JMIR Publications

ABSTRACT

Background:

Healthcare research relies increasingly on secure, multi-partner collaboration across genomic, clinical, behavioural, and social datasets. Yet regulatory, ethical, and operational constraints often prevent institutions from pooling raw data. Existing infrastructures—Trusted Research Environments (TREs), federated learning pilots, and integration platforms—remain fragmented and frequently lack runtime safeguards, interoperable standards, or equitable governance models [1]. Meanwhile, regulatory initiatives are becoming more supportive of distributed and privacy-preserving solutions, underscoring the need for scalable federated approaches.

Objective:

This paper reviews current solutions for collaborative health research and introduces Open Health, a federated computing framework and proof-of-concept (PoC) demonstrator designed to enable secure, interoperable, and person-centred collaboration while maintaining privacy and data sovereignty.

Methods:

We undertook: (1) a comparative review of collaborative platforms, assessing maturity, capabilities, and governance; (2) development of the Open Health framework, adapted from Federated Computing [2]; and (3) implementation of a PoC demonstrator and roadmap to operationalise future use cases.

Results:

Open Health builds on four pillars—distributed datasets, federated analytics-to-data services, standards-based APIs, and governance-as-code—coordinated through a federated orchestrator. The PoC validated feasibility by executing analytics and federated learning at source, exchanging anonymised outputs via APIs, and enforcing governance at runtime. A co-creation roadmap links requirements, design, testing, and evaluation, embedding fairness, transparency, and usability.

Conclusions:

Open Health demonstrates that Federated Computing can be technically feasible, policy aligned, and person centred. It provides a blueprint for learning health systems that preserve privacy, strengthen trust, and support multi-partner research across organisational and jurisdictional boundaries.