Journal article icon

Journal article

Self-supervised learning for human activity recognition using 700,000 person-days of wearable data

Abstract:
Accurate physical activity monitoring is essential to understand the impact of physical activity on one's physical health and overall well-being. However, advances in human activity recognition algorithms have been constrained by the limited availability of large labelled datasets. This study aims to leverage recent advances in self-supervised learning to exploit the large-scale UK Biobank accelerometer dataset-a 700,000 person-days unlabelled dataset-in order to build models with vastly improved generalisability and accuracy. Our resulting models consistently outperform strong baselines across eight benchmark datasets, with an F1 relative improvement of 2.5-130.9% (median 24.4%). More importantly, in contrast to previous reports, our results generalise across external datasets, cohorts, living environments, and sensor devices. Our open-sourced pre-trained models will be valuable in domains with limited labelled data or where good sampling coverage (across devices, populations, and activities) is hard to achieve.
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Files:
Publisher copy:
10.1038/s41746-024-01062-3

Authors


More by this author
Institution:
University of Oxford
Division:
MSD
Department:
Nuffield Department of Population Health
Role:
Author
ORCID:
0000-0001-5944-1925
More by this author
Institution:
University of Oxford
Division:
MSD
Department:
Nuffield Department of Population Health
Role:
Author
More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Role:
Author
More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Author
More by this author
Institution:
University of Oxford
Division:
MSD
Department:
NDM
Sub department:
Big Data Institute
Role:
Author
ORCID:
0000-0002-3241-1280


Publisher:
Springer Nature
Journal:
npj Digital Medicine More from this journal
Volume:
7
Issue:
1
Article number:
91
Place of publication:
England
Publication date:
2024-04-12
Acceptance date:
2024-02-22
DOI:
EISSN:
2398-6352
ISSN:
2398-6352
Pmid:
38609437


Language:
English
Keywords:
Pubs id:
1989639
Local pid:
pubs:1989639
Deposit date:
2024-05-02

Terms of use



Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP