Improved Hurst exponent based on genetic algorithm in schizophrenia EEG

Abstract

Schizophrenia is a serious mental disorder, and event-related potentials can effectively reflect the differences between patients and healthy individuals. However, currently, there is no effective method to analyze the Hurst index of short-term electroencephalogram (EEG) signals. This paper improves the calculation method of the Hurst index based on a multipoint fractional Brownian bridge using a genetic algorithm and verifies its feasibility on simulated signals. Through the event-related potential analysis of EEG signals in schizophrenia patients, it was found that the greatest difference between patients and healthy people was in the frontal lobe, and there were differences in the N100 wave, the P50 wave, and the mismatch negativity (MMN) wave in the frontal lobe under sound and action stimulation, and their Hurst index increased. In conclusion, the study found differences in event-related potentials between the frontal and central regions of schizophrenia patients compared to healthy subjects. This research may aid in the diagnosis of schizophrenia patients in clinical practice.