Hybrid Intelligent Pattern Recognition Systems for Mass Segmentation and Classification: A Pilot Study on Full-Field Digital Mammograms

Governments and health authorities emphasize the importance of early detection breast cancer, usually through mammography, to improve prognosis, increase therapeutic options achieve optimum outcomes. Despite technological advances advent full-field digital mammography (FFDM), diagnosis abnormalities on mammographic images remains a challenge due qualitative variations in different tissue types densities. Highly accurate computer-aided (CADx) systems could assist differentiation between normal abnormal classification as benign or malignant. In this paper, classical, advanced fuzzy sets fusion techniques for image enhancement were combined with three thresholding methods (Global, Otsu type-2 threshold) classifying (K-means, FCM ANFIS) masses FFDM. The aim paper is identify performance sets, segmentation, decisions based K-means FCM, ANFIS classifier. Sixty-three combinations evaluated ninety-seven (sixty-five thirty-two malignant). sixty-three was by estimating accuracy, F1 score, area under curve (AUC). LH-XWW method classifier outperformed all other an accuracy 95.17%, score 89.42% AUC 0.91. This algorithm seems offer promising CADx system cancer