Nonlinear dynamics of EEG in Alzheimers disease

Abstract

Nonlinear dynamical analysis has been widely applied to a variety of physiological data for last two decades. One of its major contributions is to the electroencephalogram (EEG) in Alzheimer's disease (AD). A number of studies using nonlinear dynamical methods have shown the globally decreased complexity of EEG patterns in AD patients. A prominent decrease in information transmission among cortical areas quantified by information-theoretic measures like mutual information is also found. These findings indicate decreased nonlinear processes underlying the EEG in AD. This review article focuses on nonlinear EEG abnormalities in AD patients obtained from nonlinear methods and their clinical implications. We suggest that nonlinear dynamical analysis may contribute to a deeper understanding of the neuropathological mechanism of AD in ways that are not possible by conventional spectral analysis. (C) 2002 Wiley-Liss, Inc.