Deep Residual U-Net Autoencoder with Weighted Overlapping Reconstruction for EMG Signal Denoising

Electromyography (EMG) signals, crucial for neuromuscular assessment, are frequently corrupted by noise, impairing signal fidelity and subsequent analysis across diverse applications. Conventional filters often inadequately address non-stationary noise or introduce signal distortion. This paper intr...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Signal Processing Algorithms, Architectures, Arrangements, and Applications Conference proceedings s. 198 - 203
Hlavní autoři: Mehmood, Atif, Wiora, Jozef
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: Division of Signal Processing and Electronic Syste 17.09.2025
Témata:
Autoencoders
Convolution
Convolutional Autoencoder (CAE)
Deep learning
Electromyography
Electromyography (EMG)
Noise measurement
Noise reduction
Residual Networks
Signal denoising
Signal Processing
Signal to noise ratio
Synthetic data
Synthetic Data Generation
Training
U-Net
ISBN:9788362065493, 8362065494
ISSN:2326-0262
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Electromyography (EMG) signals, crucial for neuromuscular assessment, are frequently corrupted by noise, impairing signal fidelity and subsequent analysis across diverse applications. Conventional filters often inadequately address non-stationary noise or introduce signal distortion. This paper introduces an advanced deep learning framework for EMG denoising, centred on a U-Net-inspired convolutional autoencoder with integrated residual blocks and skip connections. Training utilised synthetic EMG data, closely emulating physiological frequency bands and burst dynamics, subsequently corrupted by a comprehensive noise model encompassing electrode, crosstalk, electronic, drift, and contact artefacts. Training was guided by a custom loss function that combined weighted mean squared error (MSE) with signal-to-noise ratio (SNR). The proposed autoencoder achieved substantial improvements, SNR increased from -0.95 dB (noisy) to 14.64 dB (denoised), and MSE was drastically reduced from 0.001493 V 2 to 0.000041 V 2 on the test dataset. Qualitative analysis confirmed effective noise suppression while retaining crucial EMG burst characteristics. This advanced framework offers a promising solution for robust restoration of EMG signals in practical settings.
ISBN:9788362065493
8362065494
ISSN:2326-0262
DOI:10.23919/SPA65537.2025.11215114