Open Access Review

Article DOI: 10.7759/cureus.49560

The Utility of Artificial Intelligence in the

Diagnosis and Management of Pancreatic Cancer
Review began 11/15/2023
Vikash Kumar 1 , Mrunanjali Gaddam 1 , Amr Moustafa 1 , Rabia Iqbal 1 , Dhir Gala 2 , Mili Shah 2 , Vijay
Review ended 11/19/2023 Reddy Gayam 3 , Praneeth Bandaru 3 , Madhavi Reddy 3 , Vinaya Gadaputi 4
Published 11/28/2023

© Copyright 2023 1. Internal Medicine, The Brooklyn Hospital Center, Brooklyn, USA 2. Internal Medicine, American University of the
Kumar et al. This is an open access article Caribbean School of Medicine, Sint Maarten, SXM 3. Gastroenterology and Hepatology, The Brooklyn Hospital Center,
distributed under the terms of the Creative Brooklyn, USA 4. Gastroenterology and Hepatology, Blanchard Valley Health System, Findlay, USA
Commons Attribution License CC-BY 4.0.,
which permits unrestricted use, distribution,
and reproduction in any medium, provided
Corresponding author: Vikash Kumar, [email protected]
the original author and source are credited.

Abstract
Artificial intelligence (AI) has made significant advancements in the medical domain in recent years. AI, an
expansive field comprising Machine Learning (ML) and, within it, Deep Learning (DL), seeks to emulate the
intricate operations of the human brain. It examines vast amounts of data and plays a crucial role in
decision-making, overcoming limitations related to human evaluation. DL utilizes complex algorithms to
analyze data. ML and DL are subsets of AI that utilize hard statistical techniques that help machines
consistently improve at tasks with experience. Pancreatic cancer is more common in developed countries
and is one of the leading causes of cancer-related mortality worldwide. Managing pancreatic cancer remains
a challenge despite significant advancements in diagnosis and treatment. AI has secured an almost
ubiquitous presence in the field of oncological workup and management, especially in gastroenterology
malignancies. AI is particularly useful for various investigations of pancreatic carcinoma because it has
specific radiological features that enable diagnostic procedures without the requirement of a histological
study. However, interpreting and evaluating resulting images is not always simple since images vary as the
disease progresses. Secondly, a number of factors may impact prognosis and response to the treatment
process. Currently, AI models have been created for diagnosing, grading, staging, and predicting prognosis
and treatment response. This review presents the most up-to-date knowledge on the use of AI in the
diagnosis and treatment of pancreatic carcinoma.

Categories: Family/General Practice, Gastroenterology, Internal Medicine

Keywords: deep learning, screening, machine learning, artificial intelligence, pancreatic cancer

Introduction And Background

Pancreatic carcinoma (PC) is a lethal cancer of the pancreas. The most prevalent type of pancreatic cancer is
pancreatic ductal adenocarcinoma (PDAC). It should be considered that the majority of cases are PDAC when
the broad term “pancreatic cancer” is used. PDAC is categorized as one of the leading causes of cancer
mortality worldwide. Unfortunately, pancreatic cancer usually shows few or no symptoms until it has either
locally advanced or metastatically spread leading to poor survival rates and posing a significant diagnostic
challenge. Due to close proximity to the major blood vessels, the tumor can easily invade them [1], making
80-85% of tumors non-resectable at the time of diagnosis [2]. Therefore, in addition to addressing the risk
factors associated with it such as smoking, alcohol, chronic pancreatitis, diabetes, obesity, and Helicobacter
pylori infection, appropriate timely diagnosis and treatment also improve survival rates. According to a
multicenter study, individuals with PC identified through screening had a median survival time of 9.8 years
as opposed to 1.5 years for patients with PC identified with non-screening.

Generally, CT scan with contrast media enhancement and magnetic resonance imaging (MRI) are considered
for the diagnosis/staging of pancreatic cancer. However, Endoscopic Ultrasound (EUS) has a higher
sensitivity for identifying solid pancreatic lesions less than 2 cm when compared to CT and MRI [3]. A biopsy
is the gold standard for diagnosing pancreatic cancer. Even the best-validated PC biomarker, carbohydrate
antigen 19-9 (CA19-9), does not have enough accuracy and specificity in screening early PC, and a lack of
biomarker in early PC make it challenging to diagnose at early staging. As a result, numerous researchers are
still striving to develop new early screening methods for PCs. As the images vary with disease progression, it
is not always simple to interpret and evaluate the resulting images. There can also be interobserver
variability in reading radiographic images, which can be improved with the use of AI. Similarly, several
factors may have an influence on prognostic factors and treatment response.

