Extending BEHRT to UK Biobank: assessing transformer model performance in clinical prediction
Our results show that transformer models can achieve high predictive performance across diverse clinical scenarios, but outcomes vary considerably depending on modelling choices, particularly in long-term prediction tasks.
Authors
- Goran Nenadić (ORCID: https://orcid.org/0000-0003-0795-5363)
- M. Jani
- Yusuf Yildiz
- David A. Jenkins
Institutions
- Manchester Academic Health Science Centre (GB)
- University of Manchester (GB)
- Arthritis UK (GB)
- NIHR Manchester Biomedical Research Centre (GB)
- Centre for Epidemiology Versus Arthritis (GB)
Publication Details
- Journal
- Frontiers in Digital Health
- Published
- 2026-02-10
- DOI
- https://doi.org/10.3389/fdgth.2026.1715506
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Institute for Health and Care Research
- Department of Health and Social Care