Change detection in nonlinear biosignal processing by wavelets

Abstract

Biological information in the brain is carried by patterns of neural activity that are manifested in electrical fields of potential known as action potentials and in electroencephalographic (EEG) waves. Detecting changes of these waves is critical for an understanding of brain functions. Experiments suggest that most EEG and evoked potential phenomena are localized, distributed, and nonstationary. Here, the authors continue their previous work on wavelet application of biosignal processing published in recent IEEE and SPIE proceedings. They focus on exploration of some theoretical issues of change detection in nonlinear biosignal processing by wavelets and multiscale edge detection.