{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T00:44:01Z","timestamp":1789087441692,"version":"build-2803163510"},"reference-count":36,"publisher":"AIP Publishing","issue":"8","funder":[{"name":"Engineering Research Center of Integration and Application of Digital Learning Technology Ministry of Education","award":["1411002"],"award-info":[{"award-number":["1411002"]}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,8,1]]},"abstract":"<jats:p>Surface electromyography (sEMG) signals are commonly utilized as a control source for upper-limb rehabilitation robots. Fast and high accuracy recognition of sEMG-based movements is crucial for the real-time control of robots. Here, we report a composite transformer-long short-term memory (LSTM) model for recognizing wrist-hand movements from sEMG signals. The sEMG data were collected to construct the training dataset. The consistent dataset was generated by a sliding window method to evaluate the model performance. We investigated the impact of window size and window increment on the sEMG-based recognition of wrist-hand movements. The results indicated that the recognition accuracy of the minimum window of 150\u00a0ms and the maximum window of 300\u00a0ms reached 94.68% and 99.35%, respectively. The recognition speeds were less than 0.3\u00a0ms. Finally, the trained transformer-LSTM model was combined with a majority voting strategy to determine the control outputs. The results confirm the model\u2019s capability for real-time and high-accuracy control with minimal delay. This method holds substantial significance for rapid feedback control of rehabilitation robots.<\/jats:p>","DOI":"10.1063\/5.0271372","type":"journal-article","created":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T17:36:42Z","timestamp":1756316202000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":2,"title":["sEMG-based recognition of wrist-hand movements using a composite transformer-LSTM model"],"prefix":"10.1063","volume":"96","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2386-5413","authenticated-orcid":false,"given":"Chaonan","family":"Zhang","sequence":"first","affiliation":[{"name":"The Experimental School, The Open University of China 1 , Beijing,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5770-9366","authenticated-orcid":false,"given":"Shuzhang","family":"Liang","sequence":"additional","affiliation":[{"name":"School of Engineering, The University of Tokyo 2 , Tokyo,","place":["Japan"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"317","published-online":{"date-parts":[[2025,8,27]]},"reference":[{"key":"2025082713363783000_c1","doi-asserted-by":"publisher","first-page":"1215","DOI":"10.3390\/jcm12031215","article-title":"Shoulder joint hybrid assistive limb treatment for chronic stroke patients with upper limb dysfunction","volume":"12","year":"2023","journal-title":"J.\u00a0Clin. Med."},{"key":"2025082713363783000_c2","volume-title":"Neurological Rehabilitation Optimizing Motor Performance","year":"2010","edition":"2"},{"key":"2025082713363783000_c3","doi-asserted-by":"publisher","first-page":"025101","DOI":"10.1063\/5.0218295","article-title":"An elbow passive exoskeleton with controllable assistance: Design and experimental","volume":"96","year":"2025","journal-title":"Rev. Sci. Instrum."},{"key":"2025082713363783000_c4","doi-asserted-by":"publisher","first-page":"eadi8852","DOI":"10.1126\/scirobotics.adi8852","article-title":"Estimating human joint moments unifies exoskeleton control, reducing user effort","volume":"9","year":"2024","journal-title":"Sci. Robot."},{"key":"2025082713363783000_c5","doi-asserted-by":"publisher","first-page":"19374","DOI":"10.1021\/acsami.2c21354","article-title":"High-fidelity sEMG signals recorded by an on-skin electrode based on AgNWs for hand gesture classification using machine learning","volume":"15","year":"2023","journal-title":"ACS Appl. Mater. Interfaces"},{"key":"2025082713363783000_c6","doi-asserted-by":"publisher","first-page":"612","DOI":"10.1007\/s42235-022-00280-3","article-title":"sEMG-based lower limb motion prediction using CNN-LSTM with improved PCA optimization algorithm","volume":"20","year":"2023","journal-title":"J.\u00a0Bionic Eng."},{"key":"2025082713363783000_c7","first-page":"1","article-title":"CNN-LSTM network based prediction of human joint angles using multi-band sEMG and historical angles","year":"2021"},{"key":"2025082713363783000_c8","doi-asserted-by":"publisher","first-page":"044106","DOI":"10.1063\/5.0057478","article-title":"Continuous motion estimation of lower limbs based on deep belief networks and random forest","volume":"93","year":"2022","journal-title":"Rev. Sci. Instrum."