o-CLEAN: a novel multi-stage algorithm for the ocular artifacts' correction from EEG data in out-of-the-lab...

Area of Science:

• Neuroscience
• Bioengineering
• Signal Processing

Background:

• Ocular artifacts, such as blinks, significantly confound electroencephalographic (EEG) signal processing.
• Removing these artifacts is challenging, particularly for wearable EEG systems used in out-of-lab settings with limited channels and no electrooculography (EOG).

Purpose of the Study:

• To validate a novel, multi-stage ocular blink artifact correction method named o-CLEAN (multi-stage OCuLar artEfActs deNoising).
• To assess o-CLEAN's suitability for online processing using minimal EEG channels and without EOG templates.

Main Methods:

• The o-CLEAN method was evaluated using EEG datasets from both controlled and real-world environments.
• Performance was compared against state-of-the-art techniques, assessing artifact correction (IN-Blink) and EEG signal preservation (OUT-Blink).

Main Results:

• o-CLEAN demonstrated reliability comparable to established methods for ocular blink artifact correction.
• The algorithm exhibited superior EEG signal preservation, particularly when using a reduced number of EEG channels.

Conclusions:

• The validated o-CLEAN method provides an effective solution for correcting ocular artifacts in EEG signals.
• It is particularly beneficial for out-of-the-lab and online applications using wearable systems with limited channels, addressing a key challenge in applied neuroscience.