Mammograms are an easy and cost-effective way to detect breast cancer early. Women having high-density breast tissues are more prone to misdiagnosis during mammography. The dense tissues hide developing tumours, a phenomenon called masking, resulting in false negatives and misses early detection. Identifying mammographic masking could help recommend suitable breast cancer screening methods. In contrast to the previous approach of classifying mammograms based on breast density, we aim to identify masking using the CSAW-M dataset containing 10,000 images. Experimenting with different deep learning algorithms, we intend to create a model that could identify masking comparable to a human expert.
Here is the overleaf page that contains the annotated bibliography of the papers reviewed for the literature.
This project is done as part of my Master's Thesis supervised by Prof. M.M. Gore at MNNIT Allahabad.