Classification of Masses in Mammograms Using Selected Features Using Subclass Based Learning Neural Network
K. Dinakaran, V. Sivakrithika · Transylvanian Review · 2016
Computer aided detection (CAD) assists radiologists by providing second opinion in the mammography detection, and reduce misdiagnosis. In this work the task of automatically classifying the mass tissue into benign and malign based on the characteristics of mass is investigated. Mass is characterized by its shape, margin, density of the patient. Geometrical shape, margin and texture features are used to represent the radiological characteristics of mass. In this paper we investigated a novel subclass based learning neural network classifier which classifies the input to their respective subclasses and then to the main class. The feature space that represent mass characteristics can be grouped under multiple classes ,the proposed work makes use of this idea to classify the input vector into sub classes with respect to density, shape and margin and then into main classes benign and malign. The experiments using the proposed work have been implemented on Mammographic Image Analysis Society (MIAS) Database. The experiments were implemented in MATLAB.