A sub-region combination scheme for spatial coding in a high-frequency SSVEP-based BCI
Ruochen Hu1, Gege Ming2,3, Yijun Wang2,3,4
1Department of Biomedical Engineering, Tsinghua University, Beijing, People's Republic of China.
Journal of Neural Engineering
|July 19, 2023
Summary
This study developed a novel spatial coding scheme for high-frequency steady-state visual evoked potential (SSVEP) brain-computer interfaces (BCIs). The findings demonstrate that sub-region SSVEP responses can predict joint region responses, enabling effective BCI operation.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Human-Computer Interaction
Background:
- Spatial coding is crucial for enhancing steady-state visual evoked potential (SSVEP) based brain-computer interfaces (BCIs).
- Understanding the relationship between local and global visual field stimulation is key to optimizing SSVEP responses.
- Existing methods may not fully leverage spatial information for BCI performance.
Purpose of the Study:
- To develop and evaluate a novel sub-region combination scheme for spatial coding in high-frequency SSVEP-BCIs.
- To investigate if SSVEPs generated by stimulating individual sub-regions can predict responses from combined joint regions.
- To assess the performance of a spatially-coded BCI paradigm using this scheme.
Main Methods:
- An annular visual field was divided into eight sub-regions, with 60 Hz visual stimuli presented to individual and combined regions.
- SSVEP responses from sub-regions were superimposed to simulate responses from joint regions.
- A four-class spatially-coded BCI paradigm was used for offline and online performance evaluation.
Main Results:
- The proposed scheme successfully implemented a spatially-coded visual BCI with satisfactory performance and imperceptible flicker.
- Offline analysis showed classification accuracy of 89.69 ± 8.75% and an information transfer rate (ITR) of 24.35 ± 7.09 bits/min with 3s data length.
- Online BCI system achieved an average accuracy of 87.50 ± 9.13% and an ITR of 22.48 ± 6.71 bits/min with 3s data length.
Conclusions:
- The study validates the feasibility of predicting joint region SSVEP responses from sub-region stimulations.
- This spatial coding approach offers potential for extending to other frequency bands and developing more complex BCI paradigms.
- The findings lay the groundwork for future advancements in SSVEP-based BCI technology.
