{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T08:10:46Z","timestamp":1793175046595,"version":"4.5.0"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,11,13]],"date-time":"2025-11-13T00:00:00Z","timestamp":1762992000000},"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. Neurol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>This study aimed to develop an automated approach for assessing upper limb (UL) motor impairment severity in stroke patients using a deep learning framework applied to resting-state functional magnetic resonance imaging (rs-fMRI).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      Dynamic functional connectivity (dFC) was computed with the ipsilesional primary motor cortex (M1) as a seed and extracted from rs-fMRI data of 69 stroke patients. These dFC features were used to train a three-dimensional convolutional neural network (3D-CNN) for automatic classification of UL motor impairment severity. Patients were divided into two groups according to UL Fugl-Meyer Assessment (UL-FMA) scores: mild-to-moderate impairment (UL-FMA\u202f&amp;gt;\u202f20;\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f29, maximum\u202f=\u202f66) and severe impairment (0\u202f\u2264\u202fUL-FMA\u202f\u2264\u202f20;\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f40). UL-FMA scores served as labels for supervised learning.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The model achieved a balanced accuracy of 99.8%\u202f\u00b1\u202f0.2%, with a specificity of 99.9%\u202f\u00b1\u202f0.2% and a sensitivity of 99.7%\u202f\u00b1\u202f0.3%. Several brain regions\u2014including the angular gyrus, medial orbitofrontal cortex, dorsolateral superior frontal gyrus, superior parietal lobule, supplementary motor area, thalamus, cerebellum, and middle temporal gyrus\u2014were linked to UL motor impairment severity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These findings demonstrate that a 3D deep learning framework based on dFC features from rs-fMRI enables highly accurate and objective classification of UL motor impairment in stroke patients. This approach may provide a valuable alternative to manual UL-FMA scoring, particularly in clinical settings with limited access to experienced evaluators.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fneur.2025.1666991","type":"journal-article","created":{"date-parts":[[2025,11,13]],"date-time":"2025-11-13T12:45:51Z","timestamp":1763037951000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Assessment of functional decline in stroke patients using 3D deep learning and dynamic functional connectivity based on resting-state fMRI"],"prefix":"10.3389","volume":"16","author":[{"given":"Yingying","family":"Gao","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guojun","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Peng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengbin","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shifei","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Minmin","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hewei","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiyong","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,11,13]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"528","DOI":"10.1016\/S1474-4422(04)00851-8","article-title":"Strategies for stroke rehabilitation","volume":"3","author":"Dobkin","year":"2004","journal-title":"Lancet Neurol"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"1480","DOI":"10.1161\/01.STR.0000170706.13595.4f","article-title":"Factors influencing stroke survivors\u2019 quality of life during subacute recovery","volume":"36","author":"Nichols-Larsen","year":"2005","journal-title":"Stroke"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1224","DOI":"10.1093\/brain\/awq043","article-title":"Dynamic functional reorganization of the motor execution network after stroke","volume":"133","author":"Wang","year":"2010","journal-title":"Brain"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"3388","DOI":"10.1002\/hbm.24183","article-title":"Altered intra-and inter-network functional coupling of resting-state networks associated with motor dysfunction in stroke","volume":"39","author":"Zhao","year":"2018","journal-title":"Hum Brain Mapp"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"2527","DOI":"10.1016\/j.celrep.2019.07.100","article-title":"Structural disconnections explain brain network dysfunction after stroke","volume":"28","author":"Griffis","year":"2019","journal-title":"Cell Rep"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1016\/S1474-4422(14)70160-7","article-title":"Constraint-induced movement therapy after stroke","volume":"14","author":"Kwakkel","year":"2015","journal-title":"Lancet Neurol"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1080\/10749357.2018.1466971","article-title":"Effect of intensive motor training with repetitive transcranial magnetic stimulation on upper limb motor function in chronic post-stroke patients with severe upper limb motor impairment","volume":"25","author":"Hirakawa","year":"2018","journal-title":"Top Stroke Rehabil"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1177\/154596802401105171","article-title":"The Fugl-Meyer assessment of motor recovery after stroke: a critical review of its measurement properties","volume":"16","author":"Gladstone","year":"2002","journal-title":"Neurorehabil Neural Repair"},{"key":"ref9","doi-asserted-by":"crossref","first-page":"13","DOI":"10.2340\/1650197771331","article-title":"The post stroke hemiplegic patient. I. A method for evaluation of physical performance","volume":"7","author":"Meyer","year":"1975","journal-title":"Scand J Rehabil Med"},{"key":"ref10","doi-asserted-by":"crossref","DOI":"10.1109\/EMBC.2012.6346340","article-title":"Full body gait analysis with