Abstract
In recent years, the usage of portable electroencephalogram (EEG) devices are becoming popular for both clinical and non-clinical applications. In order to provide more comfort to the subject and measure the EEG signals for several hours, these devices usually consists of fewer EEG channels or even with a single EEG channel. However, electrooculogram (EOG) signal, also known as eye-blink artifact, produced by involuntary movement of eyelids, always contaminate the EEG signals. Very few techniques are available to remove these artifacts from single channel EEG and most of these techniques modify the uncontaminated regions of the EEG signal. In this paper, we developed a new framework that combines unsupervised machine learning algorithm (k-means) and singular spectrum analysis (SSA) technique to remove eye blink artifact without modifying actual EEG signal. The novelty of the work lies in the extraction of the eye-blink artifact based on the time-domain features of the EEG signal and the unsupervised machine learning algorithm. The extracted eye-blink artifact is further processed by the SSA method and finally subtracted from the contaminated single channel EEG signal to obtain the corrected EEG signal. Results with synthetic and real EEG signals demonstrate the superiority of the proposed method over the existing methods. Moreover, the frequency based measures [the power spectrum ratio (Gamma ) and the mean absolute error (MAE)] also show that the proposed method does not modify the uncontaminated regions of the EEG signal while removing the eye-blink artifact.
Highlights
Electroencephalography (EEG) records brain electrical activity and it plays an important role in understanding of the motor functions, the cognitive loads, the level of attention and the brain disorders[1,2,3,4,5]
We observed sudden increase in the relative root mean square error (RRMSE) for Th > 1.4. It is noticed from the simulation analysis that the threshold (T h ) acts similar to a cut-off frequency as in classic filters and whereas the number of clusters L will act as the number of decomposition levels as in a wavelet decomposition
There is no significant difference in the CC values obtained for the proposed and the FBSE-EWT methods for the conditions p > 1, it can be clearly seen that the peaks of eye-blink component were clipped-off in the estimated eye-blink artifact obtained by the FBSE-EWT method
Summary
Electroencephalography (EEG) records brain electrical activity and it plays an important role in understanding of the motor functions, the cognitive loads, the level of attention and the brain disorders[1,2,3,4,5]. Portable EEG devices with single EEG channel are widely used to measure the brain signals in non laboratory/clinical applications[12,13] The use of these devices and their performance is studied in different applications such as BCI, driver fatigue detection and brain d isorders[14,15,16,17,18]. Canonical correlation analysis (CCA), another BSS based technique, is popular to remove eye-blink artifacts from the multichannel EEG s ignals[30,31]. The ensemble empirical mode decomposition (EEMD)[37] and ICA techniques are combined ( called EEMD-ICA), to separate the sources from single channel EEG signal using I CA38 In this method, the EEMD is employed to decompose single channel EEG signal into multi-variate data. This method can work only under the assumption of EEG stationarity, which may not hold for lengthy EEG epochs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
