Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SDEMG: Score-based Diffusion Model for Surface Electromyographic Signal Denoising

Yu-Tung Liu, Kuan-Chen Wang, Kai-Chun Liu, Sheng-Yu Peng, Yu Tsao

TL;DR

This work addresses ECG contamination in sEMG recordings, which compromises clinical analysis and prosthetic control. It introduces SDEMG, a conditional score-based diffusion model that denoises sEMG by conditioning the reverse diffusion process on ECG-polluted inputs, with a two-stage training and sampling framework. The model uses a two-stream architecture with Half Normalized Filter blocks and FiLM-conditioned bridges, trained with a cosine schedule and noise-scale conditioning, and evaluated on the NINAPro DB2 dataset with MIT-BIH NSRD ECG interference. Results show that SDEMG yields higher signal quality and more faithful feature preservation than HP, TS, and FCN baselines across a range of SNRs and mismatch conditions, suggesting strong practical potential for clinical sEMG analysis.

Abstract

Surface electromyography (sEMG) recordings can be influenced by electrocardiogram (ECG) signals when the muscle being monitored is close to the heart. Several existing methods use signal-processing-based approaches, such as high-pass filter and template subtraction, while some derive mapping functions to restore clean sEMG signals from noisy sEMG (sEMG with ECG interference). Recently, the score-based diffusion model, a renowned generative model, has been introduced to generate high-quality and accurate samples with noisy input data. In this study, we proposed a novel approach, termed SDEMG, as a score-based diffusion model for sEMG signal denoising. To evaluate the proposed SDEMG approach, we conduct experiments to reduce noise in sEMG signals, employing data from an openly accessible source, the Non-Invasive Adaptive Prosthetics database, along with ECG signals from the MIT-BIH Normal Sinus Rhythm Database. The experiment result indicates that SDEMG outperformed comparative methods and produced high-quality sEMG samples. The source code of SDEMG the framework is available at: https://github.com/tonyliu0910/SDEMG

SDEMG: Score-based Diffusion Model for Surface Electromyographic Signal Denoising

TL;DR

This work addresses ECG contamination in sEMG recordings, which compromises clinical analysis and prosthetic control. It introduces SDEMG, a conditional score-based diffusion model that denoises sEMG by conditioning the reverse diffusion process on ECG-polluted inputs, with a two-stage training and sampling framework. The model uses a two-stream architecture with Half Normalized Filter blocks and FiLM-conditioned bridges, trained with a cosine schedule and noise-scale conditioning, and evaluated on the NINAPro DB2 dataset with MIT-BIH NSRD ECG interference. Results show that SDEMG yields higher signal quality and more faithful feature preservation than HP, TS, and FCN baselines across a range of SNRs and mismatch conditions, suggesting strong practical potential for clinical sEMG analysis.

Abstract

Surface electromyography (sEMG) recordings can be influenced by electrocardiogram (ECG) signals when the muscle being monitored is close to the heart. Several existing methods use signal-processing-based approaches, such as high-pass filter and template subtraction, while some derive mapping functions to restore clean sEMG signals from noisy sEMG (sEMG with ECG interference). Recently, the score-based diffusion model, a renowned generative model, has been introduced to generate high-quality and accurate samples with noisy input data. In this study, we proposed a novel approach, termed SDEMG, as a score-based diffusion model for sEMG signal denoising. To evaluate the proposed SDEMG approach, we conduct experiments to reduce noise in sEMG signals, employing data from an openly accessible source, the Non-Invasive Adaptive Prosthetics database, along with ECG signals from the MIT-BIH Normal Sinus Rhythm Database. The experiment result indicates that SDEMG outperformed comparative methods and produced high-quality sEMG samples. The source code of SDEMG the framework is available at: https://github.com/tonyliu0910/SDEMG
Paper Structure (14 sections, 8 equations, 5 figures, 1 table, 2 algorithms)

This paper contains 14 sections, 8 equations, 5 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. ECG interference removal methods
  4. Score-based diffusion models
  5. The Proposed SDEMG Method
  6. Training process
  7. Sampling process
  8. Model Architecture
  9. Experiments
  10. Datasets
  11. Data preprocessing and preparation
  12. Evaluation metrics
  13. Results and discussion
  14. Conclusion

Figures (5)

  • Figure 1: The diffusion process and the sampling in SDEMG.
  • Figure 2: Comparison of all methods with SNR$_{imp}$ under different SNR input.
  • Figure 3: Performance of simulated trunk sEMG denoising results evaluated by (a) SNR$_{imp}$, (b) RMSE, (c) RMSE$_{ARV}$, and (d) RMSE$_{MF}$
  • Figure 4: The waveform of (a) clean sEMG (b) noisy sEMG (c) denoised sEMG by SDEMG.
  • Figure :