
Ethics committee - additional information: Ethics review not required for service evaluations
Adding artificial intelligence (AI) case scoring in a breast screening programme to overcame delay in most probably
true positive cases: a retrospective study (7 min)
Andrea Nitrosi; Reggio Emilia / Italy
1034
Author Block: A. Nitrosi, R. Vacondio, C. Coriani, L. Verzellesi, N. Cucurachi, C. Campari, M. Bertolini, P. Pattacini, M. Iori; Reggio
Emilia/IT
Purpose: The objective of this study was to retrospectively evaluate an AI case score based strategy to anticipate the readings of
most probably true positive cases to keep reading times within two weeks as would be required by local regulations.
Methods or Background: We analyzed 32,012 2D mammography screening exams including 71 proven tumours and 61 pending
diagnosis, consequentially acquired in Reggio Emilia Breast Screening Program (BSP) starting from October 2022 to July 2023 and
elaborated by iCAD Inc. ProFound AI 2D system. ICAD Case Scores represent the AI algorithm’s relative confidence that a case is
malignant on a scale of 0% to 100%. A pool of nine radiologists performs double blinded plus arbitration readings; each reads about
200 mammograms per 6 hours work-shift, exams are evaluated in “virtual sessions” defined by mammograph unit and exam date.
Due to a known chronic shortage of medical personnel, the readings can exceed local regulations standards of 15 days (even to more
than 30). An AI-based prioritised reading protocol was elaborated analyzing the tumour incidence in function of Case Score.
Results or Findings: Among cases respectively having a Case Score evaluated greater than 30%, 40%, 50% and 60% the
percentage of total tumours found were 89%, 85%, 69% and 61%, while the percentage of cases to be read were 20%, 13.8%, 9.8%,
5.4%. It’s worth noting that prioritizing the readings of “case score based virtual session” comprehending only exams with case score
>40% (4406 over 32012) per week (an average of 174 exams per week), the majority of the true positives (85%) women could be
recalled within very short time.
Conclusion: This scenario would not undermine the reading screening workflow while guarantee early diagnosis and hopefully nor
influence readers competence.
Limitations: Cases refer only to Reggio Emilia BSP limiting this study.
Funding for this study: This study was partially supported by the Italian Ministry of Health—Ricerca Corrente.
Ethics committee - additional information: Compliance with Ethical Standards Institutional Review Board approval was not
required because it is a Clinical Audit about a technical development. This study was conducted in accordance with the routine quality
assurance procedures established by the Local Health Authority for its screening programmes. The Reggio Emilia Cancer Registry,
which routinely collects the screening history of each case of breast cancer, has been approved by the Provincial Ethic Committee.
1035
RC 2401 - Non-IBD intestinal inflammation: the role of imaging
Categories: GI Tract
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:

Infectious enteritis and colitis (15 min)

Jordi Rimola Gibert; Barcelona / Spain
1. To identify imaging findings of specific gastrointestinal infectious diseases.

2. To diagnose (less prevalent? or another better term?) gastrointestinal manifestations of infectious diseases.
3. To differentiate infectious from inflammatory bowel disease.
Intestinal graft versus host disease (15 min)

Francesca Maccioni; Rome / Italy
1. To list the main radiological features of intestinal acute GVHD.

2. To list the main intestinal and abdominal MRI biomarkers associated with the disease.
3. To reflect on clinical correlations and MRI scoring systems for diagnostic and prognostic purposes.
Drug-induced enteritis and colitis (15 min)

Ingrid Millet; Montpellier / France
1. To list the main drugs at risk for intestinal toxicity.

2. To know how long it takes for symptoms to appear.
3. To identify imaging signs suggestive of drug-induced digestive damage.
Panel discussion: Is GI imaging a problem-solving tool in the multidisciplina discussion of non-IBD intestinal
inflammation? (10 min)
1036
E³ 25E - Topics in aortic imaging
ETC Level: LEVEL I
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:

Atherosclerotic disease of the aorta (15 min)

1. To discuss indications for cross-sectional aortic imaging in atherosclerotic disease.

2. To discuss the role of CTA and MRA in the diagnostic workup of aortic atherosclerotic disease.
3. To discuss the elements of the radiology report essential in diagnostic workup.
Genetic aortic syndromes (15 min)

Julius Matthias Weinrich; Hamburg / Germany
1. To discuss the most common genetic aortic syndromes.

