Removal of Ballistocardiogram Artifacts Using the Cyclostationary Source Extraction Method

Abstract

Ballistocardiogram (BCG) artifact is considered here as the sum of a number of independent cyclostationary components having the same cycle frequency. Our proposed method, called cyclostationary source extraction (CSE), is able to extract these components without much destructive effect on the background electroencephalogram (EEG). It is shown that the proposed method outperforms other methods particularly in preserving the remaining signals. CSE is utilized to remove the BCG artifact from real EEG data recorded inside the magnetic resonance (MR) scanner, i.e., visual evoked potential (VEP). The results are compared to the results of benchmark BCG removal techniques. Analyzing the power spectral density of the cleaned EEG data, it is shown that CSE effectively removes the frequency components corresponding to the BCG artifact. It is also shown that VEPs recorded inside the scanner and processed using the proposed method are more correlated with the VEPs recorded outside the scanner. Moreover, there is no need for electrocardiogram (ECG) data in this method as the cycle frequency of the BCG is directly computed from the contaminated EEG signals.

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

“…First, the EEG data acquired using 128-electrodes for the analysis of peak-to-peak value in time domain were grouped over 11 scalp regions as: Prefrontal (PF): (9,14,15,21,22), Anterior Frontal (AF): (2,3,12,16,23,26,34), Frontal (F): (1,4,11,14,21,15,19,24,27,32,33 (70, 75, 83). The difference between maximum peak and minimum peak in the EEG epochs based on R-R interval of ECG has been calculated and averaged per channel and then averaged over the electrodes per region.…”

“…The artifact which is more complex, dynamic and problematic to remove is the BCG artifact. The BCG artifact varies significantly within magnetic scanners, channels and individuals [9].…”

Cyclostationarity: New trends and applications

“…Recent applications are in the following subjects: analysis of genome signals [19], neuroscience [20], ballistocardiogram analysis [128], heart and lung sound separation [129], analysis of the embolic blood Doppler signal [132], foetal PQRST extraction from electrocardiogram (ECG) recordings [143], heart and respiration rates monitoring [183], heart sound signal selection [217], heart sound cancelation from lung sound [218], detection and characterization of a runner's fatigue [38,235], analysis of electromyographic signals [296], modeling and analysis of ground reaction force signals [97,300], and analysis of myoelectric signals [3].…”

“…Applications are considered in the following fields: acoustics and mechanics [10,14], radio astronomy and astrophysics [78,100,145], optics and spectroscopy [105,310,309], analysis of genome and biological signals [19,97,128,132,235], finance and econometrics [27,43,160,204], and climatology [84,125,151,190,308].…”

Separation and reconstruction of BCG and EEG signals during continuous EEG and fMRI recordings

“…Other widely used adaptive template approaches for BCG suppression such as Forbes and Fiume (2005) can be interpreted as weighted PCA to incorporate temporal model updates. Methods based on independent component analysis (ICA) (Srivastava et al, 2005; Ghaderi et al, 2010; Liu et al, 2012) also are used widely. All such blind source separation approaches, as reviewed in Grouiller et al (2007) and Vanderperren et al (2010), are limited to performing component extraction based on the contaminated data alone, agnostic of the structural difference between BCG and EEG.…”