Analysis of Fuzzy Techniques and Neural Networks (RBF&MLP) in Classification of Epilepsy Risk Levels from EEG Signals

Abstract

Most research to date using hybrid systems (Fuzzy-Neuro) focused on the Multi-Layer Perceptron (MLP). Alternative neural network approaches such as the Radial Basis Function (RBF) network, and their representations appear to have received relatively little attention. Here we focus on RBF network as an optimizer for classification of epilepsy risk level obtained from the fuzzy techniques using the EEG signals parameters. The obtained risk level patterns from fuzzy techniques are found to have low values of Performance Index (PI) and Quality Value (QV). These neural networks are trained and tested with 480 patterns extracted from three epochs of sixteen channel EEG signals of ten known epilepsy patients. Different architectures of MLP and RBF networks are compared based on the minimum Mean Square Error (MSE), the better networks in MLP (16-16-1) and RBF (1-16-1) are selected. RBF out performs the MLP network and fuzzy techniques with the high Quality Value of 23.98 when compared to the Quality Values of 16.92 and 6.25.