Artificial intelligence (AI) is the computer simulation of the human intelligence process. The concept of AI
emerged in the 1950s, but real progress was made only a few years ago [4]. The utilization of AI is rapidly
expanding and is increasingly useful in understanding gastrointestinal diseases [5-7]. AI acquires a flexible
structure by employing mathematical formulas and fundamental statistical concepts. As it is trained with
real-world data in a particular domain, it can infer and reason.

How to cite this article

Kumar V, Gaddam M, Moustafa A, et al. (November 28, 2023) The Utility of Artificial Intelligence in the Diagnosis and Management of Pancreatic
Cancer. Cureus 15(11): e49560. DOI 10.7759/cureus.49560
 One widely used subset of AI is machine learning (ML). ML examines data to develop algorithms that can
recognize distinct behavior forms and confirm predictive models. ML focuses on developing models that
assist machines in making predictions or judgments without explicit programming. Various ML techniques,
for instance, support vector machines (SVM), artificial neural networks (ANNs), classification, and
regression trees, seem to be employed in various investigations in the medical discipline [4]. New algorithms
are seemingly introduced every day. Some show the potential to have a major impact on image
interpretation.

Another AI technique called Deep learning (DL) has become a promising tool for image analysis in medicine
[8-11]. DL is a part of the ML technique that uses deep multilayered neural network algorithms to make
predictions based on discrete inputs. The model uses an iterative process to take output predictions and
reanalyze the model to further improve accuracy, continually improving with each new data point. DL has
an edge over other conventional ML techniques as it uses “representation learning,” a technique by which
the AI model recognizes relationships within a data set that human operators cannot notice [11].

In summary, ML is a core component of AI, and to implement it, DL is used. ML and DL have been successful
at predicting the risk of gastric cancer [12].

Review
Artificial intelligence in the diagnosis of pancreatic cancer
As the medical field and knowledge continue to expand, AI has gained importance due to its efficiency in
processing and organizing data. The use of AI has been bringing new tools that have been helping in early
diagnosis, risk stratification, and improving outcomes in patients with PC [13].

With the advancement in the field of AI, many tools are available that help to measure biomarkers and
analyze complex features and tissue characteristics.

For the diagnosis of PC, imaging using CT, EUS, MRI, and positron emission tomography (PET) scan is
required, which requires an expert radiologist to interpret the image. Previously used modalities and
techniques are summarized in Table 1.

Modalities Advantage Disadvantage

EUS High resolution, can get tissue biopsy Invasive

CT scan Provides detailed imaging, easily available Can miss diagnosis sometimes

MRI Can detect small tumors that might be missed by other scans Expensive; not widely available

TABLE 1: Summary of the advantages and disadvantages of the diagnostic modalities used to
diagnose pancreatic cancer.
EUS: endoscopic ultrasound

Artificial intelligence in radiology

The use of AI in the diagnosis of cancer can help physicians save a lot of time by manually going over the
details of imaging in detail. One method to extract minute details from the detailed image that might have
been missed by the human eye is the use of radiomics. This method is used to extract information regarding
shape, characteristics, heterogeneity, and even tumor aggressiveness [13].

AI models are being made that can help analyze CT scan images in more detail and provide a more detailed
analysis of complex patterns and images [14]. AI techniques have also been used to predict the malignant
potential of intraductal mucinous papillary neoplasm (IPMN) using CT scan images [15]. In one of the
studies, a computational model integrating clinical data and imaging features was designed that was used to
predict the probability of lymph node metastasis in PDAC by integrating the clinical data and the images
extracted from CT scans [16].

Endoscopic ultrasound is also used to detect small tumors and lesions of the pancreas up to a range of 2-3
mm in size [17]. It can help to get detailed images of the pancreas and surrounding tissue. In one of the
studies carried out by Zu et al, a machine model was used to extract EUS images recorded for pancreatic
adenocarcinoma and chronic pancreatitis. The model then helped to detect features that were used to
differentiate between PC and chronic pancreatitis with a sensitivity of 94% [18].

