Breast Cancer Tissue Identification Using Deep Learning in Mammogram Images

Abstract

Breast cancer, the most common cancer in women, is best detected through screening programs. Mammography is the most common screening test, yet human error necessitates computer-assisted diagnosis. Convolutional networks, a machine learning technique, can assist detect breast masses and improve microcalcification identification in mammograms. Automatic breast cancer detection in mammography using convolutional networks has the potential to improve both the precision and timeliness of diagnosis, hence increasing survival rates. This method utilizes a single, easily-learned step of picture segmentation in order to detect breast masses and microcalcification clusters, both of which are strong indicators of breast cancer. The application of deep learning algorithms to the analysis of mammograms has the potential to greatly improve the diagnosis and treatment of breast cancer.