A robust and reliable online P300-based BCI system using Emotiv EPOC + headset

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Brain-computer interface (BCI) systems facilitate interaction without muscular activity by analyzing brain signals.
• The P300 wave, a specific brain response, is crucial for developing effective BCI communication tools.

Purpose of the Study:

• To evaluate the Emotiv EPOC+ headset's capability in detecting and recording P300 waves.
• To assess the impact of signal preprocessing on P300 detection accuracy.
• To determine the feasibility of using this affordable headset for online P300-based BCI systems.

Main Methods:

• Five participants used the Emotiv EPOC+ headset to record EEG data while performing a P300 speller task.
• EEG data were wirelessly transmitted to OpenViBE software for real-time processing.
• Two classifiers, Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM), were employed to analyze the P300 signals.

Main Results:

• The Emotiv EPOC+ headset successfully detected P300 signals.
• High accuracies were achieved: up to 90% with LDA and 70% with SVM after only two training sessions.
• Signal preprocessing was found to influence the acquired EEG data.

Conclusions:

• The Emotiv EPOC+ headset is a capable tool for detecting P300 waves.
• This portable and affordable technology can support the development of robust and reliable online P300-based BCI systems.
• The findings suggest broader accessibility for BCI technology in various applications.