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 2 of 7

 PET scan is also used in staging pancreatic cancer and monitoring response to treatment, as well as looking
for recurrent disease. Zhang used PET/CT scan imaging and developed a machine model that is used to
differentiate between autoimmune pancreatitis and pancreatic adenocarcinoma non-invasively [19].

Digital pathology is another method that can be used to analyze histological slides with accuracy and
precision that exceeds that of a human. This involves the utilization of scanners which are used to scan a
simple tissue or blood specimen and form an accurate 3D image of the entire tissue. In addition to
diagnosing biopsy samples, this technology can help us better understand the cellular structure [20].

Models have also been made that can detect tumor markers or protein-based markers that are present more
in cancer cells than in normal cells. Early work shows that AI models such as surface-enhanced laser
desorption/ionization (SELDI) have been used to detect six protein markers that are found in cancer cells
and not detected in normal pancreatic cells [21]. Building on the utilization of biomarkers, AI can also
analyze genomic data and find associations of specific genes to pancreatic cancer. Ko et al developed a
machine model, Genes Vector for Each Sample (GVES), that can be used to diagnose genes involved in the
prognosis of the disease with accuracy, irrespective of datasets and sample sizes [22].

AI is also used to predict the risk of the development of pancreatic cancer by extracting clinical data about
the patient from electronic health systems [23]. A few studies have been done that show that the use of
electronic health records (EHR) to extract patient data can screen people who are at increased risk of
developing pancreatic cancer before diagnosis [24]. In one article, logistic regression was used on EHR to
detect people at risk of cancer a decade before the diagnosis was made [24].

The use of AI and software can help physicians analyze images and tissue samples to detect pancreatic
cancer, predict grading and staging, and detect people who might develop PCs in the future.

Artificial intelligence use in the treatment of pancreatic cancer

The use of AI in the treatment of pancreatic cancer has the potential to improve patient outcomes and
accelerate the development of new treatments. By providing personalized treatment plans, identifying
targeted therapies, monitoring treatment response, optimizing radiation therapy, and matching patients to
clinical trials, AI can help physicians provide the best possible care for pancreatic cancer patients.

Personalized Treatment Plans

AI can analyze various types of data, including medical history, imaging results, and genetic data to create
personalized treatment plans [23-25]. AI can use ML algorithms to analyze CT scans, MRI scans, and other
imaging data to create a 3D model of the tumor and surrounding tissues. This can help physicians plan
intervention and radiation therapy with greater accuracy while minimizing the risk of complications. AI can
also use genetic data to identify specific mutations or molecular targets that are present in cancer cells but
not in normal cells [26,27]. This can help physicians develop personalized treatment plans that are tailored
to the unique characteristics of each patient's cancer.

Targeted Therapies

AI can be used to identify specific molecular targets that are essential for the growth and survival of
pancreatic cancer cells [28]. This information can be used to develop targeted therapies that specifically
inhibit these targets. ML algorithms can analyze genetic data to identify specific mutations that are present
in cancer cells but not in normal cells [29]. This can be used to develop drugs that specifically target those
mutations, which can be more effective and less toxic than traditional chemotherapy. AI can also be used to
predict which patients are most likely to respond to specific treatments, enabling physicians to personalize
treatment plans based on the patient's genetic profile.

Treatment Monitoring

AI can be used to monitor a patient's response to treatment and adjust the treatment plan accordingly [30].
For example, ML algorithms can analyze imaging data to track changes in tumor size over time [31]. This can
help physicians determine whether the treatment is working or if adjustments need to be made. AI can also
be used to monitor a patient's blood markers, such as CA 19-9, which can indicate the presence of pancreatic
cancer [32]. By monitoring treatment response, physicians can adjust treatment approaches to maximize the
chances of a successful outcome.

Radiation Therapy

AI can also be used to optimize radiation therapy for pancreatic cancer patients. This can include creating a
personalized radiation treatment plan based on the patient's medical history, imaging results, and other
data [33]. ML algorithms can analyze imaging data to identify areas of the tumor that require higher doses of
radiation. This can help physicians deliver more effective radiation therapy while minimizing the risk of

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 3 of 7

 damage to healthy tissue. AI can also be used to monitor the patient's response to radiation therapy and
adjust the treatment plan accordingly [31].

