EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Feature-Shuffle and Multi-Head Attention-Based Autoencoder for Eliminating Electrode Motion Noise in ECG Applications

Szu-Ting Wang iD, Wen-Yen Hsu, Shin-Chi Lai iD, Ming-Hwa Sheu iD, Chuan-Yu Chang iD, Shih-Chang Hsia iD, Szu-Hong Wang iD

Sensors · 2025

DOI: 10.3390/s25206322 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Szu-Ting Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Information Engineering, Chaoyang University of Technology, Taichung City, Wufeng 413310, Taiwan. Hauptthemen laut MEDLINE: Electrocardiography. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council, National Science and Technology Council, Taiwan, Smart Machinery and Intelligent Manufacturing Research Center, National Formosa University, Yunlin, Taiwan.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Electrocardiograms (ECGs) are critical for cardiovascular disease diagnosis, but their accuracy is often compromised by electrode motion (EM) artifacts—large, nonstationary distortions caused by patient movement and electrode-skin interface shifts. These artifacts overlap in frequency with genuine cardiac signals, rendering traditional filtering methods ineffective and increasing the risk of false alarms and misdiagnosis, particularly in wearable and ambulatory ECG applications. To address this, we propose the Feature-Shuffle Multi-Head Attention Autoencoder (FMHA-AE), a novel architecture integrating multi-head self-attention (MHSA) and a feature-shuffle mechanism to enhance ECG denoising. MHSA captures long-range temporal and spatial dependencies, while feature shuffling improves representation robustness and generalization. Experimental results show that FMHA-AE achieves an average signal-to-noise ratio (SNR) improvement of 25.34 dB and a percentage root mean square difference (PRD) of 10.29%, outperforming conventional wavelet-based and deep learning baselines. These results confirm the model’s ability to retain critical ECG morphology while effectively removing noise. FMHA-AE demonstrates strong potential for real-time ECG monitoring in mobile and clinical environments. This work contributes an efficient deep learning approach for noise-robust ECG analysis, supporting accurate cardiovascular assessment under motion-prone conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectrocardiographyHumansElectrodesSignal-To-Noise RatioSignal Processing, Computer-AssistedArtifactsMotionAlgorithmsDeep LearningAutoencoder

Stichwörter der Autor:innen

ECG denoisingautoencoderelectrode motion artifactsmulti-head self-attentiontransformer

Autor:innen und Institutionen

  • Department of Computer Science and Information Engineering, Chaoyang University of Technology, Taichung City, Wufeng 413310, Taiwan.
  • Department of Electronic Engineering, National Yunlin University of Science and Technology, Yunlin County, Douliu 64002, Taiwan.
  • Department of Automation Engineering and the Smart Machinery and Intelligent Manufacturing Research Center, National Formosa University, Yunlin County, Huwei 632301, Taiwan.
  • Department of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Yunlin County, Douliu 64002, Taiwan.

ORCID-Profile: Szu-Ting Wang, Shin-Chi Lai, Ming-Hwa Sheu, Chuan-Yu Chang, Shih-Chang Hsia, Szu-Hong Wang

Förderung

  • National Science and Technology Council
  • National Science and Technology Council, Taiwan
  • Smart Machinery and Intelligent Manufacturing Research Center, National Formosa University, Yunlin, Taiwan
  • Intelligent Recognition Industry Service Center

Zitieren

APA

Wang, S. T., Hsu, W. Y., Lai, S. C., Sheu, M. H., Chang, C. Y., Hsia, S. C., & Wang, S. H. (2025). Feature-Shuffle and Multi-Head Attention-Based Autoencoder for Eliminating Electrode Motion Noise in ECG Applications. Sensors. https://doi.org/10.3390/s25206322

Vancouver

Wang ST, Hsu WY, Lai SC, Sheu MH, Chang CY, Hsia SC, et al. Feature-Shuffle and Multi-Head Attention-Based Autoencoder for Eliminating Electrode Motion Noise in ECG Applications. Sensors. 2025. doi:10.3390/s25206322

BibTeX
@article{Wang2025,
  author = {Wang, Szu-Ting and Hsu, Wen-Yen and Lai, Shin-Chi and Sheu, Ming-Hwa and Chang, Chuan-Yu and Hsia, Shih-Chang and Wang, Szu-Hong},
  title = {Feature-Shuffle and Multi-Head Attention-Based Autoencoder for Eliminating Electrode Motion Noise in ECG Applications},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25206322}
}
Kostenlos starten →