Non-invasive EEG based mental state identification using nonlinear combination

Non-invasive EEGs are very useful in human-machine integration development and medical diagnosis. Mental state, especially mental fatigue, is one of the main causes of the tragic accidents. In order to prevent accidents caused by mental fatigue, it is crucial to identify such mental state. Based on the mental state, the human-machine systems would obtain beneficial effects for reducing their accident rate. Using non-invasive EEG recordings, the features of EEG are extracted based on nonlinear combination among EEG four frequency components. The index of mental state can be represented by a polynomial equation. The method is more flexible and provides a quantitative analysis way to acquire the more accurate mental state. The effectiveness of the method is well demonstrated through experimental results.