Clinical Trial Matching

Matching patients to clinical trials based on their medical history and other criteria is also an application of
AI [34,35]. This can help speed up the drug development process and improve patient outcomes. ML
algorithms can analyze genetic data to identify patients who are likely to respond to a particular treatment.
This can help identify patients who are eligible for a clinical trial and improve the chances of success for the
trial.

Table 2 presents a comparison of various AI models that have been used in studies for predicting the risk
and outcomes in pancreatic cancer.

Study AI model used Input data used Outcomes Predicted Performance Limitations

72.5% sensitivity,
Smoking, Alcohol intake,
59% specificity for High number of
Malhotra Weight and BMI, symptoms,
Multivariate Logistic regression Screening tool predicting risk patients <60 y; 65% false positives;
et al. Past Medical History,
(MLR), Random Forest (RF) of PC development sensitivity, 57% poor external
[24] Medication History, Primary
specificity for validity
care consultation frequency
patients >60 y.

Data from only

Pancreatic cyst identification; one medical
Natural Language Processing Pancreatic cyst occurrence
Roch et subsequent follow-up for 99.9% sensitivity, center used; no
(NLP)- based pancreatic cyst from electronic health
al. [25] patients at risk of developing 98.8% specificity. other risk factors
identification system records
PC taken into
consideration

Identified 5 biomarkers for

Variations in
Differentially expressed pancreatic ductal
Almeida sample
genes identified as risk adenocarcinoma (PDAC) 100% sensitivity,
et al. Artificial Neural Network (ANN) preparation or
factors for pancreatic based on performance 94% specificity.
[26] microarray
cancer (PC) (FAIM3, IRANK3, DENND2D,
technologies.
PLBD1, AGPAT)

Support Vector Machine

Wang, Three genes (MMP7, FOS
Recursive Feature Elimination Differentially expressed Database-based
Liu, Ma and A2M) were found to be
(SVM-RFE) and Large Margin genes identified as risk 80% Specificity population - not
et al. closely related to the survival
Distribution Machine Recursive factors for PC self-measure
[27] rate of the patients.
Feature Elimination (LDM-RFE)

TABLE 2: Comparison of various AI models used in studies for predicting risk and outcomes in
pancreatic cancer.

Pancreatic cancer overall survival prediction

Pancreatic cancer usually has a very poor prognosis of a 5-year survival rate of less than 6%. Determining the
survival rate of pancreatic cancer is hard as the tumor usually presents with different features. Research is
being done to predict the survival time of pancreatic cancer using machine models that are non-invasive
and accurate. CT and PET scan images have been used with machine models to predict the survival time of
PC patients with more accuracy [36]. AI has also been used to predict the survival time when imaging is not
available. In one of the studies carried out by Hayward et al, they used clinical features and treatment
records to predict the outcomes and survival time [37]. Biomarkers and genes have also been used to
determine prognosis via ML. Algorithms have been used to analyze the identified biomarkers and genes,
which has helped in developing prognostic classifiers [38].

AI has revolutionized the medical field and made work easier for physicians. For pancreatic cancer, the
application of this technology is wide-ranging, from early diagnosis to predicting survival rates. Many
models have been designed that can help in predicting the recurrence of pancreatic cancer after treatment
[39].

Future perspectives

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 4 of 7

 AI prediction power will improve by training deep learning models with more input data. As pancreatic
cancer is prevalent worldwide, multi-institutional collaboration could help create a larger cohort of patients
for evaluation. These studies should focus on factors related to the prognosis and treatment of pancreatic
cancer (such as identifying ambiguous pancreatic lesions, the presence of vascular invasion, and the
treatment response). Another significant aspect is the utilization of AI for the analysis of PDAC behavior in
patients with risk factors for PDAC vs normal patients. Further study is also needed to demonstrate the
efficacy of AI for medical assistance by comparing the output of medical staff using AI aid to that of experts
not using AI support. The perception and acceptance of AI by healthcare providers and patients also
determine the extent of its use in the future. AI knowledge among healthcare providers has a greater impact
on their readiness to learn and apply its technology in their field, especially in the area of pancreatic cancer.
Gaining patient trust through education also determines the future success of AI in clinical practice.

