An Intelligent System for Diagnosing Schizophrenia and Bipolar Disorder based on MLNN and RBF

This paper is concerned with the problem of discriminating the patients suffering from schizophrenia and bipolar disorder versus control based on EEG rhythms. The EEG rhythms are used to extract a feature vector for each patient. In this paper, the large set of features included in the EEG rhythms is reduced into smaller set of features using FFT segmentation. Two classes of classifiers which are multi-layer perceptron and radial basis function are used to discriminate the patient data based on features vector. Experimental studies have shown that the proposed algorithms give excellent results when applied and tested on the three classes. The multilayer neural network with backpropagation achieved a high performance rate equal to 98.67 % compared to radial basis function networks which achieved a performance rate equal to 87.33%.