EEG-fMRI Gradient Artifact Correction by Multiple Motion-Related Templates.

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Signal Processing

Background:

• Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) is powerful for studying brain activity.
• MR scanner magnetic field gradients induce significant artifacts in EEG signals, particularly with subject motion.
• Existing artifact correction methods like averaged artifact subtraction (AAS) and optimal basis sets (OBS) have limitations.

Purpose of the Study:

• To develop and evaluate a novel method for correcting gradient artifacts in EEG-fMRI recordings.
• To improve EEG signal quality in the presence of subject motion during simultaneous EEG-fMRI.
• To compare the effectiveness of the new method against established artifact correction techniques.

Main Methods:

• Gradient artifacts were modeled using multiple templates derived from head position information and motion time courses.
• Artifact correction was performed by estimating artifactual templates modulated by splines and head position data.
• EEG signal quality was assessed by comparing the new method with AAS and OBS using root-mean-square power.

Main Results:

• The proposed template-based method significantly outperformed AAS and OBS in artifact correction, as measured by reduced root-mean-square power.
• Improvements were most notable in posterior EEG channels, which typically show the most residual artifacts.
• The corrected EEG signals achieved spectral power comparable to EEG recorded without fMRI.

Conclusions:

• Gradient artifacts in EEG-fMRI can be effectively modeled and removed using multiple templates derived from head position information.
• This advanced artifact correction technique enhances the feasibility of EEG-fMRI studies, especially in subjects with inevitable motion.
• The method facilitates the investigation of high-frequency EEG activity, where gradient artifacts are particularly challenging.