Limitations
In the area of medicine, AI has its limitations. Even though AI has potential use for pancreatic cancer
diagnosis and treatment, it may not be ready for practical usage without further refinement. The
truthfulness of AI can be influenced by irrelevant databases with biases.

Therefore, it is crucial to design a non-biased, multicenter collaborative study, taking also into
consideration other important aspects, such as economics, ethical evaluation, and medical professional
regulations.

Ethical considerations
The advent of AI in pancreatic carcinoma diagnosis and treatment presents critical ethical concerns.
Protecting patient data for AI training while ensuring privacy remains paramount, necessitating stringent
data privacy measures. AI should augment healthcare professionals, not replace them, emphasizing
transparency in AI's capabilities and limitations. Mitigating biases within AI algorithms is crucial to
preventing disparities in diagnosis and treatment. Equitable access to AI-driven healthcare services and
addressing resource disparities are imperative ethical goals. Continuous discussions and regulations are
essential to govern AI's ethical implementation and impact on patient care, requiring collaboration among
regulators, healthcare professionals, technologists, and ethicists to establish ethical frameworks prioritizing
patient well-being in this evolving landscape.

Conclusions
AI has the potential to revolutionize the diagnosis and treatment of pancreatic cancer, which is a highly
lethal disease with a low survival rate. Early detection and accurate diagnosis are critical for improving
patient outcomes, and AI can analyze large amounts of patient data and medical images to identify patterns
and relationships that can lead to earlier and more accurate diagnoses. By analyzing genetic profiles,
imaging studies, and electronic health records, AI can identify specific biomarkers or genetic mutations that
are associated with pancreatic cancer, which can aid in early detection and diagnosis. Additionally, AI can
assist in the staging of pancreatic cancer by analyzing radiographic data to determine the extent of the
disease, which can guide treatment decisions.

In terms of treatment, pancreatic cancer is often resistant to chemotherapy and radiation therapy, which
makes it difficult to effectively treat. However, AI can help to identify more effective treatment by
simultaneously analyzing a wide variety of factors such as genetic profiles, treatment histories, and clinical
outcomes. This can lead to the development of more personalized and effective treatment plans, as well as
improve patient survival rates. AI can also assist in the drug discovery process by identifying new therapeutic
targets and predicting the efficacy of potential drugs.

Overall, the use of AI in pancreatic cancer is an area of active research and development that has the
potential to transform patient care and management in the field of oncology. By providing more objective
and consistent assessments of medical images and analyzing large amounts of patient data, AI has the
potential to improve early detection, increase diagnostic accuracy, and guide treatment decisions,
ultimately leading to better patient outcomes and more efficient use of healthcare resources. However, more
research is needed to fully realize the potential of AI in the field of pancreatic cancer.

Additional Information
Disclosures
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the
following: Payment/services info: All authors have declared that no financial support was received from
any organization for the submitted work. Financial relationships: All authors have declared that they have
no financial relationships at present or within the previous three years with any organizations that might
have an interest in the submitted work. Other relationships: All authors have declared that there are no
other relationships or activities that could appear to have influenced the submitted work.

