作者: Stefan Debener , Alexander Strobel , Bettina Sorger , Judith Peters , Cornelia Kranczioch
DOI: 10.1016/J.NEUROIMAGE.2006.09.031
关键词:
摘要: EEG signals recorded simultaneously with fMRI are massively compromised by severe artefacts, among them the cardiac cycle-related ballistocardiogram (BCG) artefact. Different methods have been proposed to remove BCG artefact focusing on channel-wise template subtraction procedures or spatial filtering approaches such as independent component analysis (ICA). Here we systematically compared performance of optimal basis set (OBS), a correction approach, ICA and recently combination both (OBS–ICA). The three different were applied 60-channel data from 12 subjects during acquisition in 3-T scanner. In addition examination residual artefact, signal-to-noise ratio (SNR) topography resulting auditory evoked potential N1 compared. Whereas all led significant reduction, procedure resulted significantly reduced SNR amplitude when BCG-uncorrected data, indicating rather poor performance. contrast ICA, OBS OBS–ICA corrected substantially improved N1. quality was investigated means equivalent current dipole modelling. On descriptive level, localization error data. This improvement for OBS–ICA. We conclude that can efficiently artefacts improve inside scanner, prerequisite successful integration fMRI.