{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T07:20:42Z","timestamp":1787642442780,"version":"build-2736575974"},"reference-count":35,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T00:00:00Z","timestamp":1787616000000},"content-version":"vor","delay-in-days":24,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T00:00:00Z","timestamp":1787616000000},"content-version":"tdm","delay-in-days":24,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"2024 SUIBE Young Faculty Grant","award":["24QN011"],"award-info":[{"award-number":["24QN011"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["J. Neural Eng."],"published-print":{"date-parts":[[2026,8,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:italic>Objective<\/jats:italic>\n                    . Gesture recognition using surface electromyography (sEMG) is a promising technology for wearable human\u2013machine interaction systems. However, sEMG-based gesture classification models often suffer from performance degradation in cross-user scenarios due to the variability in individual physiological signals. To address this issue, large-scale deep learning algorithms with a great amount of parameters are employed to enhance model generalization and improve recognition performance across different users. However, the use of deep learning models on wearable devices is challenging due to their limited computational power and memory capacity.\n                    <jats:italic>Methods<\/jats:italic>\n                    . This study addresses this challenge through knowledge distillation, which compresses large teacher models into lightweight student models while maintaining performance. We trained deep teacher models, including DenseNet, InceptionV3, and VggNet, and distilled their parameters into compact convolutional neural network and long short-term memory (LSTM) models. The experiment was performed on the sEMG data of 30 subjects collected from a wrist-band electrode with 32 channels.\n                    <jats:italic>Main Results<\/jats:italic>\n                    . The results demonstrated that student models, particularly LSTM-based ones, achieved classification accuracy close to or even higher than their teacher models, with a highest accuracy improvement. Among the teacher\u2013student combinations, the DenseNet121-LSTM architecture achieved the highest classification accuracy. The relationship between computational complexity floating-point operations (FLOPs) and model performance was also analyzed, showing that the distilled models can effectively approximate high-FLOP models.\n                    <jats:italic>Significance<\/jats:italic>\n                    . This research enhances the feasibility of embedding deep learning models into wearable systems (e.g. wristband or wristwatch), facilitating more responsive and efficient gesture recognition in daily-life applications. The Python implementation of the complete knowledge distillation framework is publicly available at:\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/Open-EXG\/handDistill\">https:\/\/github.com\/Open-EXG\/handDistill<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1088\/1741-2552\/ae93f4","type":"journal-article","created":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T22:52:23Z","timestamp":1785797543000},"page":"046061","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Knowledge distillation for sEMG-based gesture recognition: enhancing wearable HMI systems with lightweight models"],"prefix":"10.1088","volume":"23","author":[{"given":"Fang","family":"Qiu","sequence":"first","affiliation":[{"name":"Shanghai University of International Business and Economics"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3056-4339","authenticated-orcid":true,"given":"Chenyun","family":"Dai","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaodong","family":"Liu","sequence":"additional","affiliation":[{"name":"Guangxi University of Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinming","family":"Ye","sequence":"additional","affiliation":[{"name":"East China University of Science and Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,8,25]]},"reference":[{"key":"jneae93f4bib1","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1080\/03091902.2021.1936237","type":"journal-article","article-title":"Developments in the human machine interface technologies and their applications: a review","volume":"45","author":"Singh","year":"2021","journal-title":"J. Med. Eng. Technol."},{"key":"jneae93f4bib2","first-page":"pp 3460","type":"conference-proceedings","article-title":"A 3d-printed, adjustable armband for electromyography-based finger movement classification with haptic feedback","author":"Wang","year":"2020"},{"key":"jneae93f4bib3","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1038\/scientificamerican1096-82","type":"journal-article","article-title":"Controlling computers with neural signals","volume":"275","author":"Lusted","year":"1996","journal-title":"Sci. Am."