Breast mass recognition based on developed genetic algorithm and support vector machine

Breast cancer is regarded as one of the leading causes of death in women all over the world. Computer-aided detection (CAD) system can effectively improve the accuracy, efficiency and consistency of breast cancer diagnosis in clinical environment. For mass recognition problem, a method based on developed genetic algorithm (GA) and support vector machine (SVM) is proposed in this paper. First, gray, shape and texture features are extracted from segmented masses. Then, all these features are encoded into binary chromosome. And the individual fitness of GA is constructed by combining the classification accuracy of SVM classifier and the number of selected features. At last, the optimal feature set is obtained after the genetic operations. And it is used to be the input of the SVM classifier to classify new cases. To evaluate the performance of the proposed methods, different feature selection methods and classifiers are used. Comparing experiments are performed on 2667 cases from DDSM database. The results show that the proposed method can well improve the accuracy of mass recognition. ©, 2015, Binary Information Press. All right reserved.