EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Enhancing Trust by a Keycloak-Flower Integration for Federated Machine Learning

Matthaeus Morhart, Johanna Schwinn, Seyedmostafa Sheikhalishahi, Michael Wellnhofer, Ludwig Christian Hinske, Mathias Kaspar

Studies in Health Technology and Informatics · 2025

DOI: 10.3233/shti250406 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2025. Verfasst von Matthaeus Morhart und 5 weiteren Autor:innen. Beteiligte Einrichtung: Digital Medicine, University Hospital of Augsburg, Augsburg, Germany. Hauptthemen laut MEDLINE: Machine Learning, Computer Security, Software, Trust. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Since its introduction, federated learning (FL) has attracted a lot of attention in the medical field, but its actual application in healthcare organisations remains limited. Flower is a leading FL framework known for its good documentation and wide application. To close security gaps, we propose to integrate Keycloak with gRPC and Flower to improve identity and access management. We have developed a lightweight Python module that integrates both and also validates the client’s code with the server before execution. The system has been tested in a simple prototype, but further work and security testing is required for a complex evaluation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningComputer SecuritySoftwareTrustSystems IntegrationFederated Learning

Stichwörter der Autor:innen

FlowerKeycloakTrust

Autor:innen und Institutionen

  • Digital Medicine, University Hospital of Augsburg, Augsburg, Germany.

Zitieren

APA

Morhart, M., Schwinn, J., Sheikhalishahi, S., Wellnhofer, M., Hinske, L. C., & Kaspar, M. (2025). Enhancing Trust by a Keycloak-Flower Integration for Federated Machine Learning. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti250406

Vancouver

Morhart M, Schwinn J, Sheikhalishahi S, Wellnhofer M, Hinske LC, Kaspar M. Enhancing Trust by a Keycloak-Flower Integration for Federated Machine Learning. Studies in Health Technology and Informatics. 2025. doi:10.3233/shti250406

BibTeX
@article{Morhart2025,
  author = {Morhart, Matthaeus and Schwinn, Johanna and Sheikhalishahi, Seyedmostafa and Wellnhofer, Michael and Hinske, Ludwig Christian and Kaspar, Mathias},
  title = {Enhancing Trust by a Keycloak-Flower Integration for Federated Machine Learning},
  journal = {Studies in Health Technology and Informatics},
  year = {2025},
  doi = {10.3233/shti250406}
}
Kostenlos starten →