A Survey on Automated Detection of Breast Cancer based Histopathology Images

The increasing mortality rate in the women population is mainly due to breast cancer. Diagnosing breast cancer in its early stages will always remain crucial. Hence identifying and treating the disease at the earliest will increase the possibilities of survival. Recently, by using ultrasound images a computer-aided diagnosis (CAD) system is being developed to help radiologists to attain higher accuracy for identification. Normally, a CAD system comprises of different phases such as pre-processing, segmentation for regions of interest, feature selection & extraction, and last phase is to do classification. This paper illustrates the various methods used to deploy an automated CAD system development for the early identification of cancer disease. In this paper, various approaches used are abridged and their pros and cons are compared. The performance evaluation of the CAD system is also depicted as well. The dataset of breast cancer histology images (BACH) is made available to participate in a grand challenge aimed at the classification of microscopy and whole slide images, whereas it is made publicly available for the challenge to promote further improvements for developing an intelligent classification system in digital pathology. According to the number of diagnostic classes and image types (Microscopy and whole slide images), an intelligent system is implemented for initial detection for deploying a proper treatment for breast cancer.