Removal of FMRI environment artifacts from EEG data using optimal basis sets

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) offers synergistic strengths for neuroscientific investigation.
• EEG data acquired during fMRI is contaminated by gradient artifacts and ballistocardiographic (BCG) artifacts from cardiac activity.
• Existing artifact removal methods often fall short, limiting the utility of combined EEG/fMRI.

Purpose of the Study:

• To develop and validate novel methods for removing gradient and BCG artifacts from simultaneous EEG/fMRI data.
• To improve the quality of EEG signals for more accurate neuroimaging analysis.
• To demonstrate the feasibility of high-quality simultaneous EEG/fMRI acquisition.

Main Methods:

• Temporal principal component analysis (PCA) was employed to capture and model the temporal variations of artifacts.
• Identified artifact basis functions were fitted to and subtracted from the EEG data.
• A robust algorithm was developed for accurate heartbeat peak detection from electrocardiographic (ECG) data for BCG artifact correction.

Main Results:

• The proposed methods significantly outperformed existing techniques in artifact removal.
• Artifact-free EEG data was successfully generated, preserving signal integrity.
• The methods proved effective even with a relatively low EEG sampling frequency (2048 Hz).

Conclusions:

• The developed methods provide a robust solution for artifact reduction in simultaneous EEG/fMRI.
• This advancement facilitates cleaner and more reliable data for neuroscience research.
• The findings support the practical implementation of simultaneous EEG/fMRI studies.