Title :
Detection and localization of epileptic seizures using ICA
Author :
Sanei, Saeid ; Latif, Mohammad A. ; Shoker, Leor
Author_Institution :
Centre for Digital Signal Process. Res., King´´s Coll., London, UK
Abstract :
An effective and simple algorithm for localization of epileptiform sources of the EEG signals within the brain has been developed here. The method includes a constrained blind source separation (CBSS) algorithm, which attempts to separate and localize the epileptifom EEG sources. The significance of the new developed algorithm is the incorporation of a proper constraint into the joint approximate diagonalization formulation. The fixed locations of the electrodes together with the normalized correlations between the estimated sources and the observed EEGs provide a sufficient geometrical information to localize the sources.
Keywords :
blind source separation; diseases; electroencephalography; independent component analysis; medical signal detection; neurophysiology; CBSS algorithm; EEG signal; ICA; brain; constrained blind source separation; epileptic seizure detection; epileptic seizure localization; geometrical information; joint approximate diagonalization formulation; Blind source separation; Direction of arrival estimation; Electrodes; Electroencephalography; Epilepsy; Frequency; Independent component analysis; Signal processing; Signal processing algorithms; Working environment noise;
Conference_Titel :
Sensor Array and Multichannel Signal Processing Workshop Proceedings, 2004
Print_ISBN :
0-7803-8545-4
DOI :
10.1109/SAM.2004.1502931