Model utility and explainability in federated learning - A case study in healthcare using fundus oculi datasets
This study demonstrates the benefits of FL systems in a healthcare scenario, providing a case study for evaluating federated systems beyond idealized benchmarks. Our findings highlight the potential of FL to not only improve model utility in privacy-sensitive medical domains but also to learn medically relevant features instead of spurious correlations.
Authors
- Joachim Denzler (ORCID: https://orcid.org/0000-0002-3193-3300)
- Marco Seeland (ORCID: https://orcid.org/0000-0001-7204-3972)
- J. Haueisen
- Niklas Penzel (ORCID: https://orcid.org/0000-0001-8002-4130)
- Hannes Oppermann (ORCID: https://orcid.org/0009-0005-5567-1535)
- Daniel Scheliga (ORCID: https://orcid.org/0000-0002-6469-7068)
- Patrick Mäder (ORCID: https://orcid.org/0000-0001-6871-2707)
Institutions
- Technische Universität Ilmenau (DE)
- Ernst Abbe University of Applied Sciences Jena (DE)
- Friedrich Schiller University Jena (DE)
Publication Details
- Journal
- Journal of Biomedical Informatics
- Published
- 2026-03-07
- DOI
- https://doi.org/10.1016/j.jbi.2026.105010
- Primary Topic
- Retinal Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00