Temporally Local Weighting-Based Phase-Locked Time-Shift Data Augmentation Method for Fast-Calibration SSVEP-BCI

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Training-based spatial filtering methods are crucial for decoding steady-state visual evoked potentials (SSVEPs).
• Current methods demand extensive calibration data, limiting the practicality of SSVEP-based brain-computer interfaces (BCIs).
• Reducing calibration time is essential for advancing BCI technology.

Purpose of the Study:

• To introduce a novel data augmentation method, temporally local weighting-based phase-locked time-shift (TLW-PLTS).
• To enhance the efficiency of calculating spatial filters and temporal templates for SSVEP decoding.
• To enable faster calibration for SSVEP-based BCIs.

Main Methods:

• The TLW-PLTS method utilizes a sliding window strategy with SSVEP response periods for data augmentation.
• A time filter is employed to maximize temporal covariance and suppress noise in augmented data.
• The TLW-PLTS method was integrated with existing spatial filtering techniques for evaluation.

Main Results:

• The TLW-PLTS method demonstrated superior decoding performance compared to state-of-the-art methods.
• Significantly fewer calibration data were required when using TLW-PLTS.
• Improved classification accuracies and information transfer rates (ITRs) were observed across three SSVEP datasets.

Conclusions:

• The TLW-PLTS data augmentation method offers a promising solution for reducing calibration time in SSVEPs.
• This advancement facilitates the development of more practical and efficient brain-computer interfaces.
• The findings support the potential for rapid-calibration BCIs using advanced data augmentation techniques.