Multilayer-Aggregation Functional Network for Identifying Brain Fatigue and Diseases

Wen-Kuo Cui,Xin-Rui Qi,Yu Sun,Gang Yan

doi:10.3389/fphy.2021.822915

Abstract

Recent years have witnessed increasing interest of applying network science methodologies to analyze brain activity data. Owing to the noninvasiveness, low cost and high sampling rate, electroencephalogram (EEG) recordings have been widely used as a proxy for probing the internal states of human brains. Previous correlation-based functional networks (CFN) mainly focused on the covariance or coherence between readings from electrodes attached to different regions, largely overlooking local temporal properties of these electrical activities. Here, we propose a method to construct multilayer-aggregation functional network (MAFN) which is able to capture both temporal and topological characteristics from EEG data. We extract features from these MAFNs and incorporate them into each of 12 classification algorithms, aiming to detect mental fatigue and two brain diseases, schizophrenia and epilepsy. The results demonstrate that MAFNs consistently outperform CFN and dynamic version of CFN. In comparison to functional networks based on weighted phase lag index (wPLI), MAFNs also achieve higher or comparable accuracy in most classifiers. Moreover, the nodal features of MAFNs allow us to identify the important positions of EEG electrodes for different brain states or diseases. These findings together offer not only a framework for classifying normal and abnormal brain activities but also a general method for constructing more informative functional networks from multiple time series data.

Highlights

Human brain is one of the most delicate and complex systems, responsible for maintaining the internal regulation of human body and perception, and responding to external stimuli
The results show that in comparison to correlationbased functional networks (CFNs) and dynamic version of correlation-based functional networks (CFN) (DCFNs), multilayer-aggregation functional network (MAFN) exhibit significantly higher accuracy
In this work we demonstrate the advantages of MAFN in identifying three abnormal brain states, it can be applied to understand our healthy or diseased brain, such as detecting drivinginduced fatigue, Alzheimer’s disease, depression, and different emotions, etc

Summary

Introduction

Human brain is one of the most delicate and complex systems, responsible for maintaining the internal regulation of human body and perception, and responding to external stimuli. Constructing functional networks from the brain activities recorded with these technologies has attracted more and more attentions [10,11,12]. Increasing evidence showed that functional networks change with cognitive activities, emotion, and the development of brain diseases and so on [13, 14]. Functional networks can be applied to reveal different brain states or to detect brain illnesses whose neuropathology are not yet clear. Previous studies have found that, among others, the modules of the brain’s functional network become more isolated, and the connections within the modules become stronger when people age [15]. Functional networks of patients with schizophrenia, compared with healthy people, and exhibit abnormalities in multiple global indicators (global clustering coefficient, small-world-ness, etc.) [16]

Methods

Findings

Discussion

Conclusion