Subject(s): CX.41, CX.32

Category: Brief Article

Abstract:

A new approach based on orthogonal transformation and analytic signal analysis is utilized to characterize the hidden dynamical characteristics of EEG signal. Alpha band EEG signal is decomposed into relatively orderly (in terms of periodicity) and disorderly cyclical components using singular value decomposition based p-spectrum with moving windowed periodicity detection, followed by periodic decomposition. Normal, maniac and seizure subjects are studied. An analytic signal based approach is used for the characterization of the underlying functional structure of the brain. It is found that the normalized probability distributions of the amplitudes of the analytic signals for healthy brain show universal data collapsing phenomena (or homogeneous scaling behaviour) which is a typical characteristic of physical systems at the critical point of phase transitions; for pathological subjects lack of universality (or heterogeneous scaling) is observed.

Retrieve Manuscript

Retrieve Previous Revision's Abstract

Submit referee report/comment