Continuous assessment of daily-living gait using self-supervised learning of wrist-worn accelerometer data
Physical activity and mobility are critical for healthy aging and predict diverse health outcomes. While wrist-worn accelerometers are widely used to monitor physical activity, estimating gait metrics from wrist data remains challenging. We extend ElderNet, a self-supervised deep-learning model previously validated for walking-bout detection, to estimate gait metrics from wrist accelerometry. Validation involved 819 older adults (Rush-Memory-and-Aging-Project) and 85 individuals with gait impairments (Mobilise-D), from six medical centers. In Mobilise-D, ElderNet achieved an absolute error of 8.82 cm/s and an intra-class correlation of 0.87 for gait speed, outperforming state-of-the-art methods (p < 0.001) and models using a lower-back sensor. ElderNet outperformed (percentage error; p < 0.01) competing approaches in estimating cadence and stride length, and better (p < 0.01) classified mobility disability (AUC = 0.80) than conventional gait or physical activity metrics. These results demonstrate the potential of ElderNet a scalable tool for gait assessment using wrist-worn devices in aging and clinical populations.
Authors
- Arne Mueller (ORCID: https://orcid.org/0000-0001-6551-2283)
- Luca Palmerini (ORCID: https://orcid.org/0000-0003-4758-662X)
- Aron S. Buchman (ORCID: https://orcid.org/0000-0002-6426-2742)
- Or Perlman (ORCID: https://orcid.org/0000-0002-3566-569X)
- Yonatan E. Brand (ORCID: https://orcid.org/0000-0002-4214-0699)
- Silvia Del Din (ORCID: https://orcid.org/0000-0003-1154-4751)
- Alison Jane Yarnall (ORCID: https://orcid.org/0000-0002-3126-9163)
- Walter Maetzler
- Felix Kluge
- Lynn Rochester
- Beatrix Vereijken
- Jeffrey M. Hausdorff
- Andrea Cereatti
- Clemens Becker
Institutions
- Novartis (Switzerland) (CH)
- Rush University Medical Center (US)
- Tel Aviv University (IL)
- Politecnico di Torino (IT)
- Norwegian University of Science and Technology (NO)
- Heidelberg University (DE)
- Tel Aviv Sourasky Medical Center (IL)
- Marconi University (IT)
- Newcastle upon Tyne Hospitals NHS Foundation Trust (GB)
- Robert Bosch Stiftung (DE)
- University Hospital Heidelberg (DE)
- Laboratori Guglielmo Marconi (Italy) (IT)
- NIHR Newcastle Biomedical Research Centre (GB)
- Newcastle University (GB)
- University of Bologna (IT)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2026-03-12
- DOI
- https://doi.org/10.1038/s41746-026-02528-2
- Primary Topic
- Balance, Gait, and Falls Prevention
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- European Federation of Pharmaceutical Industries and Associations
- National Institute for Health and Care Research
- European Commission
- Tel Aviv University
- National Institutes of Health
- NIHR Newcastle Biomedical Research Centre