EEG-based mental fatigue assessment during driving by using sample entropy and rhythm energy

Two complexity parameters of EEG, i.e. sample entropy and rhythm energy are utilized to characterize the complexity and irregularity of EEG data under the different mental fatigue states. Then the wavelet transform and BP neural networks are combined to differentiate two mental fatigue states. The WT is employed to extract nonlinear features from the complexity parameters of EEG and improve the generalization performance of BPNN. The investigation suggests that sample entropy can effectively describe the dynamic complexity of EEG, which is strongly correlated with mental fatigue. Both complexity parameters are significantly decreased as the mental fatigue level increases. These complexity parameters may be used as the indices of the mental fatigue level. Moreover, the combined feature of rhythmic energy and sample entropy can attribute to higher classification accuracy of mental fatigue than one kind of feature alone. The proposed scheme could be a promising model for the estimation of mental fatigue.