TrustEMG-Net: Using Representation-Masking Transformer with U-Net for Surface Electromyography Enhancement

This dataset is specifically built for sEMG denoising research. It uses sEMG data from Ninapro DB2 as the clean sEMG reference [1]. To simulate noisy signals, five types of contaminants are included, sourced from various datasets and studies:

Baseline Wander (BW): Taken from the MIT-BIH NSTDB [2].

Electrocardiogram (ECG): Extracted from the MIT-BIH NSRD [3].

Motion Artifact (MOA): Derived from the MIT-BIH NSTDB [2] and additional research [4].

Powerline Interference (PLI) and White Gaussian Noise (WGN): Generated using mathematical simulations.

More details about this dataset are available in the paper "TrustEMG-Net: Using Representation-Masking Transformer with U-Net for Surface Electromyography Enhancement." You can access it here.

A subject-wise, four-fold dataset is available on this website. Alternatively, you can create your own sEMG denoising dataset by following the steps in the associated GitHub repository: TrustEMG Repository.

If you use this dataset in your research, please also cite the following references:

[1] Atzori, M., Gijsberts, A., Castellini, C., Caputo, B., Hager, A. G. M., Elsig, S., ... & Müller, H. (2014). Electromyography data for non-invasive naturally-controlled robotic hand prostheses. Scientific data, 1(1), 1-13.[2] Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.[3] Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.[4] Machado, Juliano, Amauri Machado, and Alexandre Balbinot. "Deep learning for surface electromyography artifact contamination type detection." Biomedical Signal Processing and Control 68 (2021): 102752.