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 5 of 7

 References
1. Canto MI, Harinck F, Hruban RH, et al.: International Cancer of the Pancreas Screening (CAPS) Consortium
summit on the management of patients with increased risk for familial pancreatic cancer. Gut. 2013,
62:339-47. 10.1136/gutjnl-2012-303108
2. Vincent A, Herman J, Schulick R, Hruban RH, Goggins M: Pancreatic cancer. Lancet. 2011, 378:607-20.
10.1016/S0140-6736(10)62307-0
3. Yu J, Sadakari Y, Shindo K, et al.: Digital next-generation sequencing identifies low-abundance mutations in
pancreatic juice samples collected from the duodenum of patients with pancreatic cancer and intraductal
papillary mucinous neoplasms. Gut. 2017, 66:1677-87. 10.1136/gutjnl-2015-311166
4. Kaul V, Enslin S, Gross SA: History of artificial intelligence in medicine . Gastrointest Endosc. 2020, 92:807-
12. 10.1016/j.gie.2020.06.040
5. Reichling C, Taieb J, Derangere V, et al.: Artificial intelligence-guided tissue analysis combined with
immune infiltrate assessment predicts stage III colon cancer outcomes in PETACC08 study. Gut. 2019,
69:681-90. 10.1136/gutjnl-2019-319292
6. Reismann J, Romualdi A, Kiss N, Minderjahn MI, Kallarackal J, Schad M, Reismann M: Diagnosis and
classification of pediatric acute appendicitis by artificial intelligence methods: an investigator-independent
approach. PLoS One. 2019, 14:e0222030. 10.1371/journal.pone.0222030
7. Maeda Y, Kudo SE, Mori Y, et al.: Fully automated diagnostic system with artificial intelligence using
endocytoscopy to identify the presence of histologic inflammation associated with ulcerative colitis (with
video). Gastrointest Endosc. 2019, 89:408-15. 10.1016/j.gie.2018.09.024
8. Litjens G, Kooi T, Bejnordi BE, et al.: A survey on deep learning in medical image analysis . Med Image Anal.
2017, 42:60-88. 10.1016/j.media.2017.07.005
9. Wang J, Yang X, Cai H, Tan W, Jin C, Li L: Discrimination of breast cancer with microcalcifications on
mammography by deep learning. Sci Rep. 2016, 6:27327. 10.1038/srep27327
10. Litjens G, Sánchez CI, Timofeeva N, et al.: Deep learning as a tool for increased accuracy and efficiency of
histopathological diagnosis. Sci Rep. 2016, 6:26286. 10.1038/srep26286
11. LeCun Y, Bengio Y, Hinton G: Deep learning. Nature. 2015, 521:436-44. 10.1038/nature14539
12. Taninaga J, Nishiyama Y, Fujibayashi K, Gunji T, Sasabe N, Iijima K, Naito T: Prediction of future gastric
cancer risk using a machine learning algorithm and comprehensive medical check-up data: a case-control
study. Sci Rep. 2019, 9:12384. 10.1038/s41598-019-48769-y
13. Mendoza Ladd A, Diehl DL: Artificial intelligence for early detection of pancreatic adenocarcinoma: the
future is promising. World J Gastroenterol. 2021, 27:1283-95. 10.3748/wjg.v27.i13.1283
14. Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts HJ: Artificial intelligence in radiology. Nat Rev
Cancer. 2018, 18:500-10. 10.1038/s41568-018-0016-5
15. Chen X, Fu R, Shao Q, et al.: Application of artificial intelligence to pancreatic adenocarcinoma . Front
Oncol. 2022, 12:960056. 10.3389/fonc.2022.960056
16. Li K, Yao Q, Xiao J, et al.: Contrast-enhanced CT radiomics for predicting lymph node metastasis in
pancreatic ductal adenocarcinoma: a pilot study. Cancer Imaging. 2020, 20:12. 10.1186/s40644-020-0288-3
17. Hameed BS, Krishnan UM: Artificial intelligence-driven diagnosis of pancreatic cancer . Cancers (Basel).
2022, 14:10.3390/cancers14215382
18. Zhu M, Xu C, Yu J, et al.: Differentiation of pancreatic cancer and chronic pancreatitis using computer-aided
diagnosis of endoscopic ultrasound (EUS) images: a diagnostic test. PLoS One. 2013, 8:e63820.
10.1371/journal.pone.0063820
19. Zhang Y, Cheng C, Liu Z, et al.: Radiomics analysis for the differentiation of autoimmune pancreatitis and
pancreatic ductal adenocarcinoma in (18) F-FDG PET/CT. Med Phys. 2019, 46:4520-30. 10.1002/mp.13733
20. Bera K, Schalper KA, Rimm DL, Velcheti V, Madabhushi A: Artificial intelligence in digital pathology - new
tools for diagnosis and precision oncology. Nat Rev Clin Oncol. 2019, 16:703-15. 10.1038/s41571-019-0252-
y
21. Yu Y, Chen S, Wang LS, et al.: Prediction of pancreatic cancer by serum biomarkers using surface-enhanced
laser desorption/ionization-based decision tree classification. Oncology. 2005, 68:79-86. 10.1159/000084824
22. Ko S, Choi J, Ahn J: GVES: machine learning model for identification of prognostic genes with a small
dataset. Sci Rep. 2021, 11:439. 10.1038/s41598-020-79889-5
23. Kenner BJ, Abrams ND, Chari ST, et al.: Early detection of pancreatic cancer: applying artificial intelligence
to electronic health records. Pancreas. 2021, 50:916-22. 10.1097/MPA.0000000000001882
24. Malhotra A, Rachet B, Bonaventure A, Pereira SP, Woods LM: Can we screen for pancreatic cancer?
Identifying a sub-population of patients at high risk of subsequent diagnosis using machine learning
techniques applied to primary care data. PLoS One. 2021, 16:e0251876. 10.1371/journal.pone.0251876
25. Roch AM, Mehrabi S, Krishnan A, et al.: Automated pancreatic cyst screening using natural language
processing: a new tool in the early detection of pancreatic cancer. HPB (Oxford). 2015, 17:447-53.
10.1111/hpb.12375
26. Almeida PP, Cardoso CP, de Freitas LM: PDAC-ANN: an artificial neural network to predict pancreatic
ductal adenocarcinoma based on gene expression. BMC Cancer. 2020, 20:82. 10.1186/s12885-020-6533-0
27. Wang Y, Liu K, Ma Q, et al.: Pancreatic cancer biomarker detection by two support vector strategies for
recursive feature elimination. Biomark Med. 2019, 13:105-21. 10.2217/bmm-2018-0273
28. You Y, Lai X, Pan Y, et al.: Artificial intelligence in cancer target identification and drug discovery . Signal
Transduct Target Ther. 2022, 7:156. 10.1038/s41392-022-00994-0
29. Qian Y, Gong Y, Fan Z, et al.: Molecular alterations and targeted therapy in pancreatic ductal
adenocarcinoma. J Hematol Oncol. 2020, 13:130. 10.1186/s13045-020-00958-3
30. Kaissis G, Ziegelmayer S, Lohöfer F, et al.: A machine learning algorithm predicts molecular subtypes in
pancreatic ductal adenocarcinoma with differential response to gemcitabine-based versus FOLFIRINOX
chemotherapy. PLoS One. 2019, 14:e0218642. 10.1371/journal.pone.0218642
31. Li X, Yang L, Yuan Z, et al.: Multi-institutional development and external validation of machine learning-
based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection. J Transl

