Image Representation Based Hybrid Intelligent Diagnosis Approach for Computer Aided Diagnosis (CAD) Systems

Abstract

Computer Aided Diagnosis (CAD) is one of the most interesting and most difficult dilemma dealing in one hand with expert (human) knowledge consideration. On the other hand, fault diagnosis is a complex and fuzzy cognitive process and soft computing approaches as modular neural networks and fuzzy logic, have shown great potential in the development of decision support systems. In this paper, a brief survey on fault diagnosis systems, knowledge representations, and modular neural networks is given. From the classification and decisionmaking problem analysis, a hybrid intelligent diagnosis approach is suggested from signal to image conversion (image representation). In this approach, each image is divided in several sub-images (local indicators) which are classified by global approximators MultiLayer feedforward Perceptron networks (MLP) and by local approximators Radial Basis Function networks (RBF). Then, the suggested approach is developed in biomedicine for a CAD, from Auditory Brainstem Response (ABR) test, and the prototype design and experimental results are presented. Finally, a discussion is given with regard to the reliability and large application field of the suggested approach.