},{"key":"2025082713363783000_c9","doi-asserted-by":"publisher","first-page":"107323","DOI":"10.1016\/j.bspc.2024.107323","article-title":"EMG feature extraction and muscle selection for continuous upper limb movement regression","volume":"103","year":"2025","journal-title":"Biomed. Signal Process. Control"},{"key":"2025082713363783000_c10","doi-asserted-by":"publisher","first-page":"6403","DOI":"10.1109\/lra.2023.3303701","article-title":"Subject-independent estimation of continuous movements using CNN-LSTM for a home-based upper limb rehabilitation system","volume":"8","year":"2023","journal-title":"IEEE Rob. Autom. Lett."},{"key":"2025082713363783000_c11","doi-asserted-by":"publisher","first-page":"4890","DOI":"10.1038\/s41598-024-55483-x","article-title":"Advanced hybrid LSTM-transformer architecture for real-time multi-task prediction in engineering systems","volume":"14","year":"2024","journal-title":"Sci. Rep."},{"key":"2025082713363783000_c12","doi-asserted-by":"publisher","first-page":"1322","DOI":"10.3390\/electronics13071322","article-title":"Improving sEMG-based hand gesture recognition through optimizing parameters and sliding voting classifiers","volume":"13","year":"2024","journal-title":"Electronics"},{"key":"2025082713363783000_c13","doi-asserted-by":"publisher","first-page":"0366","DOI":"10.34133\/research.0366","article-title":"Deep learning model coupling wearable bioelectric and mechanical sensors for refined muscle strength assessment","volume":"7","year":"2024","journal-title":"Research"},{"key":"2025082713363783000_c14","doi-asserted-by":"publisher","first-page":"11000","DOI":"10.1038\/s41598-023-36490-w","article-title":"Transformer-based hand gesture recognition from instantaneous to fused neural decomposition of high-density EMG signals","volume":"13","year":"2023","journal-title":"Sci. Rep."},{"key":"2025082713363783000_c15","doi-asserted-by":"publisher","first-page":"2510615","DOI":"10.1109\/tim.2024.3373045","article-title":"A transformer-based gesture prediction model via sEMG sensor for human\u2013robot interaction","volume":"73","year":"2024","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"2025082713363783000_c16","doi-asserted-by":"publisher","first-page":"2503809","DOI":"10.1109\/tim.2020.3036654","article-title":"A CNN-LSTM hybrid model for wrist kinematics estimation using surface electromyography","volume":"70","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"2025082713363783000_c17","doi-asserted-by":"publisher","first-page":"106686","DOI":"10.1016\/j.bspc.2024.106686","article-title":"Integration of convolutional neural network and vision transformer for gesture recognition using sEMG","volume":"98","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"2025082713363783000_c18","doi-asserted-by":"publisher","first-page":"4211","DOI":"10.1109\/tnsre.2023.3324252","article-title":"TraHGR: Transformer for hand gesture recognition via electromyography","volume":"31","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"2025082713363783000_c19","doi-asserted-by":"publisher","first-page":"e2022WR032602","DOI":"10.1029\/2022wr032602","article-title":"Transformer versus LSTM: A comparison of deep learning models for karst spring discharge forecasting","volume":"60","year":"2024","journal-title":"Water Resour. Res."},{"key":"2025082713363783000_c20","doi-asserted-by":"publisher","first-page":"3631","DOI":"10.3390\/s24113631","article-title":"Dual stream long short-term memory feature fusion classifier for surface electromyography gesture recognition","volume":"24","year":"2024","journal-title":"Sensors"},{"key":"2025082713363783000_c21","doi-asserted-by":"publisher","first-page":"104546","DOI":"10.1016\/j.bspc.2022.104546","article-title":"How do sEMG segmentation parameters influence pattern recognition process? An approach based on wearable sEMG sensor","volume":"81","year":"2023","journal-title":"Biomed. Signal Process. Control"},{"key":"2025082713363783000_c22","doi-asserted-by":"publisher","first-page":"106681","DOI":"10.1016\/j.bspc.2024.106681","article-title":"Continuous limb joint angle prediction from sEMG using SA-FAWT and Conv-BiLSTM","volume":"97","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"2025082713363783000_c23","doi-asserted-by":"publisher","first-page":"147","DOI":"10.3390\/sym9080147","article-title":"Research on lower limb motion recognition based on fusion of sEMG and accelerometer