Development of time to event prediction models using federated learning
The proposed procedures enable training of time-to-event models using data distributed across sites, without direct sharing of individual-level data and event times, while retaining a predictive performance on par with undistributed approaches.
Authors
- Rasmus Rask Kragh Jørgensen (ORCID: https://orcid.org/0009-0000-7249-2196)
- Rasmus Froberg Brøndum (ORCID: https://orcid.org/0000-0001-5537-6767)
- Lasse Hjort Jakobsen (ORCID: https://orcid.org/0000-0001-6575-7686)
- Tarec Christoffer El‐Galaly (ORCID: https://orcid.org/0000-0002-4406-380X)
- Mikkel Runason Simonsen (ORCID: https://orcid.org/0009-0001-1119-4782)
- Martin Bøgsted (ORCID: https://orcid.org/0000-0001-9192-1814)
- Jonas Faartoft Jensen
Institutions
- Aalborg University Hospital (DK)
- Odense University Hospital (DK)
- Karolinska Institutet (SE)
- Aalborg University (DK)
Publication Details
- Journal
- BMC Medical Research Methodology
- Published
- 2025-05-26
- DOI
- https://doi.org/10.1186/s12874-025-02598-y
- Citations
- 2
- Primary Topic
- Privacy-Preserving Technologies in Data
- Type
- article
- Field-Weighted Citation Impact
- 4.96
Funders
- Villum Fonden