Analysis of depressive EEG signals via symbolic phase transfer entropy with an adaptive template method

Abstract

Depression is a prevalent mental disorder in contemporary society. Symbolic phase transfer entropy can quantify the dynamic interaction and information flow between electroencephalogram (EEG) signals in depressed patients and healthy groups, which can help diagnose and treat depression. However, the traditional symbolization process of symbolic phase transfer entropy adopts the basic template method, which makes the symbolic phase transfer entropy unable to express the characteristics and changes of time series in different time periods in detail. Therefore, this paper proposes an improved symbolic phase transfer entropy algorithm, which adopts the adaptive template method in the symbolization process of the symbolic phase transfer entropy algorithm so that it can capture the dynamic changes of time series more finely. It was verified on the task EEG signals of 40 depressed patients and 40 healthy people. The experimental results show that the improved symbolic phase transfer entropy can more accurately distinguish depressed patients from healthy people in lead F4 and lead O1, which is helpful for the study of the EEG pathological characteristics of depression. The improved symbolic phase transfer entropy algorithm makes up for the shortcomings of the traditional symbolic phase transfer entropy in capturing the dynamic changes of time series and provides help for the study of dynamic changes in complex systems.