An autonomous hybrid brain-computer interface system combined with eye-tracking in virtual environment.

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Biomedical Engineering

Background:

• Traditional brain-computer interface (BCI) systems require improvements in flexibility, robustness, and accuracy.
• Hybrid BCI systems offer a promising avenue for enhanced human-machine interaction.

Purpose of the Study:

• To develop an autonomous hybrid BCI system combining electroencephalogram (EEG) and eye-tracking for improved control in virtual environments.
• To enhance the flexibility, robustness, and accuracy of BCI systems.

Main Methods:

• An autonomous control strategy using a sliding window method for eye-gaze analysis was developed.
• A novel fusion method based on particle swarm optimization (PSO) was proposed for combining EEG and eye-gaze data.
• The PSO fusion method was compared against various other fusion techniques and machine learning algorithms.

Main Results:

• The proposed hybrid BCI system demonstrated superior performance compared to single-modality systems.
• The particle swarm optimization (PSO) fusion method achieved the best results among all evaluated fusion techniques.
• The system achieved higher accuracy and information transfer rate (ITR) in steady-state visual evoked potentials (SSVEP) tasks.

Conclusions:

• The developed autonomous control and dual-modal fusion methods significantly improve the performance of hybrid BCI systems.
• The hybrid BCI system offers enhanced flexibility, robustness, and classification accuracy.
• This approach represents a significant advancement in BCI technology for virtual environment control.