Mammographic Image Classification Using Deep Neural Network for Computer-Aided Diagnosis

Breast cancer detection is a crucial topic in the healthcare sector. major reason for increased mortality rate recent years among women, specifically developed and underdeveloped countries around world. The incidence less India than countries, but awareness must be increased. This paper focuses on an efficient deep learning-based diagnosis classification technique to detect breast from mammograms. model includes preprocessing, segmentation, feature extraction, classification. At initial level, Laplacian filtering applied identify portions of edges mammogram images that are highly sensitive noise. Subsequently, segmentation done using modified adaptively regularized kernel-based fuzzy C means (ARKFCM). Feature extraction accomplished morphological, texture, moment invariants. corresponding values provided as inputs neural network (DNN) classifies normal abnormal images. performance proposed validated with Mammographic Image Analysis Society (MIAS) database. efficiency classifier experimentally proved by comparing various classifiers respect their statistical performances. On database, offered maximum highest accuracy level 99.13%.