Fusion of Spatio-Temporal and Multi-Scale Frequency Features for Dry Electrodes MI-EEG Decoding
This work addresses the challenge of enabling fast, comfortable brain-computer interfaces for at-home and wearable use by improving decoding accuracy for dry-electrode EEG, though it appears incremental as it builds on existing methods like CNNs, Transformers, and graphs.
The paper tackled the problem of decoding motor imagery EEG from dry electrodes, which suffer from lower signal-to-noise ratio and greater data distribution shifts, by introducing the STGMFM framework that fuses spatio-temporal and multi-scale frequency features, achieving consistent improvements over competitive baselines on collected dry-electrode MI-EEG data.
Dry-electrode Motor Imagery Electroencephalography (MI-EEG) enables fast, comfortable, real-world Brain Computer Interface by eliminating gels and shortening setup for at-home and wearable use.However, dry recordings pose three main issues: lower Signal-to-Noise Ratio with more baseline drift and sudden transients; weaker and noisier data with poor phase alignment across trials; and bigger variances between sessions. These drawbacks lead to larger data distribution shift, making features less stable for MI-EEG tasks.To address these problems, we introduce STGMFM, a tri-branch framework tailored for dry-electrode MI-EEG, which models complementary spatio-temporal dependencies via dual graph orders, and captures robust envelope dynamics with a multi-scale frequency mixing branch, motivated by the observation that amplitude envelopes are less sensitive to contact variability than instantaneous waveforms. Physiologically meaningful connectivity priors guide learning, and decision-level fusion consolidates a noise-tolerant consensus. On our collected dry-electrode MI-EEG, STGMFM consistently surpasses competitive CNN/Transformer/graph baselines. Codes are available at https://github.com/Tianyi-325/STGMFM.