Automatic detection of mitosis cell in breast cancer histopathology images using genetic algorithm

Nowadays, pathologist grade breast cancer histopathology slides by microscopes based on Nottingham as an international standard. The mitotic counting is one of the three scoring criteria in Nottingham standard for breast cancer grading based on histopathology slide image studies. Large number of non-mitosis organs, which exists in histopathology slide tissue, is one of the most important challenges facing mitosis detection methods. In this paper, a system for automatic mitosis detection purpose from breast cancer histopathology slide images is proposed to aid pathologists for mitotic cells counting. In the proposed algorithm the number of non-mitosis candidates are defined as a cast function and by minimization using Genetic Optimization algorithm, the most of the non-mitosis candidates will be omitted. Then some features such as co-occurrence and run-length matrices and Gabor features are extracted from the rest of candidates and finally mitotic cells are classified using support vector machine (SVM) classifier. Experimental results demonstrate the efficiency of this method to detect mitotic cells in breast cancer histology images.