Chapter 5 - Screening and early identification of microcalcifications in breast using texture-based ANFIS classification

Abstract

The objective of this investigation is to widen the robust detection schemes for the detection of breast cancer at an early stage. This is a noninvasive method called the adaptive neurofuzzy structure (ANFIS) to diagnose microcalcifications (cancerous lesions) in the mammary glands. An investigation of breast cancer by using ANFIS along with the clinical inputs from medical practioners is proposed. Outcomes of the discussions from various medical practioners and various algorithms are reviewed. Clinical images from the MIAS databased are used for testing and training of the proposed algorithm. Texture-based entropy values are also used for training and testing of features. The ANFIS-based classification is an efficient method of breast cancer diagnosis. The performance of the proposed (texture features with ANFIS) method is compared with the outputs obtained from the K-means clustering algorithm, wavelet transform, and artificial neural network-based classification using mean absolute deviation as a feature for early stage diagnosis.