Research on Sleep EEG Signals Based on IOTA

Abstract

As a nonlinear analysis method based on permutation, internal composition alignment (IOTA) algorithm can study the coupling between systems by calculating the coupling coefficient between two time series. In this paper, the internal composition alignment (IOTA) algorithm is used to study the sleep EEG signals generated by the human body in different sleep periods. Firstly, the IOTA coefficients between different time series calculated by this method are used as nodes to construct the sleep function networks in different sleep periods, and the statistical characteristics of networks such as node degree and clustering coefficient are selected to compare different sleep networks. The results show that the IOTA coefficient and the node average degree and aggregation coefficient of EEG network in NREM-I period are higher than those in awake period, indicating that the complexity of EEG network in NREM-I period is higher than that in awake period, and that the coupling degree in NREM-I period is also higher than that in awake period. This experiment proves the effectiveness of IOTA algorithm for analyzing sleep function network. This algorithm can be used to study sleep EEG staging. At the same time, it also provides an important reference for the research, clinical diagnosis and treatment of sleep diseases in the future.