Kinect","author":"Gabel","year":"2012"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"641","DOI":"10.3233\/THC-130764","article-title":"Evaluation of upper extremity reachable workspace using Kinect camera","volume":"21","author":"Kurillo","year":"2013","journal-title":"Technol Health Care"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1357","DOI":"10.1161\/STROKEAHA.110.596155","article-title":"Longitudinal changes of resting-state functional connectivity during motor recovery after stroke","volume":"42","author":"Park","year":"2011","journal-title":"Stroke"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"341","DOI":"10.1177\/1545968314547766","article-title":"Comparison of three tools to measure improvements in upper-limb function with poststroke therapy","volume":"29","author":"Thompson-Butel","year":"2015","journal-title":"Neurorehabil Neural Repair"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"15065","DOI":"10.1523\/JNEUROSCI.2364-11.2011","article-title":"Focal pontine lesions provide evidence that intrinsic functional connectivity reflects polysynaptic anatomical pathways","volume":"31","author":"Lu","year":"2011","journal-title":"J Neurosci"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"4495","DOI":"10.1523\/JNEUROSCI.3662-11.2012","article-title":"Extent of bilateral neuronal network reorganization and functional recovery in relation to stroke severity","volume":"32","author":"van Meer","year":"2012","journal-title":"J Neurosci"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"e0166210","DOI":"10.1371\/journal.pone.0166210","article-title":"Altered effective connectivity of the primary motor cortex in stroke: a resting-state fMRI study with granger causality analysis","volume":"11","author":"Zhao","year":"2016","journal-title":"PLoS One"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1211","DOI":"10.1093\/brain\/awab469","article-title":"Functional connectivity drives stroke recovery: shifting the paradigm from correlation to causation","volume":"145","author":"Cassidy","year":"2022","journal-title":"Brain"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1002\/ana.21905","article-title":"Resting interhemispheric functional magnetic resonance imaging connectivity predicts performance after stroke","volume":"67","author":"Carter","year":"2010","journal-title":"Ann Neurol"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"663","DOI":"10.1093\/cercor\/bhs352","article-title":"Tracking whole-brain connectivity dynamics in the resting state","volume":"24","author":"Allen","year":"2014","journal-title":"Cereb Cortex"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"262","DOI":"10.1016\/j.neuron.2014.10.015","article-title":"The chronnectome: time-varying connectivity networks as the next frontier in fMRI data discovery","volume":"84","author":"Calhoun","year":"2014","journal-title":"Neuron"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1207","DOI":"10.1002\/hbm.23910","article-title":"Differential patterns of dynamic functional connectivity variability of striato\u2013cortical circuitry in children with benign epilepsy with centrotemporal spikes","volume":"39","author":"Li","year":"2018","journal-title":"Hum Brain Mapp"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1016\/j.neulet.2018.09.008","article-title":"Alterations of static functional connectivity and dynamic functional connectivity in motor execution regions after stroke","volume":"686","author":"Chen","year":"2018","journal-title":"Neurosci Lett"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"298","DOI":"10.1016\/j.nicl.2014.07.003","article-title":"Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia","volume":"5","author":"Damaraju","year":"2014","journal-title":"NeuroImage"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1007\/s12021-019-09419-w","article-title":"3D-deep learning based automatic diagnosis of Alzheimer\u2019s disease with joint MMSE prediction using resting-state fMRI","volume":"18","author":"Duc","year":"2020","journal-title":"Neuroinformatics"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"16376","DOI":"10.1109\/ACCESS.2023.3244952","article-title":"AlzheimerNet: an effective deep learning based proposition for alzheimer\u2019s disease stages classification from functional brain changes in magnetic resonance images","volume":"11","author":"Shamrat","year":"2023","journal-title":"IEEE Access"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.artmed.2019.06.003","article-title":"3D-CNN based discrimination of schizophrenia using resting-state fMRI","volume":"98","author":"Qureshi","year":"2019","journal-title":"Artif Intell Med"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"103015","DOI":"10.1016\/j.bspc.2021.103015","article-title":"Using DeepGCN to identify the autism spectrum disorder from multi-site resting-state data","volume":"70","author":"Cao","year":"2021","journal-title":"Biomed Signal Process Control"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"8","DOI":"10.3389\/fnagi.2019.00008","article-title":"Evaluation of functional decline in Alzheimer\u2019s dementia using 3D deep learning and group ICA for rs-fMRI measurements","volume":"11","author":"Qureshi","year":"2019","journal-title":"Front Aging Neurosci"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"106240","DOI":"10.1016\/j.compbiomed.2022.106240","article-title":"Multi-label classification of Alzheimer's disease stages from resting-state fMRI-based correlation connectivity data and deep learning","volume":"151","author":"Alorf","year":"2022","journal-title":"Comput Biol Med"},{"key":"ref30","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1109\/TPAMI.2012.59","article-title":"3D convolutional neural networks for human action recognition","volume":"35","author":"Ji","year":"2012","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"456","DOI":"10.1016\/j.apmr.2016.06.023","article-title":"Determining levels