Automatic Classification of Benign and Malignant Breast Tumors in Ultrasound Image with Texture and Morphological Features
Mengwan Wei, Yongzhao Du, Xiuming Wu, Jianqing Zhu · 2019
Breast cancer has become one of the malignant tumors with the highest morbidity and mortality among women in the world, which seriously threatens women's health. Because the cause of breast cancer is unknown, early detection, early diagnosis and early treatment are the key to improve the cure rate of breast cancer. Computer aided diagnosis (CAD) system is a valuable assistant method for automatic detection and classification of breast cancer. In this paper, the morphological and texture features of ultrasonic images of breast tumors are extracted to train SVM classifier, so as to realize automatic classification of tumors. Of the 1061 newly collected breast ultrasound images, 589 were malignant and 472 were benign. Finally, features combined with texture and morphological get the best performance and the overall accuracy reaches 87.32%. The results show that this method has a good generalization ability and can be used as an assistant method for the classification of benign and malignant breast cancer.