An Eye Blinking Artifact Rejection Method Using Single-Channel Electroencephalographic Signals and 2 Step Non-Negative Matrix Factorization

Many biological signal analyses have been proceeded in various research fields. In particular, electroencephalographic (EEG) signal analyses have attracted attention because the EEG signal includes a mixture of endogenous brain activities. However, the EEG signal is often mixed physiological artifacts such as the eye blinking, oculogyration, heart beat, or muscle activity. Specifically, humans are unable to keep gazing at something without the eye blinking. In other words, the eye blinking artifacts absolutely invade the EEG signals while a subject wears an EEG device. Therefore, it is important to remove the artifacts from the EEG signals when researchers attempt to analyze the brain activities accurately. An independent component analysis (ICA) has been used for removing the eye blinking artifacts effectively with over 90% of the reconstruction for original EEG signals. However, a drawback of the ICA is that this method can only manage overdetermined mixtures, which entails many EEG electrodes. There is no numerical approach of eye blinking artifact rejection method for single-channel EEG signals. Therefore, in this paper, we propose an eye blinking artifact rejection method using single-channel EEG signals and 2 step non-negative matrix factorization (NMF). We acquired 14 EEG and 1 vertical electrooculographic signals from 10 subjects who blink every 5 seconds. Furthermore, we performed the proposed method to reject the eye blinking artifacts using single-channel EEG (Fp1) signals and the ICA using multichannel EEG signals to prepare the target for comparison. High SNR between reconstructed signals by the ICA and the proposed method was represented. Moreover, over 99% of the reconstruction for original EEG signals was showed by the proposed method. Therefore, we confirmed the validity of the proposed method for the eye blinking artifact rejection method using only single-channel EEG signals.