{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T04:03:46Z","timestamp":1790741026797,"version":"4.1.0"},"reference-count":24,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T00:00:00Z","timestamp":1771286400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T00:00:00Z","timestamp":1771286400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["esskajournals.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Knee surg. sports traumatol. arthrosc."],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aims to test whether week\u2010long wrist accelerometry combined with deep learning can (i) distinguish healthy individuals from people with knee osteoarthritis (KOA), (ii) separate prodromal KOA from established KOA, and (iii) identify individuals who will receive a KOA diagnosis within 5 years.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We conducted a retrospective case\u2013control study using the UK Biobank data set. After quality control, 102,120 participants with valid accelerometry were available; KOA was identified via ICD\u201010 M17.x codes (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20097262). To reduce adiposity confounding, analyses were restricted to body mass index (BMI)\u2009\u2265\u200929, with controls matched to KOA on age, sex and BMI distributions. We used preprocessed, orientation\u2010independent, hourly mean acceleration over a 24\u2010h cycle and included month, sex, age and weight as covariates. A 1D convolutional neural network modelled daily activity profiles with embeddings for categorical covariates. Fivefold cross\u2010validation assessed accuracy, macro F1, macro sensitivity and AUC.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Balanced cohorts were formed for three tasks: healthy versus KOA (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20093677 per class), prodromal versus diagnosed KOA (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20091596 vs. 2081), and healthy vs prodromal within 5 years (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20091369 per class). Daily activity patterns were similar across groups, with slightly lower daytime acceleration in KOA\/prodromal participants. Model performance was moderate for healthy versus KOA (accuracy 63.5\u2009\u00b1\u20091.2%; AUC 0.672\u2009\u00b1\u20090.017) and healthy vs prodromal within 5 years (64.5\u2009\u00b1\u20090.5%; AUC 0.675\u2009\u00b1\u20090.019). Discrimination between prodromal and diagnosed KOA was close to random (54.6%\u2009\u00b1\u20091.5%; AUC 0.552\u2009\u00b1\u20090.015).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>One week of wrist\u2010worn accelerometry contains a reproducible signal associated with KOA and can flag elevated risk up to 5 years before diagnosis. Since existing KOA cannot be distinguished from prodromal KOA, it can be assumed that patients show altered movement patterns years before diagnosis. These findings highlight the clinical relevance of early, unobtrusive movement monitoring and support the potential of wearables as a scalable, low\u2010cost component of population\u2010level KOA screening.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Level of Evidence<\/jats:title>\n                    <jats:p>Level\u202fII, prognostic study\u2014lower\u2010quality prospective cohort. The study uses a large, population\u2010based prospective cohort (UK\u202fBiobank) with retrospective analytical methods; follow\u2010up is high, but the study is a secondary analysis rather than a primary prospectively designed prognostic trial.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/ksa.70332","type":"journal-article","created":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T10:50:41Z","timestamp":1771325441000},"page":"1524-1533","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Deep learning of wrist accelerometry from UK Biobank data identifies early movement signatures of knee osteoarthritis up to 5 years before diagnosis"],"prefix":"10.1002","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9825-302X","authenticated-orcid":false,"given":"Ricardo","family":"Smits Serena","sequence":"first","affiliation":[{"name":"Department of Orthopaedics and Sports Orthopaedics TUM University Hospital Munich Germany"},{"name":"Institute for AI and Informatics in Medicine TUM University Hospital Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael T.","family":"Hirschmann","sequence":"additional","affiliation":[{"name":"Kantonsspital Baselland Bruderholz Switzerland"},{"name":"Kantonsspital Baselland Liestal Switzerland"},{"name":"Kantonsspital Baselland Laufen Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georg","family":"Matziolis","sequence":"additional","affiliation":[{"name":"Waldkliniken Eisenberg GmbH Eisenberg Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"R\u00fcdiger","family":"von Eisenhart\u2010Rothe","sequence":"additional","affiliation":[{"name":"Department of Orthopaedics and Sports Orthopaedics TUM University Hospital Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Rueckert","sequence":"additional","affiliation":[{"name":"Institute for AI and Informatics in Medicine TUM University Hospital Munich Germany"},{"name":"Department of Computing Imperial College London London UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Florian","family":"Hinterwimmer","sequence":"additional","affiliation":[{"name":"Department of Orthopaedics and Sports Orthopaedics TUM University Hospital Munich Germany"},{"name":"Institute for AI and Informatics in Medicine TUM University Hospital Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christina","family":"Valle","sequence":"additional","affiliation":[{"name":"Department of Orthopaedics and Sports Orthopaedics TUM University Hospital Munich Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,2,17]]},"reference":[{"key":"e_1_2_12_2_1","doi-asserted-by":"publisher","DOI":"10.1126\/scitranslmed.adf4428"},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.3390\/informatics9040097"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1093\/pm\/pnaa096"},{"key":"e_1_2_12_5_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0169649"},{"key":"e_1_2_12_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2024.e36825"},{"key":"e_1_2_12_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ard.2025.01.019"},{"key":"e_1_2_12_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(19)30417-9"},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-019-00619-1"},{"key":"e_1_2_12_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00167-020-06154-7"},{"key":"e_1_2_12_11_1","doi-asserted-by":"publisher","DOI":"10.2196\/50035"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12984-017-0309-z"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.maturitas.2016.04.006"},{"key":"e_1_2_12_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/acr.21754"},{"key":"e_1_2_12_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.rbo.2015.08.007"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1002\/art.40692"},{"key":"e_1_2_12_17_1","unstructured":"NHS Digital. Health Survey for England 2018\u2014overweight and obesity in adults and children. 2019. Available from:https:\/\/files.digital.nhs.uk\/52\/FD7E18\/HSE18-Adult-Child-Obesity-rep.pdf. Accessed 9 Sept 2025."},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-46663-4"},{"key":"e_1_2_12_19_1","doi-asserted-by":"publisher","DOI":"10.3390\/genes11080854"},{"key":"e_1_2_12_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/nrrheum.2015.135"},{"key":"e_1_2_12_21_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-023-02440-2"},{"key":"e_1_2_12_22_1","doi-asserted-by":"publisher","DOI":"10.2196\/60521"},{"key":"e_1_2_12_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.joca.2020.02.486"},{"key":"e_1_2_12_24_1","doi-asserted-by":"publisher","DOI":"10.3390\/s22020446"},{"key":"e_1_2_12_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.knee.2021.07.002"}],"container-title":["Knee Surgery, Sports Traumatology, Arthroscopy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/esskajournals.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/ksa.70332","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/esskajournals.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/ksa.70332","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/esskajournals.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/ksa.70332","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,5]],"date-time":"2026-04-05T11:29:09Z","timestamp":1775388549000},"score":1,"resource":{"primary":{"URL":"https:\/\/esskajournals.onlinelibrary.wiley.com\/doi\/10.1002\/ksa.70332"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,17]]},"references-count":24,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["10.1002\/ksa.70332"],"URL":"https:\/\/doi.org\/10.1002\/ksa.70332","archive":["Portico"],"relation":{},"ISSN":["0942-2056","1433-7347"],"issn-type":[{"value":"0942-2056","type":"print"},{"value":"1433-7347","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,17]]},"assertion":[{"value":"2025-09-09","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-21","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-17","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}