Research On Classification Of Mental Fatigue State Based On Brain Functional Network Characteristics Of EEG

At present,mental fatigue has become a healthy problem which is common and easily ignored among modern people.It seriously impairs people's cognitive and behavioral ability for a long time during mental fatigue state,and it affects people's physical and mental health.Therefore,it is very important to understand the mechanism of mental fatigue,identify and explore the relevant methods to relieve mental fatigue.Firstly,the electroencephalogram(EEG)experiment was designed,and brain functional network in different states was constructed and compared.In this paper,the mental fatigue model of a sustained cognitive task was built.The 64 spontaneous EEG signals in resting state,mental fatigue state and magnetic stimulation state were extracted.Brain functional network in resting state,mental fatigue state and magnetic stimulation state was constructed and comparatively studied.The results showed that,compared with resting state,the complexity of brain functional network in mental fatigue state was significantly decreased,and the randomness was obviously enhanced.Compared to mental fatigue state,the complexity of brain functional network in magnetic stimulation state was significantly enhanced,at the same time,the correlation degree of brain regions was also greatly improved.Secondly,the global and local characteristic parameters of brain functional network in resting state,mental fatigue state and magnetic stimulation state were calculated.The effect of mental fatigue on brain connectivity was studied,the mechanism of mental fatigue and the effects of magnetic stimulation at acupoints on brain functional network during mental fatigue was explored.Based on complex network theory,the global and local characteristic parameters of brain functional network in resting state,mental fatigue state and magnetic stimulation state were calculated and quantitatively analyzed.At the same time,subjective evaluation method was used to compare the three kinds of states.The results showed that the global and local characteristic parameters could both well reflect the changes of brain connectivity caused by mental fatigue.They could be used to help us understand the mechanism of mental fatigue.And magnetic stimulation at acupoints can effectively relieve mental fatigue.Finally,in order to investigate the objective indicator which was used to define mental fatigue state,the classification of resting state and mental fatigue state was studied.In this paper,the global and local characteristic parameters were used as classification features respectively.Then it classified all subjects between resting state and mental fatigue state by support vector machine(SVM)algorithm,and the accuracy rate of the classification of these two methods were analyzed comparatively.The results showed that,compared with the global characteristic parameters of brain functional network which served as the classification features,the accuracy rate of the classification when local characteristic parameters were used as the classification features was higher.The local characteristic parameters of brain functional network were more suitable to be used as the classification features for classifying resting state and mental fatigue state.And it is expected to be used as an objective indicator to evaluate mental fatigue state.