Open AccessA Novel Constrained Topographic Independent Component Analysis for Separation of Epileptic Seizure Signals
Author(s) -
Min Jing,
Saeid Sanei
Publication year - 2007
Publication title -
computational intelligence and neuroscience
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.605
H-Index - 52
eISSN - 1687-5273
pISSN - 1687-5265
DOI - 10.1155/2007/21315
Subject(s) - independent component analysis , blind signal separation , computer science , source separation , signal (programming language) , electroencephalography , pattern recognition (psychology) , epileptic seizure , interference (communication) , artificial intelligence , separation (statistics) , component (thermodynamics) , signal processing , algorithm , speech recognition , machine learning , neuroscience , psychology , physics , telecommunications , channel (broadcasting) , radar , thermodynamics , programming language
Blind separation of the electroencephalogram signals (EEGs) using topographic independent component analysis (TICA) is an effective tool to group the geometrically nearby source signals. The TICA algorithm further improves the results if the desired signal sources have particular properties which can be exploited in the separation process as constraints. Here, the spatial-frequency information of the seizure signals is used to design a constrained TICA for the separation of epileptic seizure signal sources from the multichannel EEGs. The performance is compared with those from the TICA and other conventional ICA algorithms. The superiority of the new constrained TICA has been validated in terms of signal-to-interference ratio and correlation measurement.