An independent SSVEP-based brain-computer interface in locked-in syndrome.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Steady-state visually evoked potential (SSVEP)-based brain-computer interfaces (BCIs) offer communication for healthy individuals.
• Current SSVEP-BCIs rely on gaze control, limiting use for severely disabled patients.
• Existing gaze-independent SSVEP-BCIs exhibit reduced accuracy and haven't been validated in brain-injured populations.

Purpose of the Study:

• To propose and evaluate a novel gaze-independent SSVEP-BCI utilizing covert attention.
• To enhance classification accuracy by investigating feature extraction algorithms and harmonic analysis.
• To assess the online communication feasibility of this new BCI in healthy subjects and patients with locked-in syndrome (LIS).

Main Methods:

• Employed a covert, two-class SSVEP paradigm with a custom-designed portable LED-based 'interlaced squares' stimulation pattern.
• Recruited 24 healthy subjects and 6 LIS patients for offline and online testing.
• Analyzed the influence of different feature extraction methods and harmonic numbers on classification performance.

Main Results:

• Healthy subjects achieved mean offline accuracy of 85 ± 2% and online accuracy of 74 ± 13%.
• Eight out of twelve healthy subjects successfully communicated online with 80 ± 9% accuracy.
• Two LIS patients demonstrated above-chance offline accuracy, indicating command following; one patient achieved online communication.

Conclusions:

• Demonstrated the feasibility of online communication using a covert SSVEP paradigm, independent of neuromuscular functions.
• The developed BCI system shows potential as a diagnostic and communication tool for severely brain-injured patients.
• Further research is warranted to explore the clinical utility of this BCI for LIS and other brain-injured populations.