Related Experiment Video
Updated: Sep 4, 2025
A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
Published on: July 7, 2023
Bidirectional Siamese correlation analysis method for enhancing the detection of SSVEPs
Xinyi Zhang1,2, Shuang Qiu1, Yukun Zhang1
1Research Center for Brain-Inspired Intelligence, National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, People's Republic of China.
The novel bidirectional Siamese correlation analysis (bi-SiamCA) model significantly improves steady-state visual evoked potential (SSVEP) detection accuracy for brain-computer interfaces (BCIs). This advanced method excels, particularly with limited data, paving the way for faster BCI applications.
Area of Science:
- Neuroscience and Biomedical Engineering
- Brain-Computer Interfaces (BCIs)
Background:
- Steady-state visual evoked potential (SSVEP)-based BCIs offer high information transfer rates.
- Improving SSVEP detection performance is crucial for advancing BCI technology.
Purpose of the Study:
- To introduce a novel bidirectional Siamese correlation analysis (bi-SiamCA) model for enhanced SSVEP detection.
- To improve the accuracy and efficiency of SSVEP-based BCIs, especially with short signal lengths.
Main Methods:
- Development of a long short-term memory-based Siamese architecture to measure SSVEP signal and template similarity.
- Implementation of a maximize agreement module with contrastive loss to enhance signal-template similarity.
- Utilizing a two-way signal processing mechanism for integrating temporal information and end-to-end training with raw SSVEPs.
Main Results:
- The bi-SiamCA model demonstrated significantly improved classification accuracy on 40-class and 12-class datasets compared to traditional and deep learning methods.
- Performance gains were particularly notable under short data length conditions.
- Feature visualizations confirmed enhanced similarity between SSVEP signals and reference signals within the same frequency.
Conclusions:
- The proposed bi-SiamCA model effectively enhances SSVEP detection performance.
- bi-SiamCA outperforms existing methods in SSVEP detection accuracy.
- The model's high decoding accuracy with short signals shows significant potential for high-speed BCI implementation.
