Effect of electrocardiogram interference on cortico-cortical connectivity analysis and a possible solution.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Electroencephalogram (EEG) signals are frequently corrupted by electrocardiogram (ECG) interference.
• This ECG contamination hinders accurate quantitative analysis and characterization of EEG data.

Purpose of the Study:

• To introduce and validate a novel frequency-based method, null-coherence, for attenuating ECG interference in EEG.
• To compare the efficacy of the null-coherence approach against independent component analysis (ICA) for ECG artifact removal in EEG.

Main Methods:

• The null-coherence method utilizes simultaneously recorded ECG as a reference signal to remove interference.
• The approach was validated using simulated data and subsequently applied to EEG recordings from newborns undergoing therapeutic hypothermia.
• Comparative analysis involved assessing power spectrum and cortico-cortical connectivity before and after artifact removal using both null-coherence and ICA.

Main Results:

• The null-coherence approach successfully attenuated ECG contamination in EEG signals.
• No residual ECG interference was detected in the EEG after applying the null-coherence method.
• The null-coherence method demonstrated superior performance compared to ICA by not enhancing cortico-cortical connectivity.

Conclusions:

• The null-coherence method effectively removes ECG interference from EEG without introducing redistribution artifacts, unlike ICA.
• While both methods attenuate ECG contamination, null-coherence preserves EEG cortico-cortical connectivity better than ICA.
• Null-coherence is proposed as a superior method for ECG interference attenuation in EEG, particularly for cortico-cortical connectivity analysis.