Steady-State Motion Visual Evoked Potential (SSMVEP) Enhancement Method Based on Time-Frequency Image Fusion

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Steady-state motion visual evoked potentials (SSMVEPs) are prone to noise and artifacts like EOG and EMG.
• Traditional spatial filtering methods for SSMVEP enhancement operate in the time domain.
• Improving the fusion of brain region information is crucial for SSMVEP active component enhancement.

Purpose of the Study:

• To propose and evaluate a novel SSMVEP enhancement method based on time-frequency (T-F) image fusion.
• To improve the enhancement effect on SSMVEP active components compared to traditional spatial filtering.
• To explore the feasibility of fusing SSMVEP data in the T-F domain.

Main Methods:

• Transformed two electrode signals from time to T-F domain using Short-Time Fourier Transform (STFT).
• Decomposed T-F signals (images) using 2D multiscale wavelet decomposition and fused coefficients.
• Reconstructed fused T-F images, applied mean filtering, and obtained the fused time-domain signal via inverse STFT.

Main Results:

• The proposed T-F image fusion method demonstrated a superior enhancement effect on SSMVEP active components.
• Experimental results confirmed the effectiveness of the T-F domain fusion approach.
• The method showed significant improvement over traditional time-domain spatial filtering techniques.

Conclusions:

• Time-frequency domain fusion is a feasible and effective approach for SSMVEP analysis.
• This study provides a new perspective for enhancing SSMVEP signals.
• The proposed method offers a promising alternative for improving the quality of EEG-based brain-computer interfaces.