MiDeSeC: A Dataset for Mitosis Detection and Segmentation in Breast Cancer Histopathology Images

Prof. Dr. Bilge Ayca Kirmizi, akarabork@yahoo.com 1 Introduction Nottingham Grading System [1] emphasizes three key morphological features on Hematoxylin and Eosin (H&E) stained slides to grade breast cancer: mitotic count, tubule formation, and nuclear pleomorphism. Mitotic count is the most prominent feature among them. Searching for mitosis on glass slides is a routine procedure for breast pathologists. Since there are so many high power fields (HPFs) on a single slide and mitotic cells vary in appearance, it is a tedious and time - consuming task. Additionally, mitotic cell judgment is somewhat subjective, making it difficult for pathologists to reach a consensus. Thus, it is extremely important to develop automatic detection methods that will not only save time and material resources, but will also enhance the reliability of pathological diagnosis.