{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T06:17:20Z","timestamp":1790835440030,"version":"4.1.0"},"reference-count":29,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Bioeng. Biotechnol."],"abstract":"<jats:p>\n                    Three-dimensional ground reaction force (GRF) is an important biomechanical indicator of weight-bearing, propulsion, and bilateral asymmetry in hemiplegic gait. However, conventional GRF measurement relies on laboratory-based force plates, limiting its use in continuous rehabilitation assessment. A specific methodological challenge is to estimate continuous three-dimensional GRF in post-stroke hemiplegic gait without using force-plate signals as model input, while still preserving whole-body kinematic coordination and affected\u2013unaffected side asymmetry. This study proposed a force-plate-independent, marker-based method for estimating continuous stance-phase three-dimensional GRF in patients with hemiplegia by combining a spatio-temporal graph convolutional network (ST-GCN) with two-stage transfer learning. Data were collected from 30 chronic stroke patients with hemiplegia and 60 healthy controls. The model used 39 raw Plug-in Gait markers as graph nodes, with 10-dimensional node features consisting of three-dimensional position, velocity, acceleration, and laterality encoding. The model was pretrained using healthy participant data and then fine-tuned and evaluated on hemiplegic gait data using leave-one-subject-out cross-validation. The main contribution of this work is the integration of marker-level body topology, explicit kinematic derivatives, pathological laterality encoding, and healthy-to-hemiplegic transfer learning within a unified ST-GCN framework. The proposed ST-GCN achieved Pearson\u2019s\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m1\">\n                        <mml:mrow>\n                          <mml:mi>r<\/mml:mi>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    values of 0.984, 0.956, and 0.912, and rRMSE values of 5.24%, 8.15%, and 11.05% for vertical, anterior\u2013posterior, and medio\u2013lateral GRF prediction, respectively, outperforming MLP, 2D-CNN, BiLSTM, and lightweight Transformer baselines. Bland\u2013Altman analysis further showed small mean biases for the first vertical peak force, anterior\u2013posterior peak propulsive force, weight-bearing asymmetry index, and propulsion asymmetry index, with all participants falling within the 95% limits of agreement. These findings suggest that the proposed framework can reconstruct overall GRF waveform morphology and preserve group-level kinetic asymmetry features in hemiplegic gait. The method may provide a force-plate-independent, marker-based laboratory framework for kinetic gait assessment, but it should not yet be interpreted as a fully wearable or home-based clinical monitoring system.\n                  <\/jats:p>","DOI":"10.3389\/fbioe.2026.1871692","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:06:01Z","timestamp":1784095561000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Spatio-temporal graph convolutional networks with transfer learning for continuous ground reaction force estimation in hemiparetic gait"],"prefix":"10.3389","volume":"14","author":[{"given":"Qinghua","family":"Meng","sequence":"first","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiyuan","family":"Yang","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yijia","family":"Xue","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luxing","family":"Zhou","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nan","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miaomiao","family":"Xiao","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuequan","family":"Feng","sequence":"additional","affiliation":[{"name":"The Neurosurgical Department of Tianjin First Central Hospital","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunyu","family":"Bao","sequence":"additional","affiliation":[{"name":"Tianjin University of Sport","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1016\/j.apmr.2006.10.004","article-title":"Relationship between step length asymmetry and walking performance in subjects with chronic hemiparesis","volume":"88","author":"Balasubramanian","year":"2007","journal-title":"Archives Phys. Med. Rehabilitation"},{"key":"B2","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1186\/s12984-025-01620-0","article-title":"Gait speed-dependent modulation of paretic versus non-paretic propulsion in persons with chronic stroke","volume":"22","author":"Biere","year":"2025","journal-title":"J. NeuroEngineering Rehabilitation"},{"key":"B3","doi-asserted-by":"publisher","first-page":"872","DOI":"10.1161\/01.STR.0000204063.75779.8d","article-title":"Anterior-posterior ground reaction forces as a measure of paretic leg contribution in hemiparetic walking","volume":"37","author":"Bowden","year":"2006","journal-title":"Stroke"},{"key":"B4","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1016\/j.gaitpost.2009.03.011","article-title":"Reliability, smallest real difference and concurrent validity of indices computed from GRF components in gait of stroke patients","volume":"30","author":"Campanini","year":"2009","journal-title":"Gait and Posture"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1940","DOI":"10.1109\/TBME.2007.901024","article-title":"Opensim: open-source software to create and analyze dynamic simulations of movement","volume":"54","author":"Delp","year":"2007","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"B6","doi-asserted-by":"publisher","first-page":"588","DOI":"10.3390\/bioengineering12060588","article-title":"Estimation of three-dimensional ground reaction force and center of pressure during walking using a machine-learning-based markerless motion capture system","volume":"12","author":"Feng","year":"2025","journal-title":"Bioengineering"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"B8","doi-asserted-by":"publisher","first-page":"75","DOI":"10.3390\/s17010075","article-title":"Estimation of ground reaction forces and moments during gait using only inertial motion