BREAST CANCER DETECTION USING BAYESIAN NETWORK

Abstract

Breast cancer remains a major global health problem requiring effective diagnostic methods for early detection. Mammograms, although essential, may miss a significant percentage of cancer indicators. This thesis explores the potential of using machine learning techniques, specifically Bayesian Networks (BN) combined with Convolutional Neural Networks (CNN), to improve breast cancer diagnosis. The literature review highlights the urgency of improving diagnostic methods and identifies thermography as a promising, cost-effective, and non-invasive and alternative detection method. Systematically, many data sets have been collected and used to develop and train robust BN models. The so-called Model A, combining thermal images and medical records, achieved an accuracy of 84.07%, while Model B, incorporating CNN predictions on top of the datasets, achieved an accuracy of 90.9341%. These results demonstrate the potential of machine learning to transform breast cancer diagnosis, increasing accuracy and reducing the risk of misdiagnosis. Future research aims to increase dataset sizes and improve model performance, ultimately improving healthcare outcomes in breast cancer detection, in order to achieve WHO’s ultimate goal of breast self-examination (BSE).