Aug 2026· Scientific Reports· Vol 16· 1 citation· 28 references
TL;DR
Proposed Fed-ZTMA, a Federated Zero-Trust Medical Access Framework that combines the Byzantine-resilient weighted federated trust aggregation with adaptive zero-trust policy enforcement via Blockchain deployed smart contracts, demonstrates practical feasibility for multi-institutional 5G healthcare deployments where both privacy-preservation and Byzantine fault tolerance are mandatory requirements.
Abstract
The advent of the Internet of Things (IoT) technologies in healthcare has significantly changed the ways in which patient health information is collected, communicated, and managed within a wide array of geographically distributed healthcare ecosystem. The technological convergence not only increases clinical responsiveness but also creates significant issues related to trusted, interoperable cross-domain sharing of electronic health records (EHRs). Existing federated learning systems do not have tools to handle bad actors, implement zero-trust policies for access across organizational boundaries, or offer formally provable security guarantees. To address these challenges, this research work introduces Fed-ZTMA, a Federated Zero-Trust Medical Access Framework that combines the Byzantine-resilient weighted federated trust aggregation with adaptive zero-trust policy enforcement via Blockchain deployed smart contracts. The proposed framework splits the Synthea Synthetic EHR dataset across four independent logical healthcare organizations and assesses performance using a multi-node Ganache Ethereum testnet with seven different adversarial attack scenarios. Compared to existing approach such as FedHealthDP, BCMedShare, ZeroTrust-IoT, BlockEHR, and ABAC-Chain, Proposed Fed-ZTMA performs better, including a cross-organization access success rate of 98.7%, attack suppression rate of 97.9%, inter-domain access latency of 18.4 ms at 1000 devices, and security stability index of 96.4%. This research work demonstrates practical feasibility for multi-institutional 5G healthcare deployments where both privacy-preservation and Byzantine fault tolerance are mandatory requirements.
In a smart healthcare environment, these devices produce a tremendous amount of health-related data that is extremely sensitive for patients, including data generated by wearable, electronic health records (EHRs), medical imaging systems and healthcare IoT sensors. Although federated learning allows for model training...
M. Sathiyamoorthy· International Conference on...· 0 citations
Health data generated by Internet of Things (IoT) devices outside institutional boundaries—consumer wearables and home-based clinical monitors—expands the attack surface for unauthorized access and complicates patient-side control over data sharing, a problem that conventional, centrally gated Electronic Health Record...
J. C. Ferreira, Marco Esperança, Diogo Freitas et al.· Italian National Conference...· 1 citation
Federated Learning (FL) reduces direct exposure of sensitive clinical data, but decentralized healthcare deployments remain vulnerable to Byzantine and data-poisoning attacks and are complicated by non-independent and identically distributed (non-IID) data across institutions. Existing defenses typically address these...
Vivek Kumar Polurouthu, Tarun Kumar Jha, Rajya Lakshmi Kollipara et al.· Journal of Computers, Mechan...· 0 citations
Findings indicate that while ZTA offers transformative security benefits for healthcare cloud environments, adoption barriers including implementation complexity, workforce skill gaps, legacy EHR integration, and regulatory ambiguity require coordinated policy and industry responses.
Michael Akintomiwa Oyedeji, T. Dawotola, O. Oladapo· Iconic research and engineer...· 0 citations
The swift proliferation of Medical Internet of Things (MIoT) devices has made a remarkable difference to the remote monitoring of patients, intelligent diagnosis and delivery of healthcare services. But while more and more sensitive medical data is exchanged via personal devices, a number of challenges regarding privac...
A. P. Janani, S. Vairachilai· International Conference on...· 0 citations
Healthcare analytics increasingly depends on geographically distributed clinical records, connected medical devices, imaging repositories, and institutional data platforms. Conventional centralized learning can increase exposure of sensitive information and amplify governance challenges when data must cross organizatio...
A. Holzinger· Computer Science Bulletin· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.