Breast Cancer Prognosis Using Machine Learning Approaches

Abstract

Cancer is a group of diseases and breast cancer is one of the leading cancer types in which cells in breast tissue change or reform abnormally and divides in an uncontrolled manner, typically emerged in a lump or mass. Every year more than 2.7 million new cases of breast cancer are enrolled worldwide, and according to GLOBOCAN statistics of 2020, breast cancer is the leading type of cancer in the world followed by lung cancer. In the present medical scenario, detection of breast cancer in its premature stage is immensely challenging. So, early detection and diagnosis are required to reduce mortality gains demand. Thanks to artificial intelligence and machine-learning–based automatic approaches which make it possible to detect and diagnose breast cancer using radiographic mammograms in early stages without human intervention. This study aims to provide a deep insight into breast cancer and the role of machine learning approaches to early detection and diagnosis of tumour masses. Different state-of-the-art techniques and algorithms are overviewed to get a proper understanding of how to utilize them for breast cancer prognosis. The pipeline of the general approach has also been explored which proves valuable to help the expert medical community and researchers in automatic diagnosis and detection of breast cancer at an early stage to reduce death rate.