Development of a hybrid mental speller combining EEG-based brain-computer interface and webcam-based eye-tracking

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Biomedical Engineering

Background:

• Brain-computer interfaces (BCIs) offer communication pathways for individuals with severe motor impairments.
• Existing BCIs, like those using steady-state visual evoked potentials (SSVEP), can be prone to errors.
• Eye-tracking technology provides complementary spatial information that can potentially improve BCI accuracy.

Purpose of the Study:

• To develop and evaluate a novel hybrid mental spelling system.
• To integrate SSVEP-based BCI with eye-tracking for enhanced accuracy.
• To reduce the need for manual error correction in BCI-assisted communication.

Main Methods:

• A hybrid system was developed combining SSVEP BCI and webcam-based eye-tracking.
• Character selection via SSVEP was validated by eye gaze direction (left/right).
• Online experiments were conducted with ten healthy participants typing English words.

Main Results:

• The hybrid system prevented an average of 16.6 typing errors per participant.
• The integration of eye-tracking data significantly reduced misspellings.
• Participants did not require the 'BACKSPACE' key for error correction during the experiment.

Conclusions:

• The developed hybrid mental spelling system demonstrates enhanced practicality and accuracy.
• Combining SSVEP BCI with eye-tracking is a feasible approach to improve BCI performance.
• This hybrid BCI technology holds promise for more efficient and user-friendly assistive communication.