{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T22:51:54Z","timestamp":1781563914688,"version":"3.54.5"},"reference-count":0,"publisher":"Politechnika Wroclawska Oficyna Wydawnicza","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Acta of Bioengineering and Biomechanics"],"published-print":{"date-parts":[[2026,5,18]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>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\u2014support vector machine, random forest, and multilayer perceptron\u2014were 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.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The random forest model achieved the highest performance in datasets with lower variability, with macro F1 scores of 0.9033 \u00b1 0.0048 on the UCI dataset and 0.6000 \u00b1 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 \u00b1 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 \u00b1 0.0141 vs. 0.5203 \u00b1 0.0371).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.37190\/abb\/221221","type":"journal-article","created":{"date-parts":[[2026,5,24]],"date-time":"2026-05-24T14:01:01Z","timestamp":1779631261000},"source":"Crossref","is-referenced-by-count":0,"title":["Multi-Dataset Benchmarking and Preliminary Transfer Learning for Lower-Limb EMG Classification Toward Adaptive Prosthesis Control"],"prefix":"10.37190","volume":"28","author":[{"given":"Jafar","family":"Jafar","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cezary","family":"Rzymkowski","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrzej","family":"Kordecki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"23140","published-online":{"date-parts":[[2026,5,18]]},"container-title":["Acta of Bioengineering and Biomechanics"],"original-title":[],"link":[{"URL":"https:\/\/actabio.pwr.edu.pl\/d\/xGBUKOTtQKxVvA0RqREFBPF0WUXJ9XFlOSylDC0QHFm9PFQk5Eht1FRMSWHoFPFgHFkhCcHVeX1RYOA\/15.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/actabio.pwr.edu.pl\/d\/xGBUKOTtQKxVvA0RqREFBPF0WUXJ9XFlOSylDC0QHFm9PFQk5Eht1FRMSWHoFPFgHFkhCcHVeX1RYOA\/15.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T21:52:36Z","timestamp":1781560356000},"score":1,"resource":{"primary":{"URL":"https:\/\/actabio.pwr.edu.pl\/d\/xGBUKOTtQKxVvA0RqREFBPF0WUXJ9XFlOSylDC0QHFm9PFQk5Eht1FRMSWHoFPFgHFkhCcHVeX1RYOA\/15.pdf"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":0,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026]]}},"alternative-id":["221221"],"URL":"https:\/\/doi.org\/10.37190\/abb\/221221","relation":{},"ISSN":["1509-409X"],"issn-type":[{"value":"1509-409X","type":"print"}],"subject":[],"published":{"date-parts":[[2026,5,18]]}}}