signals","volume":"9","year":"2017","journal-title":"Symmetry"},{"key":"2025082713363783000_c24","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1016\/j.bspc.2012.08.008","article-title":"Bispectrum-based features classification for myoelectric control","volume":"8","year":"2013","journal-title":"Biomed. Signal Process. Control"},{"key":"2025082713363783000_c25","doi-asserted-by":"publisher","first-page":"4000910","DOI":"10.1109\/tim.2022.3225015","article-title":"Subject-independent continuous estimation of sEMG-based joint angles using both multisource domain adaptation and BP neural network","volume":"72","year":"2023","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"2025082713363783000_c26","doi-asserted-by":"publisher","first-page":"025102","DOI":"10.1063\/5.0240880","article-title":"Design of a multi-sensor walking boot to quantify the forefoot rocker motion as a function of walking speed","volume":"96","year":"2025","journal-title":"Rev. Sci. Instrum."},{"key":"2025082713363783000_c27","doi-asserted-by":"publisher","first-page":"12256","DOI":"10.1109\/access.2021.3050704","article-title":"EMG biometric systems based on different wrist-hand movements","volume":"9","year":"2021","journal-title":"IEEE Access"},{"key":"2025082713363783000_c28","doi-asserted-by":"publisher","first-page":"6502909","DOI":"10.1109\/tim.2023.3276522","article-title":"A two-stage GA-based sEMG feature selection method for user-independent continuous estimation of elbow angles","volume":"72","year":"2023","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"2025082713363783000_c29","doi-asserted-by":"publisher","first-page":"64","DOI":"10.3390\/life12010064","article-title":"sEMG-based motion recognition of upper limb rehabilitation using the improved Yolo-v4 algorithm","volume":"12","year":"2022","journal-title":"Life"},{"key":"2025082713363783000_c30","doi-asserted-by":"publisher","first-page":"109120","DOI":"10.1016\/j.agwat.2024.109120","article-title":"Comparison of transformer, LSTM and coupled algorithms for soil moisture prediction in shallow-groundwater-level areas with interpretability analysis","volume":"305","year":"2024","journal-title":"Agric. Water Manag."},{"key":"2025082713363783000_c31","article-title":"Attention is all you need","year":"2017"},{"key":"2025082713363783000_c32","first-page":"352","article-title":"An intrusion detection method based on transformer-LSTM model","year":"2023"},{"key":"2025082713363783000_c33","doi-asserted-by":"publisher","first-page":"39589","DOI":"10.1109\/access.2024.3375365","article-title":"A hybrid transformer-LSTM model with 3D separable convolution for video prediction","volume":"12","year":"2024","journal-title":"IEEE Access"},{"key":"2025082713363783000_c34","doi-asserted-by":"publisher","first-page":"11397","DOI":"10.1007\/s00521-024-09780-1","article-title":"Deep learning based ankle\u2013foot movement classification for prosthetic foot","volume":"36","year":"2024","journal-title":"Neural Comput. Appl."},{"key":"2025082713363783000_c35","doi-asserted-by":"publisher","first-page":"616","DOI":"10.3390\/sym16050616","article-title":"A transformer and LSTM-based approach for blind well lithology prediction","volume":"16","year":"2024","journal-title":"Symmetry"},{"key":"2025082713363783000_c36","doi-asserted-by":"publisher","first-page":"1956","DOI":"10.1109\/tbme.2008.919734","article-title":"Support vector machine-based classification scheme for myoelectric control applied to upper limb","volume":"55","year":"2008","journal-title":"IEEE Trans. Biomed. Eng."}],"container-title":["Review of Scientific Instruments"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/aip\/rsi\/article-pdf\/doi\/10.1063\/5.0271372\/20671908\/085208_1_5.0271372.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/rsi\/article-pdf\/doi\/10.1063\/5.0271372\/20671908\/085208_1_5.0271372.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T17:36:42Z","timestamp":1756316202000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/rsi\/article\/96\/8\/085208\/3360919\/sEMG-based-recognition-of-wrist-hand-movements"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,1]]},"references-count":36,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2025,8,1]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0271372","relation":{},"ISSN":["0034-6748","1089-7623"],"issn-type":[{"value":"0034-6748","type":"print"},{"value":"1089-7623","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,8]]},"published":{"date-parts":[[2025,8,1]]},"article-number":"085208"}}