Wavelet-based artifact identification and separation technique for EEG signals during galvanic vestibular stimulation
1Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, BC, Canada. mani.adib@gmail.com
Computational and Mathematical Methods in Medicine
|August 20, 2013
Summary
We developed a new method to remove artifacts from electroencephalography (EEG) recordings during Galvanic Vestibular Stimulation (GVS). This technique improves the analysis of brain activity by effectively eliminating GVS-induced noise in EEG signals.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Signal Processing
Background:
- Galvanic Vestibular Stimulation (GVS) is used to study brain input.
- Electroencephalography (EEG) monitors brain activity during GVS.
- GVS current distribution causes artifacts in EEG signals, hindering analysis.
Purpose of the Study:
- To develop a novel method for removing GVS-induced artifacts from EEG signals.
- To enable clearer analysis of brain activity during GVS.
Main Methods:
- Combined time-series regression and wavelet decomposition to estimate GVS current contribution.
- Used wavelet transform to analyze EEG signals across frequency bands.
- Optimized the method using simulated signals and compared it with existing techniques.
Main Results:
- The proposed method demonstrated superior performance in GVS artifact removal compared to ICA-based methods, regression, and adaptive filters.
- Achieved a higher signal-to-artifact ratio of -1.625 dB.
Conclusions:
- The novel wavelet and regression-based method effectively removes GVS artifacts from EEG.
- This advancement facilitates more accurate analysis of brain responses to GVS.
