Deep Learning in Breast Cancer Detection and Classification

Abstract

Breast cancer is considered one of the primary causes of mortality among women aged 20–59 worldwide. Early detection and treatment can allow patients to have proper treatment and consequently reduce rate of morbidity of breast cancer. Research indicates that most experienced physicians can diagnose cancer with 79% accuracy while 91% correct diagnosis is achieved using machine learning techniques. In this paper, we present the most recent breast cancer detection and classification models that are machine learning based models by analyzing them in the form of comparative study. Also, in this paper, the datasets that are public for use and popular as well are listed in the recent work to facilitate any new experiments and comparisons. The comparative analysis shows that the recent highest accuracy models based on simple detection and the classification architectures are You Only Look Once (YOLO) and RetinaNet.