Classification and medical diagnosis using wavelet-based fuzzy neural networks

Abstract

In this paper, we propose a Wavelet-based Fuzzy Neural Network (WFNN) for classification and medical diagnosis. The proposed WFNN integrates the wavelet transform functions into a fuzzy system. We use non-orthogonal and compactly supported functions as the wavelet neural network (WNN) bases. The goal of the WFNN model combined with WNN is to improve the accuracy of function approximation. An on-line structure/parameter learning algorithm is used in the WFNN. Structure learning is based on the input partitions to determine the number of fuzzy rules and wavelet functions, and parameter learning is based on the supervised gradient descent method to adjust the shape of the membership functions and the connection weights of the wavelet neural networks. Computer simulations were conducted to test the performance and applicability of the proposed system.