A Novel Real-time Algorithm Based on Phase-Locked Data Alignment for Continuously Controlled SSVEP-BCI.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Steady-state visual evoked potential (SSVEP) brain-computer interfaces (BCIs) offer high performance.
• Current SSVEP-BCI algorithms decode discrete EEG segments, limiting real-time continuous control.
• There is a need for algorithms that translate continuous EEG into control commands for enhanced BCI functionality.

Purpose of the Study:

• To propose a novel algorithm for SSVEP-BCI that enables real-time continuous control.
• To achieve real-time monitoring of user intentions through continuous EEG decoding.
• To enhance the compatibility of BCIs with human control habits.

Main Methods:

• Employed a phase synchronicity maximum strategy to capture SSVEP epochs aligned with template phases.
• Utilized a small-step sliding window update strategy for near real-time command recognition and output.
• Developed an SSVEP-BCI system with continuous stimulation for algorithm evaluation.

Main Results:

• The proposed algorithm successfully decoded continuously evoked SSVEP signals in nine subjects.
• Achieved an online average accuracy of 92.03% in the SSVEP-BCI system.
• Attained an information transfer rate (ITR) of 143.38 bits/min, demonstrating efficient continuous control.

Conclusions:

• The algorithm can theoretically decode SSVEP signals at any time, increasing command output density.
• High recognition accuracy is maintained while enabling continuous decoding of SSVEP.
• This research offers novel methods for real-time external device control via SSVEP-BCIs, promoting more intuitive human-computer interaction.