capture","volume":"17","author":"Karatsidis","year":"2017","journal-title":"Sensors"},{"key":"B9","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1016\/j.gaitpost.2010.11.024","article-title":"Minimal detectable change for gait variables collected during treadmill walking in individuals post-stroke","volume":"33","author":"Kesar","year":"2011","journal-title":"Gait and Posture"},{"key":"B10","doi-asserted-by":"publisher","first-page":"138932","DOI":"10.1109\/ACCESS.2024.3434513","article-title":"Deep learning for quantified gait analysis: a systematic literature review","volume":"12","author":"Khan","year":"2024","journal-title":"IEEE Access"},{"key":"B11","doi-asserted-by":"publisher","first-page":"45672","DOI":"10.1038\/s41598-025-25146-6","article-title":"Leveraging blockchain with zero knowledge proofs in wearable health technologies for personalized healthcare","volume":"15","author":"Khan","year":"","journal-title":"Sci. Rep."},{"key":"B12","doi-asserted-by":"publisher","first-page":"17305","DOI":"10.1038\/s41598-025-01408-1","article-title":"A cost-effective approach using generative AI and gamification to enhance biomedical treatment and real-time biosensor monitoring","volume":"15","author":"Khan","year":"","journal-title":"Sci. Rep."},{"key":"B13","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1016\/S0966-6362(02)00122-4","article-title":"Symmetry in vertical ground reaction force is accompanied by symmetry in temporal but not distance variables of gait in persons with stroke","volume":"18","author":"Kim","year":"2003","journal-title":"Gait and Posture"},{"key":"B14","doi-asserted-by":"publisher","first-page":"140","DOI":"10.1016\/j.gaitpost.2003.07.002","article-title":"Magnitude and pattern of 3d kinematic and kinetic gait profiles in persons with stroke: relationship to walking speed","volume":"20","author":"Kim","year":"2004","journal-title":"Gait and Posture"},{"key":"B15","doi-asserted-by":"publisher","first-page":"3428","DOI":"10.3390\/s23073428","article-title":"A deep learning model for 3d ground reaction force estimation using shoes with three uniaxial load cells","volume":"23","author":"Kim","year":"2023","journal-title":"Sensors"},{"key":"B16","doi-asserted-by":"publisher","first-page":"3357","DOI":"10.3390\/s25113357","article-title":"Estimation of 3d ground reaction force and 2d center of pressure using deep learning and load cells across various gait conditions","volume":"25","author":"Kim","year":"2025","journal-title":"Sensors"},{"key":"B17","doi-asserted-by":"publisher","first-page":"196","DOI":"10.1186\/s12984-025-01728-3","article-title":"Effective assistance timing using characteristics of paretic ground reaction force in ankle robotic gait training for individuals with hemiparesis after stroke: a cross-sectional study","volume":"22","author":"Kuwabara","year":"2025","journal-title":"J. NeuroEngineering Rehabilitation"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1199459","DOI":"10.3389\/fbioe.2023.1199459","article-title":"Estimating vertical ground reaction forces during gait from lower limb kinematics and vertical acceleration using wearable inertial sensors","volume":"11","author":"Mart\u00ednez-Pascual","year":"2023","journal-title":"Front. Bioeng. Biotechnol."},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.36227\/techrxiv.175951373.33024130\/v1","article-title":"Real-time estimation of anterior-posterior ground reaction forces in individuals with stroke using wearable sensors and deep learning","author":"Nurse","year":"2025","journal-title":"TechRxiv"},{"key":"B21","doi-asserted-by":"publisher","first-page":"2372","DOI":"10.1016\/j.jbiomech.2013.07.036","article-title":"Prediction of ground reaction forces during gait based on kinematics and a neural network model","volume":"46","author":"Oh","year":"2013","journal-title":"J. Biomechanics"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"B23","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1186\/s12984-020-00700-7","article-title":"Indirect measurement of anterior-posterior ground reaction forces using a minimal set of wearable inertial sensors: from healthy to hemiparetic walking","volume":"17","author":"Revi","year":"2020","journal-title":"J. NeuroEngineering Rehabilitation"},{"key":"B24","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2273","DOI":"10.1109\/TNSRE.2025.3577961","article-title":"Multi-modal sensing for propulsion estimation in people post-stroke across speeds","volume":"33","author":"Swaminathan","year":"2025","journal-title":"IEEE Trans. Neural Syst. Rehabilitation Eng."},{"key":"B26","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B27","doi-asserted-by":"publisher","first-page":"7444","DOI":"10.1609\/aaai.v32i1.12328","article-title":"Spatial temporal graph convolutional networks for skeleton-based action recognition","volume":"32","author":"Yan","year":"2018","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"B28","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1007\/s44196-024-00603-8","article-title":"Auto-correlation and channel attention enhanced deep graph convolution networks for gait phase prediction based on multi-IMU system","volume":"17","author":"Yan","year":"2024","journal-title":"Int. J. Comput. Intell. Syst."},{"key":"B29","doi-asserted-by":"publisher","first-page":"3352","DOI":"10.3390\/s24113352","article-title":"FP-GCN: frequency pyramid graph convolutional network for enhancing pathological gait classification","volume":"24","author":"Zhao","year":"2024","journal-title":"Sensors"}],"container-title":["Frontiers in Bioengineering and Biotechnology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1871692\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:06:02Z","timestamp":1784095562000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1871692\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,15]]},"references-count":29,"alternative-id":["10.3389\/fbioe.2026.1871692"],"URL":"https:\/\/doi.org\/10.3389\/fbioe.2026.1871692","relation":{},"ISSN":["2296-4185"],"issn-type":[{"value":"2296-4185","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,15]]},"article-number":"1871692"}}