An expert system of prediction for the classification of schizophrenic patients:: Esquizor

The purpose of the present study was to develop and test the efficacy of an expert system aimed at differentiating between resistant and non-resistant paranoid schizophrenics to typical neuroleptic treatment, and a group of healthy controls, classified according to their brain electrical activity (EEG). The system involves the generation of a decision tree based on learning of examples. Absolute and relative power values, as well as inter and intrahemispheric correlation values were used to feed the system, which resulted in Successful classification of patients and controls. Frequency values provided more information than broad bands. Esquizor could also be useful to classify other groups based on their EEG.