Investigation of Quantitative Electroencephalography Markers for Schizophrenia Diagnosis using Variational Mode decomposition

Analyzing the human brain and detection of brain disorders using Electroencephalography (EEG) have gained impetus over past few years. Schizophrenia is a mental disorder which affects the functioning of brain. In this paper, we aim to determine quantitative significant EEG features which can be used to differentiate person having schizophrenia disorder and the healthy ones. Variational mode decomposition method (VMD) is used for feature extraction followed by statistical significance analysis of the extracted values. VMD is a popular decomposition method due to its adaptive nature. In this study, we also aim to examine the efficacy of the method in EEG signal analysis. Six different features were calculated and analyzed over the decomposed EEG signal. Analysis is done by performing statistical significance test on SPSS version 23 and by observing the p-values and error bar graphs. Results indicate that five out of six features turned out to be a very good indicator/markers. Hence, this method could be further extended by applying machine learning approach for the classification based on obtained feature set.