Information Technology Reference
In-Depth Information
Chapter 18
Analysis of Multichannel EEG Recordings
Based on Generalized Phase Synchronization
and Cointegrated VAR
Alla R. Kammerdiner and Panos M. Pardalos
Abstract
Synchronization is shown to be a characteristic feature of electroen-
cephalogram data collected from patients affected by neurological diseases, such
as epilepsy. Phase synchronization has been applied successfully to investigate syn-
chrony in neurophysiological signal. The classical approach to phase synchroniza-
tion is inherently bivariate. We propose a novel
multivariate
approach to phase syn-
chronization, by extending the bivariate case via cointegrated vector autoregression,
and then apply the new concept to absence epilepsy data.
keywords
Electroencephalogram, Phase synchronization, Cointegrated vector au-
toregressive processes
18.1 Introduction
The temporal integration of various functional areas in different parts of the brain
is believed to be essential for normal cognitive processes. Many studies stress a
significant role of neural synchrony in such large-scale integration [6, 42, 41, 43].
Specifically, it was discovered that oscillation of various neuronal groups in given
frequency bands leads to temporary phase-locking between such groups of neu-
rons. This discovery prompted the development of robust approaches that allow one
to measure the phase synchrony in a given frequency band from experimentally
recorded biomedical signals such as EEG.