},{"key":"jneae93f4bib4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2022.3180434","type":"journal-article","article-title":"Measuring neuromuscular electrophysiological activities to decode HD-sEMG biometrics for cross-application discrepant personal identification with unknown identities","volume":"71","author":"Jiang","year":"2022","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"jneae93f4bib5","doi-asserted-by":"publisher","first-page":"eaaz8693","DOI":"10.1126\/sciadv.aaz8693","type":"journal-article","article-title":"Haptic-feedback smart glove as a creative human-machine interface (HMI) for virtual\/augmented reality applications","volume":"6","author":"Zhu","year":"2020","journal-title":"Sci. Adv."},{"key":"jneae93f4bib6","doi-asserted-by":"publisher","DOI":"10.1002\/adfm.202008347","type":"journal-article","article-title":"Nanowire-based soft wearable human\u2013machine interfaces for future virtual and augmented reality applications","volume":"31","author":"Wang","year":"2021","journal-title":"Adv. Funct. Mater."},{"key":"jneae93f4bib7","doi-asserted-by":"publisher","first-page":"10292","DOI":"10.1109\/JSEN.2022.3169492","type":"journal-article","article-title":"Non-invasive human-machine interface (HMI) systems with hybrid on-body sensors for controlling upper-limb prosthesis: a review","volume":"22","author":"Zhou","year":"2022","journal-title":"IEEE Sens. J."},{"key":"jneae93f4bib8","doi-asserted-by":"publisher","DOI":"10.20935\/AcadEng7350","type":"journal-article","article-title":"Human\u2013robot interaction via wearable device\u2014a wireless glove system for remote control of 7-DoF robotic arm","volume":"1","author":"Manolescu","year":"2024","journal-title":"Acad. Eng."},{"key":"jneae93f4bib9","doi-asserted-by":"publisher","first-page":"9198","DOI":"10.1109\/JSEN.2022.3165988","type":"journal-article","article-title":"Surface electromyography as a natural human\u2013machine interface: a review","volume":"22","author":"Zheng","year":"2022","journal-title":"IEEE Sens. J."},{"key":"jneae93f4bib10","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1109\/TII.2020.3041618","type":"journal-article","article-title":"Electromyography-based gesture recognition: is it time to change focus from the forearm to the wrist?","volume":"18","author":"Botros","year":"2020","journal-title":"IEEE Trans. Ind. Inform."},{"key":"jneae93f4bib11","doi-asserted-by":"publisher","first-page":"4186","DOI":"10.1109\/TNSRE.2024.3502624","type":"journal-article","article-title":"Ye Li and Yishan Wang. A novel dual-model adaptive continuous learning strategy for wrist-sEMG real-time gesture recognition","volume":"32","author":"Liu","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"jneae93f4bib12","doi-asserted-by":"publisher","first-page":"300","DOI":"10.1016\/j.irbm.2022.06.001","type":"journal-article","article-title":"Real-time classification of emg myo armband data using support vector machine","volume":"43","author":"Tepe","year":"2022","journal-title":"IRBM"},{"key":"jneae93f4bib13","article-title":"Discrete gesture recognition using multi-modal PPG, IMU and single-channel EMG recorded at the wrist","author":"Eddy","year":"2024","type":"conference-proceedings"},{"key":"jneae93f4bib14","doi-asserted-by":"publisher","first-page":"2198","DOI":"10.1109\/TBME.2008.923917","type":"journal-article","article-title":"A comparison of the effects of electrode implantation and targeting on pattern classification accuracy for prosthesis control","volume":"55","author":"Farrell","year":"2008","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"jneae93f4bib15","doi-asserted-by":"publisher","first-page":"1649","DOI":"10.1109\/TBME.2012.2191551","type":"journal-article","article-title":"High-density myoelectric pattern recognition toward improved stroke rehabilitation","volume":"59","author":"Zhang","year":"2012","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"jneae93f4bib16","first-page":"pp 174","type":"conference-proceedings","article-title":"Hand gesture recognition based on sEMG signals using support vector machines","author":"Pomboza-Junez","year":"2016"},{"key":"jneae93f4bib17","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1016\/j.patrec.2017.12.005","type":"journal-article","article-title":"A multi-stream convolutional neural network for semg-based gesture recognition in muscle-computer interface","volume":"119","author":"Wei","year":"2019","journal-title":"Pattern Recognit. Lett."},{"key":"jneae93f4bib18","doi-asserted-by":"publisher","DOI":"10.1142\/S2424905X20410019","type":"journal-article","article-title":"Surface EMG-based hand gesture recognition via hybrid and dilated deep neural network architectures for neurorobotic prostheses","volume":"5","author":"Rahimian","year":"2020","journal-title":"J. Med. Robot. Res."