2. To discuss the role of CTA and MRA in the genetic aortic syndromes.
3. To discuss the elements of the radiology report essential in patients with suspected or known genetic aortic syndromes.
Vasculitis (15 min)

1. To discuss the most common vasculitides involving the aorta and aortic branch arteries.
2. To discuss the role of cross-sectional imaging (CTA, MRA and nuclear medicine) in the workup of suspected or known vasculitis.
3. To discuss the elements of the radiology report essential in patients with suspected or known vasculitis.
1037
RC 2412 - The paediatric liver
Categories: Abdominal Viscera, Paediatric

ETC Level: LEVEL II
Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:

Imaging of the paediatric liver (12 min)

Seema Toso; Genève 14 / Switzerland
1. To learn about diagnostic imaging workup of the paediatric liver: indications, applications, obstacles and problem-solving.
Liver transplant (12 min)

Martijn V Verhagen; Groningen / Netherlands
1. To name the relevant vascular focus points when assessing a paediatric liver transplantation
2. To be aware of the types of vascular complications and how to assess these with imaging.
3. To understand the significance of complications and advice on the appropriate treatments.
Fontan-associated liver disease (FALD) (12 min)

Charlotte de Lange; Gothenburg / Sweden
1. To learn about the basic pathophysiology of FALD.

2. To describe the use of different imaging techniques in the assessment of FALD.
3. To discuss the role of imaging in the current surveillance strategies of FALD.
Quantification of fat and iron in the liver (12 min)

Greg Chambers; Leeds / United Kingdom
1. To recognise patterns of fat deposition on USS.

2. To appreciate why fat and iron quantification is clinically important
3. To understand the main MRI sequences used in detecting fat and iron.
Panel discussion: Where does ultrasound end and MRI begin? (10 min)
1038
RC 2410 - Ultrasound in musculoskeletal radiology

Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:

High-frequency and ultra-high frequency ultrasound (15 min)

1. To discuss the current advances in state-of-the-art high-frequency and ultra-high-frequency ultrasound.

2. To discuss the main indications and clinical applications.
3. To demonstrate the ultrasound semiology of pathologic conditions.
MSK elastography (15 min)

1. To explain the techniques used in ultrasound elastography of the musculoskeletal system.

2. To describe ultrasound elastography findings in musculoskeletal radiology.
Interventional MSK ultrasound (15 min)

Elena E. Drakonaki; Heraklion / Greece
1. To discuss the indications for ultrasound-guided treatment and intervention.

2. To describe the prerequisites and technique.
3. To discuss the advantages and disadvantages of ultrasound-guided intervention compared to other imaging-guided modalities.
Panel discussion: New clinical roles for musculoskeletal ultrasound (10 min)
1039
BS 24b - Formal and informal advanced practice roles for radiographers
Categories: Management/Leadership, Multidisciplinary, Professional Issues, Radiographers, Students

Date: March 3, 2024 | 11:30 - 12:30 CET
CME Credits: 1
Moderator:
Safora Johansen; Oslo / Norway

Safora Johansen; Oslo / Norway
Established and novel advanced practice roles for radiographers (18 min)
Claire Senior; Exeter / United Kingdom
1. To define the meaning and components of advanced clinical practice.

2. To summarise the range of advanced practice roles in radiography.
3. To consider how to apply the concepts of advanced clinical radiography practice to their career pathway.
Exploring opportunities and informal roles to advance radiographers' practice (18 min)
Patrick Vorlet; Lausanne / Switzerland
1. To define the strategy for detecting opportunities.

2. To describe two kinds of opportunities and informal roles.
3. To describe the strategy for moving from informal to formal.
Advance practice roles in radiotherapy (18 min)

Celeste Marques Oliveira; Porto / Portugal
1. To identify advanced practice (AP) roles and advanced practitioners in radiotherapy.

