Spectrum and Phase Adaptive CCA for SSVEP-based Brain Computer Interface.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Steady State Visual Evoked Potential (SSVEP) based Brain Computer Interfaces (BCIs) offer fast information transfer and minimal training.
• Existing BCIs face challenges with subject and device variability, impacting accuracy and adaptability.

Purpose of the Study:

• To propose and evaluate a novel adaptive algorithm, Spectrum and Phase Adaptive CCA (SPACCA), for subject- and device-specific SSVEP-based BCIs.
• To enhance prediction accuracy by addressing cross-subject heterogeneity in spectrum distribution and accommodating response time lags.

Main Methods:

• Developed SPACCA, incorporating spectrum and phase adaptation for SSVEP detection.
• Designed a library of phase-shifting reference signals to manage individual and device-specific response time lags.
• Validated SPACCA using three datasets with LED, computer screen, and mobile device (tablet) as stimuli sources.

Main Results:

• SPACCA demonstrated consistent performance improvements over the standard CCA baseline across all tested datasets.
• The proposed method effectively handles variations in flickering sources, including LEDs, computer screens, and mobile devices.
• Statistical analysis confirmed a significant improvement with a p-value of 1.66e-6.

Conclusions:

• SPACCA offers a robust and adaptable solution for SSVEP-based BCIs, outperforming traditional methods.
• The algorithm's ability to adapt to subject and device variations makes it suitable for real-world, self-paced BCI applications.
• This advancement holds promise for more efficient and personalized brain-computer interface systems.