A novel method for extracting time-varying rhythms of electroencephalography via wavelet packet analysis

A new method for detecting time-varying rhythms of nonstationary electroencephalogram is proposed in this paper. Multiresolution decomposition is used to investigate the transition of clinical EEG signals. Wavelet packet transformation is used to design the filters with different frequency characteristics in order to extract different kinds of dynamic EEG rhythms. Several actual EEG signals with different brain function states are tested and analyzed. The parameters of the wavelet packet transformation corresponding to the rhythms are developed to construct the time-varying electrical brain activity mapping (TVBEAM). From the experimental results, the dynamic characteristics of clinical brain electrical activities can be demonstrated by using wavelet packet decomposition. The method presented in this paper can be used as a new way for the analysis of other biomedical signals.