Developement of computer aided system for detection and classification of mitosis using SVM

Abstract

In modern medicine, digital pathology represents one of the major and challenging research field. Pathological exams have a critical role in diagnosis process. Histopathological grading of breast cancer provides prior knowledge to the patients prognosis and helps to make further treatment plans. But, manual analysis of numerous biopsy slides is a labor-intensive work for pathologists. In the case of breast cancer grading, Nottingham grading system (NGS) is the standard. Tubule formation, nuclear atypia, and mitosis count are the three criterion used for NGS. Mitosis counts play an essential role for breast cancer grading. The recent advances in histopathology has the potential to assist the pathologist. This paper presents a comprehensive review on the state-of-the-art image analysis research focusing on automated detection of mitotic cells. The main goal of this paper is the development of a system to provide automatic detection and classification of mitosis from histopathological images.