2. To distinguish between formal vs informal AP amongst therapeutic radiographers' practice.
3. To describe the four pillars of AP and its challenges in radiotherapy at the European level.
1040

ECR 2024 Programme

Uploaded by

Copyright:

Available Formats

ECR 2024 Programme

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ECR 2024 Programme

Uploaded by

Copyright:

Available Formats

AWA K E N I N G

FOREWORD BY THE ESR PRESIDENT

Our theme for this year, 'Next Generation Radiology,' reflects

In this congress, we will delve deep into the frontiers of

Of course, the ECR is not just a meeting for radiologists; it's a

FOREWORD BY THE EFRS PRESIDENT

Clinical Trials in Radiology Sessions E³ - The Beauty of Basic Knowledge Sessions

ESR Meets Sessions Joint Sessions

Amidst the dynamic landscape of medical imaging, New Horizons Sessions

Refresher Courses State of the Art Symposia

Preliminary programme printed on February 16, 2024

Date: January 30, 2024 | 08:56 - 19:00

ST 1 - Morning Welcome with Carlo Catalano

Categories: Education, General Radiology, Multidisciplinary, Professional Issues

Interview (30 min)

E³ 24A - Hybrid imaging in oncology (part 1): where is it standard-of-care?

Categories: Hybrid Imaging, Nuclear Medicine, Oncologic Imaging

Chairperson's introduction (5 min)

Hybrid imaging in haematological malignancies (15 min)

1. To understand the role of imaging in haematological malignancies therapy guidance.

PET/CT for lung cancer therapy management (15 min)

Panel discussion: Hybrid imaging as standard-of-care: accessibility guaranteed? (10 min)

RPS 101 - Pre- and post-treatment imaging in cirrhosis and HCC

Categories: Abdominal Viscera, Imaging Methods, Oncologic Imaging

Author Block: S. Kim, N. K. Lee, S-B. Hong, J. Han; Busan/KR

Author Block: R. Huang; Guangzhou/CN

Author Block: Y. H. Ji, S-B. Hong, N. K. Lee, S. Kim; Busan/KR

RPS 102 - Breast imaging biomarkers

Categories: Artiﬁcial Intelligence & Machine Learning, Breast, Oncologic Imaging

Prediction of cancer aggressiveness based on breast MRI features (7 min)

RPS 105 - Automatic segmentation techniques

Multi-organ CT-based automatic segmentation via semi-supervised learning (7 min)

BS 1 - Pearls and pitfalls in paediatric and foetal imaging

Categories: Imaging Methods, Musculoskeletal, Neuro, Paediatric

Chairperson's introduction (5 min)

Pearls and pitfalls of imaging paediatric skeletal dysplasias (15 min)

1. To become familiar with age-related patterns of traumatic brain injury.

Pearls and pitfalls of imaging the foetus (15 min)

Questions and answers (10 min)

HW 1Pa - Peripheral zone: how to detect typical and not-so-typical tumours

Categories: Genitourinary, Oncologic Imaging

In this session participants will be able:

Instructors (60 min)

E³ 22A - Hip and pelvis

Categories: Education, General Radiology, Musculoskeletal

Chairperson's introduction (2 min)

Athletic pubalgia and extraarticular soft tissue injuries (25 min)

1. To list the possible causes of athletic pubalgia and groin-related pain.

Panel discussion (8 min)

RC 111 - Post-neurosurgery: what to expect on imaging

Chairperson's introduction (5 min)

1. To name three routes of CSF leakage post-operatively.

Post-neurosurgery artifacts on MRI imaging (15 min)

CSF shunts: post-operative evaluation and complications (15 min)

Panel discussion: How to improve post-operative reporting in neuroradiology (10 min)

RC 116 - Normal variants and mimics in oncologic imaging

Categories: Chest, Genitourinary, GI Tract, Oncologic Imaging

Chairperson's introduction (5 min)

Chest imaging (15 min)

1. To understand how benign pathology may mimic cancer.