BulletTrain is a Health Information Exchange built on the OpenHIE (Open Health Information Exchange) framework. It implements the mediator pattern — where a central orchestration layer mediates between clinical systems, AI services, and operational platforms — delivering standards-compliant interoperability, full governance, and national-scale reliability.
Data enters from clinical systems. BulletTrain classifies, routes, and governs it. AI services receive only what they need, only what they are authorised for.
"Connect once. Orchestrate every clinical and AI workflow."
Every message audited, every route validated, every exception logged. Governance is not a feature you enable — it is the foundation the platform is built on.
AI models connect safely through governed channels, not ad-hoc application programming interfaces. Patient data stays protected while AI services get exactly the data they are authorised to use.
Fast Healthcare Interoperability Resources (FHIR), Health Level Seven (HL7), Clinical Document Architecture (CDA), X12 transaction sets, Digital Imaging and Communications in Medicine (DICOM) — not bolted on after the fact, native from the transport layer.
Each country runs its own instance. Patient data never crosses borders. Full regulatory compliance without compromising platform capabilities.
Fully managed within national data boundaries. No patient data leaves the jurisdiction.
Core services on-premises with cloud-based orchestration and analytics where policy allows.
Self-contained deployment within your own infrastructure. Complete operational control.
Governed connectivity for clinical and AI workflows. The transport layer that guarantees delivery across every system in your healthcare estate.
View BulletTrain →Agent discovery and trust. A universal registry that lets AI agents find, verify, and connect with each other safely across healthcare networks.
View Agent Registry →Secure agent communication. Direct, governed messaging between AI agents operating within and across healthcare organisations.
View Nexus →GHARRA: managed service • Nexus A2A: governed protocol • BulletTrain: free to self-host
| Capability | BulletTrain | Legacy Integration Engines | Generic AI Frameworks | Build Your Own |
|---|---|---|---|---|
| OpenHIE architecture | Complete implementation | Partial or proprietary | None | Years to build |
| AI agent orchestration | 25 clinical agents, multi-model | Not available | Generic, no healthcare | Must build from scratch |
| Global agent registry | GHARRA federation | Not available | Not available | Does not exist elsewhere |
| Agent-to-agent protocol | Nexus A2A, 13-point admission | Not available | No healthcare safety | Must design from scratch |
| WHO standards alignment | IHE PIX/PDQ, MHD, FHIR R4 | Varies | None | Months per profile |
| Patient data sovereignty | Sovereign deployment, 3-layer detection | Cloud-only options | Not addressed | Must architect from scratch |
| Emergency break-glass | Absolute safety constraint | Manual override | Not addressed | Rarely considered |
| Time to production | Weeks | Months to years | Months (no compliance) | 3+ years |
| Accessibility | BulletTrain free to self-host | Proprietary | Varies | N/A |
See the platform in action
Beyond BulletTrain, GHARRA, Nexus A2A and Bevan LLM, the platform now orchestrates a growing suite of clinical and operational systems — provider and citizen portals, prescribing, pharmacy, laboratory, radiology, ambulance dispatch, claims, ERP and more.
Explore the ecosystemDiscover how BulletTrain delivers governed connectivity and guaranteed delivery for your clinical and AI workflows.