Early detection of breast cancer using optimized ANFIS and features selection

Abstract

Breast cancer is one of the widespread scourges amongst women worldwide. Breast cancer is the most prominent known killer of women between the ages of 35 and 54. Effective diagnosis of breast cancer remains a major challenge and early diagnosis is extremely important in helping prevent the most serious manifestations of the disease. In this paper a new method is presented for early detection of breast cancer based on adaptive neuro-fuzzy inference system (ANFIS) and feature selection. In this method, ANFIS is used as intelligent classifier and association rules (AR) technique is used as feature selection algorithm. In ANFIS, the value of radius has significant effect on system accuracy. Therefore, in the proposed method we used cuckoo optimization algorithm (COA) to find the optimal value of radius. The proposed method is applied on Wisconsin Breast Cancer Database (WBCD) and the results show that the proposed method has high detection accuracy.