Detection and Classification of Microcalcifications Based on DWT and ANFIS

Nowadays, breast cancer has become one of the most dangerous tumors for middle-aged and older women in China. Mammography plays an important role in the clinical diagnosis of breast cancer, and microcalcifications (MCs) are one of the main symptoms in the mammograms. In order to assist the radiologists to detect the MCs accurately and rapidly, a novel computer-aided diagnosis method was proposed in this paper. DWT was used to extract the high-frequency signal of the images firstly, and thresholding with hysteresis was applied to locate the suspicious MCs. Then, filling dilation was applied to segment those desired regions. During the detection, ANFIS was used to adjust the parameters, making the CAD algorithm more adaptiv, and precise. At last, the suspicious MCs were classified with MLP, and the experiments showed the advantages of the proposed method over the conventional ones.