Development of a Brain-Computer Interface Toggle Switch with Low False-Positive Rate Using Respiration-Modulated...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Human-Computer Interaction

Background:

• Asynchronous electroencephalography (EEG)-based brain-computer interfaces (BCIs) often have low accuracy and high false-positive rates (FPR).
• Existing electrooculogram (EOG)-based BCI toggle switches can have high FPR or are unsuitable for patients with oculomotor impairments.

Purpose of the Study:

• To develop a novel, accurate, and reliable BCI toggle switch to enable on/off control of synchronous BCI systems.
• To overcome the limitations of current BCI toggle switch technologies, particularly high FPR and applicability to diverse patient groups.

Main Methods:

• Developed a BCI toggle switch utilizing photoplethysmography (PPG) signals, detecting breath-holding states.
• Employed linear discriminant analysis (LDA) to classify normal breathing versus breath-holding states using respiration-modulated PPG features.
• Integrated the PPG-based toggle switch with a steady-state visual evoked potential (SSVEP)-based BCI system for controlling external devices.

Main Results:

• Achieved 100% accuracy in activating the BCI by holding breath for approximately 10 seconds.
• Demonstrated a significantly low FPR of 0.02 false operations per minute, the lowest reported to date.
• All participants successfully controlled external devices using the synchronous BCI mode, confirming system usability.

Conclusions:

• The proposed photoplethysmography (PPG)-based BCI toggle switch offers a practical and highly effective solution for controlling synchronous BCI systems.
• This novel approach significantly improves BCI reliability by minimizing false positives and offers a viable alternative for users with oculomotor impairments.
• The system's real-time implementation with minimal calibration data suggests broad applicability in developing practical BCI solutions.