Mitosis detection techniques in H&E stained breast cancer pathological images: A comprehensive review

Abstract Quantifying mitosis in pathological sections is of great significance in the pathological diagnosis of breast cancer as it is used to evaluate the aggressiveness of the tumor and to provide more comprehensive and reliable information for accurate diagnosis and treatment. In this paper, we summarized the current mainstream methods of mitosis detection and divided them into four categories, namely traditional methods, deep learning methods, methods combining deep learning with traditional methods and other methods. Each method is introduced, and the performance indicators achieved by some of these methods and their results are discussed, compared, and evaluated. We summarize some solutions to the problem of positive and negative sample imbalance in mitotic datasets. Through the review of the research methods in this field, the existing methods of mitosis research in breast cancer are summarized, and the future developments are prospected.