{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T06:03:03Z","timestamp":1790661783863,"version":"4.1.0"},"reference-count":29,"publisher":"Wiley","issue":"10","license":[{"start":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T00:00:00Z","timestamp":1790553600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T00:00:00Z","timestamp":1790553600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/100009619","name":"Japan Agency for Medical Research and Development","doi-asserted-by":"publisher","award":["JP256f0137006"],"award-info":[{"award-number":["JP256f0137006"]}],"id":[{"id":"10.13039\/100009619","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Brain and Behavior"],"published-print":{"date-parts":[[2026,10]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Ataxia severity in degenerative cerebellar diseases (DCDs) is usually assessed with semi\u2010quantitative clinical rating scales such as the Scale for the Assessment and Rating of Ataxia (SARA), which are subject to variability. Deep learning\u2010based gait analysis offers an objective alternative, but labeled data in DCDs are limited. We investigated whether pretraining on gait videos from patients with Parkinson's disease (PD) could improve SARA score prediction in DCD.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      Gait videos from patients with DCD were processed using pose estimation, and the extracted time\u2010series keypoint coordinates were input to a transformer\u2010based model. Separate models were developed for the SARA total score and the posture and gait subscore (sum of Items 1\u20133). Models were pretrained on PD gait videos by supervised learning with the Movement Disorder Society\u2013Sponsored Unified Parkinson's Disease Rating Scale (MDS\u2013UPDRS) Part III axial subscore or by self\u2010supervised masked keypoint reconstruction. Performance was evaluated by leave\u2010one\u2010participant\u2010out cross\u2010validation (mean absolute error [MAE], coefficient of determination [\n                      <jats:italic>R<\/jats:italic>\n                      <jats:sup>2<\/jats:sup>\n                      ]), and between\u2010model differences by participant\u2010level Wilcoxon signed\u2010rank tests and bootstrap 95% confidence intervals (CIs).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      We analyzed 75 patients with DCD and 141 with PD. For the SARA total score, MAE and\n                      <jats:italic>R<\/jats:italic>\n                      <jats:sup>2<\/jats:sup>\n                      were 1.91 \u00b1 0.06 and 0.78 \u00b1 0.01 with supervised pretraining, 1.95 \u00b1 0.04 and 0.77 \u00b1 0.01 with self\u2010supervised pretraining, and 2.04 \u00b1 0.07 and 0.74 \u00b1 0.02 with training from scratch; neither strategy significantly reduced participant\u2010level prediction error (\n                      <jats:italic>p<\/jats:italic>\n                      = 0.25 and 0.24). For the posture and gait subscore, only supervised pretraining significantly reduced prediction error (mean difference: \u22120.078; 95% CI: \u22120.127 to \u22120.031;\n                      <jats:italic>p<\/jats:italic>\n                      = 0.003).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Pretraining on PD gait data modestly reduced SARA prediction error, reaching statistical significance for the posture and gait subscore with supervised pretraining. Pretraining across neurological disorders may help models acquire transferable representations of pathological gait, but its clinical significance requires validation in larger, multicenter datasets.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/brb3.71781","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:29:33Z","timestamp":1790659773000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Improving Video\u2010Based Prediction of Cerebellar Ataxia Severity Using a Pretrained Deep Learning Model"],"prefix":"10.1002","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2988-2532","authenticated-orcid":false,"given":"Katsuki","family":"Eguchi","sequence":"first","affiliation":[{"name":"Department of Neurology Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"},{"name":"Division of AI Support for Medical Research Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hiroaki","family":"Yaguchi","sequence":"additional","affiliation":[{"name":"Department of Neurology Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hisashi","family":"Uwatoko","sequence":"additional","affiliation":[{"name":"Department of Neurology Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuki","family":"Iida","sequence":"additional","affiliation":[{"name":"Hokuyukai Neurological Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Moemi","family":"Yamada","sequence":"additional","affiliation":[{"name":"Sapporo Neurology Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shinsuke","family":"Hamada","sequence":"additional","affiliation":[{"name":"Hokuyukai Neurological Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kazunori","family":"Sato","sequence":"additional","affiliation":[{"name":"Sapporo Neurology Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanae","family":"Honma","sequence":"additional","affiliation":[{"name":"Hokuyukai Neurological Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Asako","family":"Takei","sequence":"additional","affiliation":[{"name":"Hokuyukai Neurological Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Atsushi","family":"Kawashima","sequence":"additional","affiliation":[{"name":"Sapporo Neurology Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fumio","family":"Moriwaka","sequence":"additional","affiliation":[{"name":"Hokuyukai Neurological Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Toshiyuki","family":"Fukazawa","sequence":"additional","affiliation":[{"name":"Sapporo Neurology Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ren","family":"Togo","sequence":"additional","affiliation":[{"name":"Faculty of Information Science and Technology Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takahiro","family":"Ogawa","sequence":"additional","affiliation":[{"name":"Faculty of Information Science and