An Adaptive SSVEP-Based Brain-Computer Interface to Compensate Fatigue-Induced Decline of Performance in Practical...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Steady-state visual evoked potential (SSVEP) brain-computer interfaces (BCIs) offer communication solutions for individuals with motor and speech impairments.
• Eye fatigue from low-frequency stimuli degrades SSVEP BCI performance over time.
• High-frequency stimuli (>25 Hz) can mitigate eye fatigue but may impact BCI performance.

Purpose of the Study:

• To quantitatively assess the performance decrease of high-frequency SSVEP BCIs over time.
• To develop and evaluate adaptive algorithms for compensating performance degradation.
• To establish a robust SSVEP BCI communication system for practical use.

Main Methods:

• Implementation of an asynchronous spelling system using a threshold-based LASSO algorithm for frequency recognition.
• Conducting long online experiments with participants typing sentences using the BCI system.
• Comparing fixed thresholds versus two proposed adaptive algorithms (tolerance-based and relational) to assess typing speed and accuracy.

Main Results:

• Fixed thresholds led to a significant decrease in typing speed over time across all subjects.
• Proposed adaptive algorithms maintained consistent typing speed without significant degradation.
• The adaptive relational algorithm achieved a higher average typing speed (8.7 char/min) compared to the tolerance-based algorithm (8.1 char/min) and fixed thresholds (>6 char/min).

Conclusions:

• Adaptive algorithms effectively compensate for fatigue-induced performance decline in high-frequency SSVEP BCIs.
• The proposed relational algorithm offers a robust solution for practical BCI communication systems.
• This research enhances the usability and reliability of SSVEP BCIs for individuals with disabilities.