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 6 of 7

 Med. 2021, 19:281. 10.1186/s12967-021-02955-7
32. Nasief H, Hall W, Zheng C, Tsai S, Wang L, Erickson B, Li XA: Improving treatment response prediction for
chemoradiation therapy of pancreatic cancer using a combination of delta-radiomics and the clinical
biomarker CA19-9. Front Oncol. 2019, 9:1464. 10.3389/fonc.2019.01464
33. Han X, Hong J, Reyngold M, et al.: Deep-learning-based image registration and automatic segmentation of
organs-at-risk in cone-beam CT scans from high-dose radiation treatment of pancreatic cancer. Med Phys.
2021, 48:3084-95. 10.1002/mp.14906
34. Alexander M, Solomon B, Ball DL, et al.: Evaluation of an artificial intelligence clinical trial matching
system in Australian lung cancer patients. JAMIA Open. 2020, 3:209-15. 10.1093/jamiaopen/ooaa002
35. Zhao Y, Kosorok MR, Zeng D: Reinforcement learning design for cancer clinical trials . Stat Med. 2009,
28:3294-315. 10.1002/sim.3720
36. Toyama Y, Hotta M, Motoi F, Takanami K, Minamimoto R, Takase K: Prognostic value of FDG-PET
radiomics with machine learning in pancreatic cancer. Sci Rep. 2020, 10:17024. 10.1038/s41598-020-73237-
3
37. Hayward J, Alvarez SA, Ruiz C, Sullivan M, Tseng J, Whalen G: Machine learning of clinical performance in a
pancreatic cancer database. Artif Intell Med. 2010, 49:187-95. 10.1016/j.artmed.2010.04.009
38. Yokoyama S, Hamada T, Higashi M, et al.: Predicted prognosis of patients with pancreatic cancer by
machine learning. Clin Cancer Res. 2020, 26:2411-21. 10.1158/1078-0432.CCR-19-1247
39. Baek B, Lee H: Prediction of survival and recurrence in patients with pancreatic cancer by integrating multi-
omics data. Sci Rep. 2020, 10:18951. 10.1038/s41598-020-76025-1

2023 Kumar et al. Cureus 15(11): e49560. DOI 10.7759/cureus.49560 7 of 7