Reference-free harmonic regression technique to remove EEG-fMRI ballistocardiogram artifacts.

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Signal Processing

Background:

• Simultaneous electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) offer high spatiotemporal resolution for studying brain states.
• Ballistocardiogram (BCG) artifacts, caused by cardiac pulsation and head motion in the magnetic field, significantly degrade EEG quality during fMRI.
• Existing BCG removal methods often rely on reference signals that can be corrupted or difficult to obtain.

Purpose of the Study:

• To develop and validate a novel model-based harmonic regression technique for removing BCG artifacts from simultaneous EEG-fMRI data.
• To provide a robust and reference-free method for artifact removal, enhancing the utility of EEG-fMRI.
• To improve the signal-to-noise ratio (SNR) and data quality for cognitive and clinical brain state research.

Main Methods:

• A model-based harmonic regression technique was developed using physically motivated parametric models for BCG artifacts and true EEG signals.
• Maximum likelihood approaches were employed to identify model parameters, estimate, and subtract BCG artifacts from corrupted EEG.
• The method was evaluated on its ability to remove artifacts, restore simulated signals, and improve SNR.

Main Results:

• The novel technique effectively removed ballistocardiogram artifacts from EEG data acquired during MR scanning.
• Simulated oscillatory signatures in the EEG were successfully restored after artifact removal.
• A significant improvement in signal-to-noise ratio (SNR), exceeding 20-fold in relevant frequency bands, was achieved.
• The reference-free nature of the method proved advantageous when reference signals were unavailable or compromised.

Conclusions:

• The proposed model-based harmonic regression technique offers an effective and robust solution for BCG artifact removal in simultaneous EEG-fMRI.
• This reference-free approach enhances the quality and reliability of EEG data acquired during fMRI, facilitating advanced brain activity research.
• The method holds significant potential for improving the study of cognitive and clinical brain states using high-resolution neuroimaging techniques.