of upper extremity movement impairment by applying a cluster analysis to the Fugl-Meyer assessment of the upper extremity in chronic stroke","volume":"98","author":"Woytowicz","year":"2017","journal-title":"Arch Phys Med Rehabil"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1089\/brain.2012.0080","article-title":"Trouble at rest: how correlation patterns and group differences become distorted after global signal regression","volume":"2","author":"Saad","year":"2012","journal-title":"Brain Connect"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"5579","DOI":"10.1523\/JNEUROSCI.4903-14.2015","article-title":"Network structure shapes spontaneous functional connectivity dynamics","volume":"35","author":"Shen","year":"2015","journal-title":"J Neurosci"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1016\/j.neuroimage.2010.06.003","article-title":"Cost function masking during normalization of brains with focal lesions: still a necessity?","volume":"53","author":"Andersen","year":"2010","journal-title":"NeuroImage"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"339","DOI":"10.1007\/s12021-016-9299-4","article-title":"DPABI: data processing and analysis for (resting-state) brain imaging","volume":"14","author":"Yan","year":"2016","journal-title":"Neuroinformatics"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1016\/j.neuroimage.2015.11.055","article-title":"Can sliding-window correlations reveal dynamic functional connectivity in resting-state fMRI?","volume":"127","author":"Hindriks","year":"2016","journal-title":"NeuroImage"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"e52727","DOI":"10.1371\/journal.pone.0052727","article-title":"Patterns in cortical connectivity for determining outcomes in hand function after subcortical stroke","volume":"7","author":"Yin","year":"2012","journal-title":"PLoS One"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"3343","DOI":"10.1002\/hbm.22406","article-title":"Altered topological properties of the cortical motor-related network in patients with subcortical stroke revealed by graph theoretical analysis","volume":"35","author":"Yin","year":"2014","journal-title":"Hum Brain Mapp"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1007\/s10334-010-0197-8","article-title":"A method for evaluating dynamic functional network connectivity and task-modulation: application to schizophrenia","volume":"23","author":"Sako\u011flu","year":"2010","journal-title":"MAGMA"},{"key":"ref40","doi-asserted-by":"crossref","DOI":"10.4324\/9780203774441","volume-title":"Applied multiple regression\/correlation analysis for the behavioral sciences","author":"Cohen","year":"2013"},{"key":"ref41","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref42","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014"},{"key":"ref43","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2015.123","article-title":"Delving deep into rectifiers: surpassing human-level performance on imagenet classification","author":"He","year":"2015"},{"key":"ref44","doi-asserted-by":"crossref","DOI":"10.1007\/3-540-44503-X_27","article-title":"On the surprising behavior of distance metrics in high dimensional space","author":"Aggarwal","year":"2001"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"e0129569","DOI":"10.1371\/journal.pone.0129569","article-title":"Multiclass support vector machine-based lesion mapping predicts functional outcome in ischemic stroke patients","volume":"10","author":"Forkert","year":"2015","journal-title":"PLoS One"},{"key":"ref46","doi-asserted-by":"publisher","first-page":"4553","DOI":"10.1002\/hbm.22936","article-title":"Individual prediction of chronic motor outcome in the acute post-stroke stage: behavioral parameters versus functional imaging","volume":"36","author":"Rehme","year":"2015","journal-title":"Hum Brain Mapp"},{"key":"ref47","doi-asserted-by":"publisher","first-page":"372","DOI":"10.1016\/j.nicl.2016.07.014","article-title":"Decoding post-stroke motor function from structural brain imaging","volume":"12","author":"Rondina","year":"2016","journal-title":"NeuroImage"},{"key":"ref48","doi-asserted-by":"publisher","first-page":"428","DOI":"10.1177\/1545968320909796","article-title":"Machine learning methods predict individual upper-limb motor impairment following therapy in chronic stroke","volume":"34","author":"Tozlu","year":"2020","journal-title":"Neurorehabil Neural Repair"},{"key":"ref49","doi-asserted-by":"publisher","first-page":"1077","DOI":"10.1161\/STROKEAHA.113.003168","article-title":"Upper limb recovery after stroke is associated with ipsilesional primary motor cortical activity: a meta-analysis","volume":"45","author":"Favre","year":"2014","journal-title":"Stroke"},{"key":"ref50","doi-asserted-by":"publisher","first-page":"596","DOI":"10.1002\/ana.21636","article-title":"Early imaging correlates of subsequent motor recovery after stroke","volume":"65","author":"Marshall","year":"2009","journal-title":"Ann Neurol"},{"key":"ref51","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky","year":"2012","journal-title":"Adv Neural Inf Process Syst"},{"key":"ref52","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1016\/j.neuroimage.2016.12.061","article-title":"The dynamic functional connectome: state-of-the-art and perspectives","volume":"160","author":"Preti","year":"2017","journal-title":"NeuroImage"}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2025.1666991\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,13]],"date-time":"2025-11-13T12:45:52Z","timestamp":1763037952000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2025.1666991\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,13]]},"references-count":52,"alternative-id":["10.3389\/fneur.2025.1666991"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2025.1666991","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,13]]},"article-number":"1666991"}}