Sources of non-physiologic noise in simultaneous EEG-fMRI data: a phantom study

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Signal Processing

Background:

• Simultaneous electroencephalography-functional magnetic resonance imaging (EEG-fMRI) is a powerful technique for studying brain activity.
• Understanding and mitigating noise artifacts in EEG data acquired during fMRI is crucial for data integrity.
• Non-physiologic noise from the MRI environment can significantly contaminate EEG signals.

Purpose of the Study:

• To develop and demonstrate a general methodology for characterizing non-physiologic noise in EEG signals within an MRI environment.
• To identify specific noise sources impacting different EEG frequency bands.
• To provide a foundation for effective artifact removal strategies in simultaneous EEG-fMRI.

Main Methods:

• A phantom approach was employed to simulate and measure EEG signals in the presence of MRI noise.
• The methodology was validated using a specific MR scanner and EEG data acquisition system.
• Noise characteristics were analyzed across different EEG frequency bands (δ, β, γ).

Main Results:

• The δ frequency band was found to be significantly impacted by baseline drift and associated correction algorithms.
• The β and γ frequency bands were affected by residual gradient artifacts and gradient correction procedures.
• The developed methodology successfully characterized non-physiologic noise in the EEG signal.

Conclusions:

• A robust phantom-based methodology for characterizing non-physiologic EEG noise in MRI environments has been established.
• Different EEG frequency bands are susceptible to distinct types of MRI-related artifacts.
• This characterization is essential for optimizing pre-processing pipelines in simultaneous EEG-fMRI studies.