EEG Headset Evaluation for Detection of Single-Trial Movement Intention for Brain-Computer Interfaces

Area of Science:

• Neuroscience
• Rehabilitation Technology

Background:

• Brain-computer interfaces (BCIs) show promise for neurorehabilitation.
• Limited research exists on transferring BCI technology to clinical settings.
• Headset selection is crucial for effective BCI performance.

Purpose of the Study:

• To assess the efficacy of four commercial headsets for recording and classifying movement intentions.
• To evaluate the reliability of EEG recordings from these headsets for movement-related cortical potentials (MRCPs).

Main Methods:

• Twelve healthy participants performed 100 movements each.
• Continuous electroencephalography (EEG) was recorded using four different headsets over two days.
• Key metrics included single-trial classification accuracy, rejected epochs, and signal-to-noise ratio.

Main Results:

• Headsets positioned over the motor cortex yielded the highest classification accuracies (73%-77%).
• Moderate to good reliability was observed with the best-performing headset, a gel-based model.
• MRCPs were successfully recorded, demonstrating feasibility for BCI applications.

Conclusions:

• Reliable MRCP recording for BCIs necessitates headset channels close to the motor cortex.
• Gel-based headsets may offer superior performance for BCI-based neurorehabilitation.
• Further research is needed to optimize BCI hardware for clinical neurorehabilitation.