EOG artifact correction from EEG recording using stationary subspace analysis and empirical mode decomposition

Abstract

Ocular contamination of EEG data is an important and very common problem in the diagnosis of neurobiological events. An effective approach is proposed in this paper to remove ocular artifacts from the raw EEG recording. First, it conducts the blind source separation on the raw EEG recording by the stationary subspace analysis, which can concentrate artifacts in fewer components than the representative blind source separation methods. Next, to recover the neural information that has leaked into the artifactual components, the adaptive signal decomposition technique EMD is applied to denoise the components. Finally, the artifact-only components are projected back to be subtracted from EEG signals to get the clean EEG data. The experimental results on both the artificially contaminated EEG data and publicly available real EEG data have demonstrated the effectiveness of the proposed method, in particular for the cases where limited number of electrodes are used for the recording, as well as when the artifact contaminated signal is highly non-stationary and the underlying sources cannot be assumed to be independent or uncorrelated.

Figure 1.

Block diagram of the proposed approach.

Figure 2.

Placement of the EEG electrodes on the scalp according to the recording 10–20 system.

Figure 3.

Results of EOG artifact removal on an example EEG data set from the 20 artificially contaminated ones. (a) The artificially contaminated EEG data with eye movement and blink artifacts (shown in the first six EEG channels) and the corresponding EOG signals used in the mixing procedure (shown in the last two channels); (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The mixed, the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (g) The power spectrum estimates for the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (h) The zoomed artifacts-corrected signals in the artifact-free epoch.

Figure 3.

Results of EOG artifact removal on an example EEG data set from the 20 artificially contaminated ones. (a) The artificially contaminated EEG data with eye movement and blink artifacts (shown in the first six EEG channels) and the corresponding EOG signals used in the mixing procedure (shown in the last two channels); (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The mixed, the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (g) The power spectrum estimates for the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (h) The zoomed artifacts-corrected signals in the artifact-free epoch.

Figure 3.

Results of EOG artifact removal on an example EEG data set from the 20 artificially contaminated ones. (a) The artificially contaminated EEG data with eye movement and blink artifacts (shown in the first six EEG channels) and the corresponding EOG signals used in the mixing procedure (shown in the last two channels); (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The mixed, the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (g) The power spectrum estimates for the pre-contaminated EEG and the corrected EEG signals by each methods on channel Fp1; (h) The zoomed artifacts-corrected signals in the artifact-free epoch.

Figure 4.

Results of EOG artifact removal on the real EEG data. (a) The raw EEG recordings with eye movement and blink artifacts, as well as the simultaneous recordings of the EOG signals; (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The raw EEG recording and the corrected EEG signals by each methods on channel Fp1; (g) The zoomed artifacts-corrected signals in the artifact-free epoch on channel Fp1.

Figure 4.

Results of EOG artifact removal on the real EEG data. (a) The raw EEG recordings with eye movement and blink artifacts, as well as the simultaneous recordings of the EOG signals; (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The raw EEG recording and the corrected EEG signals by each methods on channel Fp1; (g) The zoomed artifacts-corrected signals in the artifact-free epoch on channel Fp1.

Figure 4.

Results of EOG artifact removal on the real EEG data. (a) The raw EEG recordings with eye movement and blink artifacts, as well as the simultaneous recordings of the EOG signals; (b) Components separated by SSA; (c) Components separated by SOBI; (d) Components separated by ICA; (e) IMFs decomposed by applying EMD on the last two components in (b); (f) The raw EEG recording and the corrected EEG signals by each methods on channel Fp1; (g) The zoomed artifacts-corrected signals in the artifact-free epoch on channel Fp1.