Narrowband vs broadband phase synchronization analysis applied to independent components of ictal and interictal EEG

Gupta, D. and James, C.J. (2007) Narrowband vs broadband phase synchronization analysis applied to independent components of ictal and interictal EEG. In, 29th International Conference of IEEE Engineering in Medicine and Biology Society (EMBC2007), Lyon, France, 23 - 26 Aug 2007. Southampton, UK, Institute of Sound and Vibration Research4pp, 3864-3867.


Full text not available from this repository.


Phase synchronization analysis of electroencephalogram (EEG) and its Independent Components (IC) are presently being explored in the context of seizure onset prediction in epilepsy. The theory of phase synchronization analysis has been based on narrowband data and it is usually extended for use on broadband data, which may not be correct. This raises doubts about the applicability and interpretation of results of phase synchronization analysis on EEG signals or their ICs, which may be broadband signals. The current work shows that using phase synchronization analysis on broadband data may fail to gauge the changes in phase synchronization occurring only in specific frequency bands, as broadband phase synchronization has been observed to show an averaging effect of narrow band phase synchronies. This concept has been assessed in the context of seizure onset detection and prediction, where phase synchronization analysis in narrow band is found to be better performed than broadband analysis, in showing the trend of increasing or decreasing synchronization at seizure onset near epileptogenic area. This information is not always found to be consistent in analysis with the raw EEG signals, which may show spurious synchronization happening mainly due to volume conduction effects. Independent Component Analysis has been used to decompose the EEG into underlying least dependent sources, which in effect also removes the effect of volume conduction that is inherent in the scalp EEG. The analytical approach using the Hilbert Transform has been used for phase synchronisation analysis along with the Phase Locking Statistic (i.e. a measure of statistically significant phase synchronization). These observations lead us to believe that tracking changes in phase synchronization of narrow band activity, on ICs of continuous EEG data records will be of great value in the context of seizure prediction.

Item Type: Conference or Workshop Item (Paper)
Related URLs:
Subjects: Q Science > Q Science (General)
T Technology > TA Engineering (General). Civil engineering (General)
Divisions : University Structure - Pre August 2011 > Institute of Sound and Vibration Research > Signal Processing and Control
ePrint ID: 49439
Accepted Date and Publication Date:
Date Deposited: 13 Nov 2007
Last Modified: 31 Mar 2016 12:26

Actions (login required)

View Item View Item