Analysis Of Fatigue Relief Functional Brain Network Based On Self-organized Criticality

Fatigue is a kind of subjective maladjustment,which easily leads to the loss of the ability to complete the normal activities or work.For people who drive continuously for a long time,with the accumulation of fatigue,physiological and psychological functions will produce some degree of imbalance,which will lead to traffic accidents.In order to reduce the risk index of fatigue driving,people have many prevention methods,among which sound stimulation,such as listening to radio,is the most effective prevention measure in the relevant research of the vehicle.However,the internal neural mechanism of alerting effect is still unknown.Electroencephalogram(EEG)is a real record of the changes of electric waves when the brain is active.It can directly reflect the activities of human brain.Using EEG,we can explore the physiological and psychological activities in the driving process.Therefore,the study of neural mechanism based on EEG is helpful to improve the effect of fatigue detection and the effective implementation of fatigue countermeasures.In this paper,we designed a control experiment of radio and normal driving simulation to explore the internal neural mechanism of alertness maintenance,and established a selforganized criticality model to complete the automatic calibration of fatigue critical point,which mainly studied from the following aspects:(1)The traditional functional brain network model is constructed,and the correlation degree of each node of EEG signal after preprocessing is calculated by using correlation function,Granger causality,phase locking value or phase lag index.The traditional functional brain network model fully shows the difference of the distribution of internal functional connections between vigilance maintenance and fatigue driving.In order to make further statistical analysis of the brain network,extract network features,such as: degree,clustering coefficient,feature path length and global efficiency.The results show that under the condition of maintaining vigilance,the network connectivity is more intensive,the network flexibility is better,and the information transmission efficiency is higher.(2)The dynamic characteristics of brain network are introduced into dynamics to construct self-organized criticality model.Calculate the comprehensive fatigue index of dynamic functional brain network,and complete the process of particle addition and particle collapse according to the degree distribution and network characteristics,find the critical point of fatigue.Finally,make statistical analysis on avalanche behavior to determine whether its probability distribution has self-organized criticality.After calculation,the comprehensive fatigue index shows a gradual upward trend,and the size and probability distribution of avalanche in each stage meet the power-law distribution,reflecting the self-organized criticality.Therefore,we can analyze the mechanism of neural network in the brain by analyzing the collapse mechanism and power-law distribution.(3)The behavior data is introduced,and the characteristics of the behavior data are extracted according to the related processing results of the EEG signal.The change trend of driving behavior in the long-time radio and normal driving process is analyzed,and the abnormal change of the behavior data near the fatigue critical point is observed.It is found that under the condition of maintaining vigilance,the dispersion and confusion of behavior data are lower and the driving safety index is higher.At the same time,near the critical point of fatigue,behavioral data fluctuated abnormally,which further verified the related research on the maintenance of alertness in the internal neurophysiology of brain.In a word,based on the analysis of self-organized criticality of fatigue relieving functional brain network,this paper uses the traditional functional brain network model and self-organized criticality model to study the internal neurophysiological mechanism of vigilance maintenance,and combines the behavioral data for statistical analysis to further study the vigilance maintenance.Research shows that under the condition of vigilance maintenance,the functional brain network is more closely connected,has higher ability to deal with events,and improves driving safety.At the same time,to verify the self-organized criticality of the functional brain network provides the possibility for the study of neurophysiological mechanism in the brain based on the maintenance of alertness.