{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T05:01:24Z","timestamp":1784782884560,"version":"3.55.0"},"reference-count":41,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T00:00:00Z","timestamp":1784505600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Accurate gesture recognition is crucial for precision control of upper limb prostheses. High-density surface electromyography (HD-sEMG) enhances spatial resolution and information richness of human gesture representation, thus improving myoelectric control of bionic limbs. Recently, deep learning has been increasingly applied to HD-sEMG to enhance gesture recognition performance. However, the black-box nature of neural networks limits their interpretability and model optimization, hindering their practical application. In this paper, we developed a spatiotemporal dual-channel interpretable hybrid neural network (STDC-Net), and validated it using the Capgmyo DB-a dataset. STDC-Net uses Feature Channels and Spatial Channels to process the feature and spatial information of sEMG signals respectively for increased interpretability. SHapley Additive exPlanations (SHAP) values are used to rank the feature importance, aiding feature filtering and reducing the impact of irrelevant features. Graph attention layers are used to calculate the connections between each Spatial Channel, illustrating the relationships between channels. Our results demonstrated the superior performance of our STDC-Net compared to state-of-the-art (SOTA) methods. STDC-Net achieved 99.8% accuracy with a 150 ms sliding window, exceeding real-time implementation requirements for intra-subject tasks. It reached an accuracy of 97.33 \u00b1 2.53% after fine-tuning for inter-subject tasks, outperforming the SOTA methods. Importantly, the SHAP value maps and the channel connection maps enhance the interpretability of the neural networks by offering detailed insights into the contribution of input features and parameter interactions of the network. These findings suggest STDC-Net holds significant promise for real-time prosthetic control.<\/jats:p>","DOI":"10.3390\/s26144602","type":"journal-article","created":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T16:19:27Z","timestamp":1784564367000},"page":"4602","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Spatiotemporal Dual-Channel Interpretable Hybrid Neural Network for HD-sEMG-Based Gesture Recognition"],"prefix":"10.3390","volume":"26","author":[{"given":"Zhefei","family":"Cai","sequence":"first","affiliation":[{"name":"College of Information Engineering, China Jiliang University, Hangzhou 310018, China"},{"name":"Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA"},{"name":"School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8714-081X","authenticated-orcid":false,"given":"Su","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA"},{"name":"Department of Neurology, University of Miami Miller School of Medicine, Miami, FL 33136, USA"},{"name":"Desai Sethi Urology Institute, University of Miami Miller School of Medicine, Miami, FL 33136, USA"},{"name":"Miami Project to Cure Paralysis, University of Miami, Miami, FL 33136, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyue","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1951-084X","authenticated-orcid":false,"given":"Michael","family":"Houston","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingle","family":"Fan","sequence":"additional","affiliation":[{"name":"School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1927-4103","authenticated-orcid":false,"given":"Yingchun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, University of Miami, Coral Gables, FL 33146, USA"},{"name":"Desai Sethi Urology Institute, University of Miami Miller School of Medicine, Miami, FL 33136, USA"},{"name":"Miami Project to Cure Paralysis, University of Miami, Miami, FL 33136, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1487","DOI":"10.1109\/TNSRE.2024.3383857","article-title":"Continuous Motion Intention Prediction Using sEMG for Upper-Limb Rehabilitation: A Systematic Review of Model-Based and Model-Free Approaches","volume":"32","author":"Wei","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1550024","DOI":"10.1142\/S0129065715500240","article-title":"Three-dimensional innervation zone imaging from multi-channel surface EMG recordings","volume":"25","author":"Liu","year":"2015","journal-title":"Int. J. Neural Syst."