Reference-free reduction of ballistocardiogram artifact from EEG data using EMD-PCA

Concurrent electroencephalograph (EEG) and functional magnetic resonance image (fMRI) led researchers to acquire neuronal activities in detail over the past few decades. Regardless of the advantages of combining these modalities, artifacts posed a greater challenge to attain good quality data. One such problematic artifact which contaminates EEG recordings is Ballistocardiogram (BCG) artifact. A reference-free composite algorithm which combines empirical mode decomposition (EMD) and principal component analysis (PCA) named as EMD-PCA has been introduced in this study. The results show that the algorithm can efficiently reduce the BCG artifact by preserving original neuronal signals. The proposed algorithm showed improvement in reducing the BCG artifact as well as in the preservation of brain activities, when compared with two renowned existing methods that are average artifact subtraction (AAS) and optimal basis set (OBS).