Multi-Dataset Benchmarking and Preliminary Transfer Learning for Lower-Limb EMG Classification Toward Adaptive Prosthesis Control
Acta of Bioengineering and Biomechanics · 18.05.2026 · 28 (2)
DOI: 10.37190/abb/221221 ↗Kurz gefasst
Fachartikel in der Zeitschrift Acta of Bioengineering and Biomechanics, online erschienen am 18.05.2026. Verfasst von Jafar Jafar und 2 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Warsaw University of Technology, Nowowiejska 24, 00-66…. Hauptthemen laut MEDLINE: Electromyography, Benchmarking, Lower Extremity, Artificial Limbs.
Abstract (Originalsprache)
Purpose
Reliable decoding of lower-limb motor intent from surface electromyography signals remains challenging due to electrode placement variability, muscle fatigue, inter-subject differences, and differences between able-bodied and amputee populations. This study evaluates the robustness of commonly used machine-learning approaches for lower-limb electromyography classification under these sources of variability.
Methods
A unified benchmarking analysis was conducted across four lower-limb electromyography datasets: UCI Lower Limb Electromyography, Vastus Medialis Electromyography, K2MUSE, and the Above-Knee Sit-to-Stand amputee dataset. Three classification models—support vector machine, random forest, and multilayer perceptron—were evaluated using standardized preprocessing and feature extraction pipelines. Cross-subject robustness was assessed using leave-one-subject-out validation on the K2MUSE dataset. A preliminary transfer-learning experiment was also performed by fine-tuning a multilayer perceptron pretrained on able-bodied data using the amputee sit-to-stand dataset.
Results
The random forest model achieved the highest performance in datasets with lower variability, with macro F1 scores of 0.9033 ± 0.0048 on the UCI dataset and 0.6000 ± 0.0087 on the Vastus dataset. Under cross-subject non-ideal acquisition conditions, performance decreased for all models. The multilayer perceptron showed competitive robustness in the K2MUSE Ideal-versus-Fatigue LOSO evaluation (0.476 ± 0.058 macro-F1), while random forest showed the most compact subject-wise distribution (median 0.476 [IQR 0.049]). In the transfer-learning experiment, fine-tuning improved performance on the Above-Knee STS dataset relative to training from scratch (macro-F1 0.5540 ± 0.0141 vs. 0.5203 ± 0.0371).
Conclusions
These results highlight the different strengths of classical and neural machine-learning models under varying electromyography variability conditions and provide a foundation for more robust intention-decoding systems for adaptive lower-limb prosthetic control.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Faculty of Power and Aeronautical Engineering, Institute of Aeronautics and Applied Mechanics, Warsaw University of Technology, Nowowiejska 24, 00-665 Warsaw, Poland.
- Centre of Excellence for Health Technologies (HealthTEC), Warsaw University of Technology, Pl. Politechniki 1, 00-661 Warsaw, Poland.
Zitieren
APA
Jafar, J., Rzymkowski, C., & Kordecki, A. (2026). Multi-Dataset Benchmarking and Preliminary Transfer Learning for Lower-Limb EMG Classification Toward Adaptive Prosthesis Control. Acta of Bioengineering and Biomechanics, 28(2). https://doi.org/10.37190/abb/221221
Vancouver
Jafar J, Rzymkowski C, Kordecki A. Multi-Dataset Benchmarking and Preliminary Transfer Learning for Lower-Limb EMG Classification Toward Adaptive Prosthesis Control. Acta of Bioengineering and Biomechanics. 2026;28(2). doi:10.37190/abb/221221
BibTeX
@article{Jafar2026,
author = {Jafar, Jafar and Rzymkowski, Cezary and Kordecki, Andrzej},
title = {Multi-Dataset Benchmarking and Preliminary Transfer Learning for Lower-Limb EMG Classification Toward Adaptive Prosthesis Control},
journal = {Acta of Bioengineering and Biomechanics},
year = {2026},
volume = {28},
number = {2},
doi = {10.37190/abb/221221}
}
EUVIMED