ITEM METADATA RECORD
Title: Improving MEG source localizations: an automated method for complete artifact removal based on independent component analysis
Authors: Mantini, Dante ×
Franciotti, Raffaella
Romani, Gian Luca
Pizzella, Vittorio #
Issue Date: Mar-2008
Series Title: NeuroImage vol:40 issue:1 pages:160-73
Abstract: The major limitation for the acquisition of high-quality magnetoencephalography (MEG) recordings is the presence of disturbances of physiological and technical origins: eye movements, cardiac signals, muscular contractions, and environmental noise are serious problems for MEG signal analysis. In the last years, multi-channel MEG systems have undergone rapid technological developments in terms of noise reduction, and many processing methods have been proposed for artifact rejection. Independent component analysis (ICA) has already shown to be an effective and generally applicable technique for concurrently removing artifacts and noise from the MEG recordings. However, no standardized automated system based on ICA has become available so far, because of the intrinsic difficulty in the reliable categorization of the source signals obtained with this technique. In this work, approximate entropy (ApEn), a measure of data regularity, is successfully used for the classification of the signals produced by ICA, allowing for an automated artifact rejection. The proposed method has been tested using MEG data sets collected during somatosensory, auditory and visual stimulation. It was demonstrated to be effective in attenuating both biological artifacts and environmental noise, in order to reconstruct clear signals that can be used for improving brain source localizations.
ISSN: 1053-8119
Publication status: published
KU Leuven publication type: IT
Appears in Collections:Research Group Neurophysiology
× corresponding author
# (joint) last author

Files in This Item:
File Description Status SizeFormat
Mantini 2008 Neuroimage.pdf Published 2828KbAdobe PDFView/Open

 


All items in Lirias are protected by copyright, with all rights reserved.

© Web of science