Scribd
Upload
12 views1 page

Brain Image Processing Project

The document discusses a project that aims to detect lung cancer stages from CT images and predict patient survival risk. It involves developing models for lung cancer stage identification and survival risk assessment. The models were tested on a dataset of 262 CT images and results show the proposed method achieves higher accuracy compared to existing methods.

Uploaded by

santoshpatil841994
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
12 views1 page

Brain Image Processing Project

The document discusses a project that aims to detect lung cancer stages from CT images and predict patient survival risk. It involves developing models for lung cancer stage identification and survival risk assessment. The models were tested on a dataset of 262 CT images and results show the proposed method achieves higher accuracy compared to existing methods.

Uploaded by

santoshpatil841994
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 1

Lung cancer stage detection and survival risk assessment

Ganesh Naik - 2BV12IS026 Santosh Patil - 2BV12IS085


Shrivatsa S -2BV12IS093 Sunil Chikkud - 2BV12IS112 Swati Ankolekar - 2BV12IS116
Guide: Prof. Shrinivas D Desai

Abstract 4. Implementation details


Cancer is one of the fatal diseases, posing threat to human life. Among different types
Implementation details consists of two main modules one is lung cancer stage identification.
of cancer, lung cancer can be considered as one of the most deadly one. Patient suffering from
Another one is life risk prediction of cancer patients
cancer and his/her relatives will be keen to understand the survival rate. This information is
also important for oncologist to give better treatment to patients. Lung cancer stage identification. Lung cancer survival risk assessment.

stages of lung cancer Criteria Total score Risk


Project aims at detecting stage of the lung cancer by using CT images and to predict
1A r<=2mm 9-15 Low risk
the risk factor of cancer in scale of four grade system viz; low ,average, high, and very High
Risk. Comparative analysis of proposed solution with current existing method as well as 1B r>2mm and r<=3mm
16-24 Average Risk
accuracy of the proposed method is done. 2A r>3 mm and r<=5mm
2B r>5mm and r<=7mm 25-32 High risk
Stage 3 r>7mm 33 Very high risk

1. Problem Statement
5. Results and Discussions
To identify the lung cancer stages from CT images and to find probability of
life risk factor.
Lung Cancer Stage detection

2. Introduction and Objectives


Lung cancer is most common cause of death across the world. Lung cancer is essentially
investigated by lung nodules, which appears as a small white spot in lung parenchyma. The
sensing and diagnosis of suspicious lung nodules in CT images are one of the primary
challenges in medical image analysis.
Objectives
• Design and develop interactive and effective model for lung cancer stage detection.
• To predict the risk factor of cancer in terms of quantitative measures.
• To perform comparative analysis of proposed solution with existing current methods.
• Evaluate the performance of proposed method.
Accuracy for lung cancer stage detection ROC curve for stage detection of lung cancer

Predicted Actual

Image No Stage of Stage of Confusion


3. Methodology 1
cancer
1A
cancer
1A
matrix
TP
2 1B 1B TP
The system model consists of sub modules such as pre processing ,segmentation, Feature, Accuracy =81.25%

extraction, ROI selection and Life risk assessment of lung cancer patients
Accuracy for lung cancer survival risk assessment. Comparative study of cancer survival prediction
Predicted Actual Publication Cancer type No of Accurac
patients y
1 low low TP
Chen Y-C .Lung cancer 440 83.5%
2 Average Average TP
proposed Lung cancer 8 87.5%
Accuracy =87.5%. metod

6. Conclusion
Conventionally life risk prediction is carried out using genetic and habitual attributes. In
this paper we attempted to consider cancer related attribute during prediction process. For
experiments, data set were collected from Cancer archieve.net website. It consists of 262 CT
images of lung cancer affected patients. To predict the life risk of cancer patients we have
considered two more important attributes such as “stage of cancer” and “work place”. We
customized Roper St.Fransis model to fit these modifications. By considering these attribute our
proposed method is found to provide more accurate results of life risk prediction. Comparative
Training model. Testing model.
analysis of existing method with the proposed method shows that accuracy of proposed method is
better than existing one.

Attribute considered along with Roper st,Fransis model.

Parameter Weight age Score


Work place (Hazards) Yes 3 References
No 1 [1] .G. Vijaya and A. Suhasini. “An Adaptive Preprocssing of Lung CT Images with
Various Filters for better Enhancement". pages 6:179–184, 2014.
Stage 1A 1,
1B 2 [2] .Al Gindi, Amal M., Tawfik A. Attiatalla, and Moustafa M. Sami. "A Comparative
2A 3
Study for Comparing Two Feature Extraction Methods and Two Classifiers in
Classification of Early stage Lung Cancer Diagnosis of chest x-ray images." J Am
2B 4 Sci 10 (2014): 13-22.
Stage 3 5

You might also like