Deep learning of wrist accelerometry from UK Biobank data identifies early movement signatures of knee osteoarthritis up to 5 years before diagnosis
Knee Surgery, Sports Traumatology, Arthroscopy · 2026-04 · 34 (4) · 1524-1533
DOI: 10.1002/ksa.70332 ↗Kurz gefasst
Fachartikel in der Zeitschrift Knee Surgery, Sports Traumatology, Arthroscopy, online erschienen am 17.02.2026. Verfasst von Ricardo Smits Serena und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Orthopaedics and Sports Orthopaedics, TUM University Hospital, Munich, Germany. Hauptthemen laut MEDLINE: Osteoarthritis, Knee, Accelerometry, Deep Learning, Wrist. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: TUM Universitätsstiftung (TUM University Foundation).
Abstract (Originalsprache)
Purpose
This study aims to test whether week‐long 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.
Methods
We conducted a retrospective case–control study using the UK Biobank data set. After quality control, 102,120 participants with valid accelerometry were available; KOA was identified via ICD‐10 M17.x codes ( n = 7262). To reduce adiposity confounding, analyses were restricted to body mass index (BMI) ≥ 29, with controls matched to KOA on age, sex and BMI distributions. We used preprocessed, orientation‐independent, hourly mean acceleration over a 24‐h 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‐validation assessed accuracy, macro F1, macro sensitivity and AUC.
Results
Balanced cohorts were formed for three tasks: healthy versus KOA ( n = 3677 per class), prodromal versus diagnosed KOA ( n = 1596 vs. 2081), and healthy vs prodromal within 5 years ( n = 1369 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 ± 1.2%; AUC 0.672 ± 0.017) and healthy vs prodromal within 5 years (64.5 ± 0.5%; AUC 0.675 ± 0.019). Discrimination between prodromal and diagnosed KOA was close to random (54.6% ± 1.5%; AUC 0.552 ± 0.015).
Conclusions
One week of wrist‐worn 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‐cost component of population‐level KOA screening. Level of Evidence Level II, prognostic study—lower‐quality prospective cohort. The study uses a large, population‐based prospective cohort (UK Biobank) with retrospective analytical methods; follow‐up is high, but the study is a secondary analysis rather than a primary prospectively designed prognostic trial.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Orthopaedics and Sports Orthopaedics, TUM University Hospital, Munich, Germany.
- Institute for AI and Informatics in Medicine, TUM University Hospital, Munich, Germany.
- Kantonsspital Baselland, Bruderholz, Switzerland.
- Kantonsspital Baselland, Liestal, Switzerland.
- Kantonsspital Baselland, Laufen, Switzerland.
- Waldkliniken Eisenberg GmbH, Eisenberg, Germany.
ORCID-Profile: Ricardo Smits Serena
Förderung
- TUM Universitätsstiftung (TUM University Foundation)
Zitieren
APA
Smits Serena, R., Hirschmann, M. T., Matziolis, G., von Eisenhart-Rothe, R., Rueckert, D., Hinterwimmer, F., & Valle, C. (2026). Deep learning of wrist accelerometry from UK Biobank data identifies early movement signatures of knee osteoarthritis up to 5 years before diagnosis. Knee Surgery, Sports Traumatology, Arthroscopy, 34(4), 1524-1533. https://doi.org/10.1002/ksa.70332
Vancouver
Smits Serena R, Hirschmann MT, Matziolis G, von Eisenhart-Rothe R, Rueckert D, Hinterwimmer F, et al. Deep learning of wrist accelerometry from UK Biobank data identifies early movement signatures of knee osteoarthritis up to 5 years before diagnosis. Knee Surgery, Sports Traumatology, Arthroscopy. 2026;34(4):1524-1533. doi:10.1002/ksa.70332
BibTeX
@article{SmitsSerena2026,
author = {Smits Serena, Ricardo and Hirschmann, Michael T and Matziolis, Georg and von Eisenhart-Rothe, Rüdiger and Rueckert, Daniel and Hinterwimmer, Florian and Valle, Christina},
title = {Deep learning of wrist accelerometry from UK Biobank data identifies early movement signatures of knee osteoarthritis up to 5 years before diagnosis},
journal = {Knee Surgery, Sports Traumatology, Arthroscopy},
year = {2026},
volume = {34},
number = {4},
pages = {1524--1533},
doi = {10.1002/ksa.70332}
}
EUVIMED