Secure Federated Learning for Privacy-Preserving Medical Record Sharing across Hospitals

Sivakumar Ponnusamy, R. Mohemmed Yousuf, G. Visalaxi, S. Sumithra, B. Sushma, Shree Yoghitha S.

2025

Abstract

In the rapidly-changing era of eHealth, secure and effective inter-institutional data sharing is still a major issue. In this work, we introduce a federated learning method for privacy-preserving medical record sharing between a collection of hospitals. With the ability to train models in a decentralized manner, all without moving aggregated sensitive patient data, the proposed method respects data sovereignty, complies with regulation and promotes collective intelligence. The design includes efficient encryption and lightweight communication algorithms, so in secure way computational overhead is optimally minimized. The utility is evaluated with a variety of hospital databases and shows excellent performance in terms of data privacy, the diagnostic accuracy, and the interoperability. This work not only addresses the restrictions of centralized data systems, but creates a foundation for scalable, secure, and responsible AI deployment in health systems.

Paper Citation

in Harvard Style

Ponnusamy S., Yousuf R., Visalaxi G., Sumithra S., Sushma B. and S. S. (2025). Secure Federated Learning for Privacy-Preserving Medical Record Sharing across Hospitals. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 194-201. DOI: 10.5220/0013860100004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sivakumar Ponnusamy and R. Yousuf and G. Visalaxi and S. Sumithra and B. Sushma and Shree S.},
title={Secure Federated Learning for Privacy-Preserving Medical Record Sharing across Hospitals},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25},
year={2025},
pages={194-201},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013860100004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25
TI - Secure Federated Learning for Privacy-Preserving Medical Record Sharing across Hospitals
SN - 978-989-758-777-1
AU - Ponnusamy S.
AU - Yousuf R.
AU - Visalaxi G.
AU - Sumithra S.
AU - Sushma B.
AU - S. S.
PY - 2025
SP - 194
EP - 201
DO - 10.5220/0013860100004919
PB - SciTePress