A Practical Hybrid Brain–Computer Interface Combining Eye Tracking and SSVEP for Real-World Applications

Background: Steady-state visual evoked potential (SSVEP)–based brain-computer interfaces (BCIs) have a wide range of applications in fields such as healthcare and automation control. However, these systems are limited by low encoding efficiency, and their reliance on individualized calibration data results in prolonged user preparation. A hybrid system that integrates eye-tracking technology provides a potential approach to address these challenges. Methods: In this study, we developed a hybrid BCI system that integrates an eye-tracking interface (ETI) based on webcam input with an SSVEP-based BCI system. Ten participants were involved in offline experiments comparing the three systems and subsequently provided subjective evaluations. Four healthy individuals and one patient with impaired hand function were recruited in online experiments utilizing the hybrid BCI system. Results: The hybrid BCI system achieved the highest ITR, the shortest training time, and the highest score in a comprehensive evaluation model. The average classification accuracy among healthy participants was 85.70%, with an information transfer rate of 260.07 bits/min. The patient successfully completed the online tasks, achieving an accuracy of 81.82%. Conclusions: By integrating eye-tracking technology, the usability of the BCI system was significantly improved, enabling participants without prior BCI experience to achieve spelling accuracies comparable to those of experienced users.