Breast Cancer Detection Using Mammographic Images over Convolutional Neural Network

The biggest cause of death worldwide, particularly for women, is breast cancer. Early cancer detection and treatment can significantly increase the chances of survival and reduce costs for patients. The first important point of decision in cancer treatment is histology, which depends on a histologist confirming the presence of cancer. The time it takes to review the slides, the amount of data, and the potential for errors make these the main use cases of automatic detection in the health care setting. The ability to accurately and quickly identify potentially cancerous areas for review by histologists can and helps optimize the use of their time. This detection can be performed using CNN.MIAS dataset for mammograms is employed for this purpose, with 322 mammograms, almost all of which 189 photos of normal breasts and 133 photographs of aberrant breasts are shown. In this article, computer vision models like the convolutional neural network are used to predict whether mammary gland tissues are benign or cancerous. The investigation is ongoing, and new advancements are being made by enhancing the CNN architecture and utilising trained neural networks that, ideally, will result in more accurate measurements.