Web Based Real Time Bowel Cancer Stage Suggestion Using CNN

Cancer that develops in the colon or rectum of the large intestine is called bowel cancer (CRC), although it goes by other names as well. The early detection of bowel cancer relies largely on the microscopic inspection of entire tissues samples. Due to its inherent subjectivity, the conventional visual examination is both laborious and wildly inaccurate. Using a deep learning method based on Convolutional Neural Networks (CNN) to distinguish between tumor tissue and healthy tissues, we provide a web-based flask application that may predict the stage of bowel cancer. 4550 image data of seven different tissue types in total are collected and analyzed using pandas and sklearn. A CNN model is trained after preprocessing and cleaning of data, the accuracy of the model is calculated and analyzed using Confusion Matrix. The model is deployed to a production environment using Flask as a Web Application. Though the information may not replace the advice of a doctor, the Application may serve as an intermediate between traditional and automated analysis of tissue for fast prediction of the stage of the cancer by Doctors.