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3059","DOI":"10.1016\/j.clinph.2016.06.014","article-title":"Motor unit number estimation based on high-density surface electromyography decomposition","volume":"127","author":"Peng","year":"2016","journal-title":"Clin. Neurophysiol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"471","DOI":"10.1109\/JBHI.2014.2328497","article-title":"Surface EMG decomposition based on K-means clustering and convolution kernel compensation","volume":"19","author":"Ning","year":"2014","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"3641","DOI":"10.1109\/TNSRE.2023.3309546","article-title":"High-density surface EMG decomposition by combining iterative convolution kernel compensation with an energy-specific peel-off strategy","volume":"31","author":"Zheng","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"596","DOI":"10.1109\/TBME.2023.3309969","article-title":"Enhanced dynamic surface EMG decomposition using the non-negative matrix factorization and three-dimensional motor unit localization","volume":"71","author":"He","year":"2023","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"6979","DOI":"10.1109\/JIOT.2020.2979328","article-title":"Learning effective spatial\u2013temporal features for sEMG armband-based gesture recognition","volume":"7","author":"Zhang","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2065","DOI":"10.1007\/s00371-022-02465-7","article-title":"An approach to continuous hand movement recognition using SEMG based on features fusion","volume":"39","author":"Li","year":"2023","journal-title":"Vis. Comput."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"36571","DOI":"10.1038\/srep36571","article-title":"Gesture recognition by instantaneous surface EMG images","volume":"6","author":"Geng","year":"2016","journal-title":"Sci. Rep."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Atzori, M., Cognolato, M., and M\u00fcller, H. (2016). Deep learning with convolutional neural networks applied to electromyography data: A resource for the classification of movements for prosthetic hands. Front. Neurorobot., 10.","DOI":"10.3389\/fnbot.2016.00009"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"512","DOI":"10.1109\/JAS.2021.1003865","article-title":"Deep learning for EMG-based human-machine interaction: A review","volume":"8","author":"Xiong","year":"2021","journal-title":"IEEE\/CAA J. Autom. Sin."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"6591035","DOI":"10.1155\/2021\/6591035","article-title":"A Hierarchical View Pooling Network for Multichannel Surface Electromyography-Based Gesture Recognition","volume":"2021","author":"Wei","year":"2021","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"3878","DOI":"10.1109\/TNSRE.2024.3487216","article-title":"A novel multi-feature fusion network with spatial partitioning strategy and cross-attention for armband-based gesture recognition","volume":"32","author":"Hu","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1821","DOI":"10.1109\/TNSRE.2017.2687520","article-title":"A framework of temporal-spatial descriptors-based feature extraction for improved myoelectric pattern recognition","volume":"25","author":"Khushaba","year":"2017","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"112235","DOI":"10.1016\/j.asoc.2024.112235","article-title":"A survey on hand gesture recognition based on surface electromyography: Fundamentals, methods, applications, challenges and future trends","volume":"166","author":"Ni","year":"2024","journal-title":"Appl. Soft Comput."},{"key":"ref_16","unstructured":"Kipf, T.N., and Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A comprehensive survey on graph neural networks","volume":"32","author":"Wu","year":"2020","journal-title":"IEEE Trans. Neural Net. Learn. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Massa, S.M., Riboni, D., and Nazarpour, K. (2022, January 7\u201310). Graph neural networks for hd emg-based movement intention recognition: An initial investigation. Proceedings of the 2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE), Tainan, Taiwan.","DOI":"10.1109\/RASSE54974.2022.9989657"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Yu, B., Yin, H., and Zhu, Z. (2017). Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. arXiv.","DOI":"10.24963\/ijcai.2018\/505"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1109\/TNSRE.2023.3342050","article-title":"Cross-user electromyography pattern recognitionn based on a novel spatial-temporal graph convolutional network","volume":"32","author":"Xu","year":"2023","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zhong, W., Zhang, Y., Fu, P., Xiong, W., and Zhang, M. (2023). A spatio-temporal graph convolutional network for gesture recognition from high-density electromyography. arXiv.","DOI":"10.1109\/M2VIP58386.2023.10413402"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"58256","DOI":"10.1109\/ACCESS.2021.3073107","article-title":"Multi-scale spatial temporal graph neural network for skeleton-based action recognition","volume":"9","author":"Feng","year":"2021","journal-title":"IEEE Access"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"109172","DOI":"10.1016\/j.engappai.2024.109172","article-title":"Dynamic graph topology generating mechanism: Framework for feature-level multimodal information fusion applied to lower-limb activity recognition","volume":"137","author":"Yu","year":"2024","journal-title":"Eng. Appli. Artif. Intell."