Colorectal abnormalities detection and classification using Deep Learning techniques

Abstract

This dissertation aims to present deep learning techniques, specifically Convolutional Neural Networks (CNNs), for the detection and classification of colorectal polyps in medical images. Colorectal cancer is a significant global health concern, and early detection of polyps is crucial for effective treatment and prevention. The research focuses on developing a CNN-based model to analyze colonoscopy images and accurately identify regions containing polyps. Various CNN architectures and training strategies are investigated, including transfer learning and fine-tuning, to optimize the model's performance. The model is evaluated on the CVC-ClinicDB dataset, and the results demonstrate high accuracy, precision, sensitivity, and specificity in polyp detection and classification. The research contributes to the advancement of computer-aided diagnosis tools for colorectal cancer screening and has the potential to improve patient outcomes through earlier detection and intervention.