Analyses Of Mental Fatigue Monitored Based On Eeg Spectrum-related Features

With the increasingly fierce social competition,and increasingly fast pace of life,mental fatigue has become a common phenomenon,affecting people's work and learning,and even lead to a variety of accidents.The objective evaluation index of mental fatigue can effectively monitor the state of mental fatigue and reduce the occurrence of accidents.Magnetic stimulation at acupoints is a combination of magnetic stimulation and traditional Chinese Medicine,with which may regulate the nervous system and relieve the effect of mental fatigue.In this thesis,the mental fatigue model was built aiming at the fatigue caused by a long-time heavy load mental activity.The 64 spontaneous EEG signals were extracted before and after mental fatigue state and after magnetic stimulation state.The EEG rhythms was extracted by wavelet packet decomposition,and the power spectrum analysis of EEG was performed using the fast Fourier transform based on sliding window.And the difference of the spectrum-related features of EEG before and after mental fatigue and the effect of magnetic stimulation at acupoints on relieving mental fatigue were researched.The main work of this thesis is as follows:(1)The EEG experiment was designed and the EEG data were collected and processed.The mental fatigue state was assessed by subjective rating scales and the results were statistically analyzed.The results showed that subjective rating scales were significantly different from before and after mental fatigue,and there was significant difference between before and after magnetic stimulation at acupoints.(2)There were significant differences in the spectrum-related features of EEG before and after mental fatigue: the dominant frequencies and barycenter frequencies of all electrodes were decreased,the relative power in ? and ? rhythm were increased,but the relative power in ? and ? rhythm were decreased after mental fatigue.So the spectrum-related features of EEG can reflect the degree of mental fatigue,and may hopefully become the objective index for monitoring the mental fatigue degree.There were significant differences in the spectrum-related features of EEG before and after magnetic stimulation: the dominant frequencies and barycenter frequencies of all electrodes were increased,the relative power in ? and ? rhythm were decreased,but the relative power in ? and ? rhythm were increased after magnetic stimulation.Therefore,magnetic stimulation at acupoints can effectively relieve mental fatigue.(3)The spectrum-related features of EEG were used as classification features.Then the classification of mental fatigue was studied using support vector machine classification algorithm.The optimal combination of spectrum-related features of EEG and the best combination of spontaneous were obtained,that is,the dominant frequency,barycenter frequency,the relative power in ? and ? rhythm of the 1-30 spontaneous were more effective to monitor the state of mental fatigue.