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"107761","DOI":"10.1016\/j.engappai.2023.107761","article-title":"Interpretable Dual-branch EMGNet: A transfer learning-based network for inter-subject lower limb motion intention recognition","volume":"130","author":"Zhang","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"3197","DOI":"10.1007\/s10115-022-01756-8","article-title":"Interpretable deep learning: Interpretation, interpretability, trustworthiness, and beyond","volume":"64","author":"Li","year":"2022","journal-title":"Knowl. Inf. Syst."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22\u201329). Grad-cam: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Ravanelli, M., and Bengio, Y. (2018, January 18\u201321). Speaker recognition from raw waveform with sincnet. Proceedings of the 2018 IEEE Spoken Language Technology Workshop (SLT), Athens, Greece.","DOI":"10.1109\/SLT.2018.8639585"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"4499","DOI":"10.1109\/TCE.2023.3333421","article-title":"Explainable AI-powered Graph Neural Networks for HD EMG-Based Gesture Intention Recognition","volume":"70","author":"Massa","year":"2024","journal-title":"IEEE Trans. Consum. Electron."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"108053","DOI":"10.1016\/j.knosys.2021.108053","article-title":"XAI for myo-controlled prosthesis: Explaining EMG data for hand gesture classification","volume":"240","author":"Gozzi","year":"2022","journal-title":"Knowl.-Based Syst."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"404","DOI":"10.1109\/TNSRE.2024.3523943","article-title":"Decoding gestures in electromyography: Spatiotemporal Graph neural networks for generalizable and interpretable classification","volume":"33","author":"Lee","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabi. Eng."},{"key":"ref_31","unstructured":"Dauphin, Y.N., Fan, A., Auli, M., and Grangier, D. (2017, January 6\u201311). Language modeling with gated convolutional networks. Proceedings of the 34th International Conference on Machine Learning (ICML), Sydney, Australia."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"102210","DOI":"10.1016\/j.bspc.2020.102210","article-title":"EMG hand gesture classification using handcrafted and deep features","volume":"63","author":"Fajardo","year":"2021","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"5246","DOI":"10.1109\/JSEN.2023.3344700","article-title":"Enhancing gesture classification using active EMG band and advanced feature extraction technique","volume":"24","author":"Rani","year":"2023","journal-title":"IEEE Sens. J."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"101872","DOI":"10.1016\/j.bspc.2020.101872","article-title":"Surface EMG signal classification using ternary pattern and discrete wavelet transform based feature extraction for hand movement recognition","volume":"58","author":"Tuncer","year":"2020","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_35","unstructured":"Lundberg, S. (2017). A unified approach to interpreting model predictions. arXiv."},{"key":"ref_36","unstructured":"Veli\u010dkovi\u0107, P., Cucurull, G., Casanova, A., Romero, A., Li\u00f2, P., and Bengio, Y. (2017). Graph attention networks. arXiv."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Hu, Y., Wong, Y., Wei, W., Du, Y., Kankanhalli, M., and Geng, W. (2018). A novel attention-based hybrid CNN-RNN architecture for sEMG-based gesture recognition. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0206049"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Chen, J., Bi, S., Zhang, G., and Cai, G. (2020). High-density surface EMG-based gesture recognition using a 3D convolutional neural network. Sensors, 20.","DOI":"10.3390\/s20041201"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"100308","DOI":"10.1016\/j.medntd.2024.100308","article-title":"A memory-friendly class-incremental learning method for hand gesture recognition using HD-sEMG","volume":"22","author":"Bai","year":"2024","journal-title":"Med. Nov. Technol. Devices"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"2514716","DOI":"10.1109\/TIM.2024.3381288","article-title":"Surface EMG-based inter-session\/inter-subject gesture recognition by leveraging lightweight all-ConvNet and transfer learning","volume":"73","author":"Islam","year":"2024","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"3469","DOI":"10.1007\/s10055-022-00741-4","article-title":"A low-cost AR application to control arm prosthesis","volume":"27","author":"Costa","year":"2023","journal-title":"Virtual Real."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/14\/4602\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T04:16:10Z","timestamp":1784780170000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/14\/4602"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,20]]},"references-count":41,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["s26144602"],"URL":"https:\/\/doi.org\/10.3390\/s26144602","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,20]]}}}