},{"key":"jneae93f4bib19","doi-asserted-by":"publisher","first-page":"646","DOI":"10.1109\/TNSRE.2020.2966749","type":"journal-article","article-title":"Unsupervised cross-subject adaptation for predicting human locomotion intent","volume":"28","author":"Zhang","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"jneae93f4bib20","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/acb7a0","type":"journal-article","article-title":"Multi-source domain generalization and adaptation toward cross-subject myoelectric pattern recognition","volume":"20","author":"Zhang","year":"2023","journal-title":"J. Neural Eng."},{"key":"jneae93f4bib21","doi-asserted-by":"publisher","first-page":"1119","DOI":"10.1109\/TNSRE.2024.3372002","type":"journal-article","article-title":"EMG-based multi-user hand gesture classification via unsupervised transfer learning using unknown calibration gestures","volume":"32","author":"Shi","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"jneae93f4bib22","first-page":"pp 176","type":"conference-proceedings","article-title":"Sparsification and separation of deep learning layers for constrained resource inference on wearables","author":"Bhattacharya","year":"2016"},{"key":"jneae93f4bib23","first-page":"pp 7","type":"conference-proceedings","article-title":"An early resource characterization of deep learning on wearables, smartphones and internet-of-things devices","author":"Lane","year":"2015"},{"key":"jneae93f4bib24","doi-asserted-by":"publisher","first-page":"2295","DOI":"10.1109\/JPROC.2017.2761740","type":"journal-article","article-title":"Efficient processing of deep neural networks: a tutorial and survey","volume":"105","author":"Sze","year":"2017","journal-title":"Proc. IEEE"},{"key":"jneae93f4bib25","article-title":"Efficient acceleration of deep learning inference on resource-constrained edge devices: review","author":"Shuvo","year":"2022","type":"conference-proceedings"},{"key":"jneae93f4bib26","doi-asserted-by":"publisher","first-page":"3048","DOI":"10.1109\/TPAMI.2021.3055564","type":"journal-article","article-title":"Knowledge distillation and student-teacher learning for visual intelligence: a review and new outlooks","volume":"44","author":"Wang","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"jneae93f4bib27","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.1109\/THMS.2022.3175408","type":"journal-article","article-title":"Optimization of HD-sEMG-based cross-day hand gesture classification by optimal feature extraction and data augmentation","volume":"52","author":"Jiang","year":"2022","journal-title":"IEEE Trans. Hum.-Mach. Syst."},{"key":"jneae93f4bib28","first-page":"pp 4700","type":"conference-proceedings","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"jneae93f4bib29","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","type":"preprint"},{"key":"jneae93f4bib30","first-page":"pp 2818","type":"conference-proceedings","article-title":"Rethinking the inception architecture for computer vision","author":"Szegedy","year":"2016"},{"key":"jneae93f4bib31","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1016\/j.isprsjprs.2020.12.010","type":"journal-article","article-title":"Review on convolutional neural networks (CNN) in vegetation remote sensing","volume":"173","author":"Kattenborn","year":"2021","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"jneae93f4bib32","doi-asserted-by":"publisher","first-page":"1235","DOI":"10.1162\/neco_a_01199","type":"journal-article","article-title":"A review of recurrent neural networks: LSTM cells and network architectures","volume":"31","author":"Yong","year":"2019","journal-title":"Neural Comput."},{"key":"jneae93f4bib33","article-title":"Distilling the knowledge in a neural network","author":"Hinton","year":"2015","type":"preprint"},{"key":"jneae93f4bib34","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2022.3217868","type":"journal-article","article-title":"User-tailored hand gesture recognition system for wearable prosthesis and armband based onsurface electromyogram","volume":"71","author":"Meng","year":"2022","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"jneae93f4bib35","article-title":"Open Access Dataset and Toolbox of High-Density Surface Electromyogram Recordings","author":"Jiang","year":"2021","type":"journal-article"}],"container-title":["Journal of Neural Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T06:47:13Z","timestamp":1787640433000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae93f4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":35,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,8,25]]},"published-print":{"date-parts":[[2026,8,1]]}},"URL":"https:\/\/doi.org\/10.1088\/1741-2552\/ae93f4","relation":{},"ISSN":["1741-2560","1741-2552"],"issn-type":[{"value":"1741-2560","type":"print"},{"value":"1741-2552","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,1]]},"assertion":[{"value":"Knowledge distillation for sEMG-based gesture recognition: enhancing wearable HMI systems with lightweight models","name":"article_title","label":"Article Title"},{"value":"Journal of Neural Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2025-11-26","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-03","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-25","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}