Application of bispectral techniques in the analysis of EEG signals
Hui Li · Zhonghua xingwei yixue yu naokexue zazhi · 2004
ObjectiveTo detect the high-order information in EEG signals by bispectral analysis based on high-order spectral techniques. MethodsAccording to the non-Gaussianilty and nonlinearity of EEG, the bispectral techniques is used to analyze EEG signals in different brain states (rest with eyes closed, rest with eyes open, computing by heart with eyes closed) and in schizophrenia. ResultsAll the EEG signals in different brain states show obvious bispectral structure, and there are differences between different brain states.ConclusionBeing a quantitative method of non-Gaussianilty and nonlinearity for EEG study, the bispectral analysis can give more useful high-order information.