Technology Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miki","family":"Haseyama","sequence":"additional","affiliation":[{"name":"Faculty of Information Science and Technology Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kenji","family":"Hirata","sequence":"additional","affiliation":[{"name":"Division of AI Support for Medical Research Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"},{"name":"Department of Diagnostic Imaging Faculty of Medicine Hokkaido University  Sapporo Japan"},{"name":"Global Center for Biomedical Science and Engineering Faculty of Medicine Hokkaido University  Sapporo Japan"},{"name":"Medical AI Research and Development Center Hokkaido University Hospital  Sapporo Japan"},{"name":"Healthcare AIX Innovation Center (HAIXIC) Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kohsuke","family":"Kudo","sequence":"additional","affiliation":[{"name":"Department of Diagnostic Imaging Faculty of Medicine Hokkaido University  Sapporo Japan"},{"name":"Global Center for Biomedical Science and Engineering Faculty of Medicine Hokkaido University  Sapporo Japan"},{"name":"Medical AI Research and Development Center Hokkaido University Hospital  Sapporo Japan"},{"name":"Healthcare AIX Innovation Center (HAIXIC) Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shinya","family":"Tanaka","sequence":"additional","affiliation":[{"name":"Department of Cancer Pathology Faculty of Medicine Hokkaido University  Sapporo Japan"},{"name":"Institute for Chemical Reaction Design and Discovery (WPI\u2010ICReDD) Hokkaido University  Sapporo Japan"},{"name":"Department of Surgical Pathology Hokkaido University Hospital  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ichiro","family":"Yabe","sequence":"additional","affiliation":[{"name":"Department of Neurology Faculty of Medicine and Graduate School of Medicine Hokkaido University  Sapporo Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,9,28]]},"reference":[{"key":"e_1_2_12_2_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41531\u2010025\u201000876\u20106"},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.1002\/mds.29439"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.gaitpost.2017.11.024"},{"key":"e_1_2_12_5_1","doi-asserted-by":"publisher","DOI":"10.1111\/jnc.12318"},{"key":"e_1_2_12_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2929257"},{"key":"e_1_2_12_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41531\u2010024\u201000742\u2010x"},{"key":"#cr-split#-e_1_2_12_8_1.1","doi-asserted-by":"crossref","unstructured":"Devlin J. M. W.Chang K.Lee andK.Toutanova.2019. \"BERT: Pre\u2010Training of Deep Bidirectional Transformers for Language Understanding.\" InProceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies Volume 1 (Long and Short Papers)","DOI":"10.18653\/v1\/N19-1423"},{"key":"#cr-split#-e_1_2_12_8_1.2","unstructured":"June 2-7 2019 J.Burstein C.Doran andT.Solorio 4171-4186.Association for Computational Linguistics.https:\/\/doi.org\/10.18653\/v1\/N19\u20101423."},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jns.2024.123158"},{"key":"e_1_2_12_10_1","doi-asserted-by":"publisher","DOI":"10.1002\/mds.30113"},{"key":"e_1_2_12_11_1","doi-asserted-by":"publisher","DOI":"10.1001\/jamaneurol.2024.5326"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/mds.22340"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010020\u201017807\u2010z"},{"key":"e_1_2_12_14_1","doi-asserted-by":"publisher","DOI":"10.1007\/s40120\u2010024\u201000625\u20106"},{"key":"e_1_2_12_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/mds.30327"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1474\u20104422(19)30044\u20104"},{"key":"e_1_2_12_17_1","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0000000000011654"},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.1212\/01.wnl.0000219042.60538.92"},{"key":"e_1_2_12_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mds.28740"},{"key":"e_1_2_12_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.parkreldis.2020.09.007"},{"key":"e_1_2_12_21_1","doi-asserted-by":"publisher","DOI":"10.3390\/s21217315"},{"key":"e_1_2_12_22_1","doi-asserted-by":"publisher","DOI":"10.1212\/01.WNL.0000156802.15466.79"},{"key":"e_1_2_12_23_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12880\u2010024\u201001253\u20100"},{"key":"e_1_2_12_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3337398"},{"key":"e_1_2_12_25_1","first-page":"5998","article-title":"Attention Is All You Need","volume":"30","author":"Vaswani A.","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_12_26_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586\u2010024\u201007441\u2010w"},{"key":"e_1_2_12_27_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591\u2010025\u201003747\u2010y"},{"key":"e_1_2_12_28_1","doi-asserted-by":"crossref","unstructured":"Zerveas G. S.Jayaraman D.Patel A.Bhamidipaty andC.Eickhoff.2021. \u201cA Transformer\u2010Based Framework for Multivariate Time Series Representation Learning.\u201d InProceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining; August 14\u201318 2021 2114\u20132124.ACM.https:\/\/doi.org\/10.1145\/3447548.3467401.","DOI":"10.1145\/3447548.3467401"},{"key":"e_1_2_12_29_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i12.17325"}],"container-title":["Brain and Behavior"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/brb3.71781","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/brb3.71781","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/brb3.71781","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:29:35Z","timestamp":1790659775000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/brb3.71781"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,28]]},"references-count":29,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2026,10]]}},"alternative-id":["10.1002\/brb3.71781"],"URL":"https:\/\/doi.org\/10.1002\/brb3.71781","archive":["Portico"],"relation":{},"ISSN":["2162-3279","2162-3279"],"issn-type":[{"value":"2162-3279","type":"print"},{"value":"2162-3279","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,28]]},"assertion":[{"value":"2026-06-25","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-07","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-28","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e71781"}}