FHIR Ontology to Collibra Integration
Overview
FHIR (Fast Healthcare Interoperability Resources) is the HL7® standard for exchanging healthcare data. It defines resources such as Patient, Practitioner, CareTeam and Device, each with a hierarchy of elements and their definitions, and it is used by Epic, Google Cloud Healthcare API, AWS HealthLake and many other systems.
This integration imports the FHIR ontology into Collibra, so that healthcare and life sciences (HCLS) organisations get a single, governed reference for the meaning of their health data.
How It Works
- Download the JSON definition of a FHIR resource from hl7.org (for example the Patient resource).
- The Python script reads every element of the resource and creates it in Collibra as a Data Attribute asset.
- The element’s definition and short description are added as attributes.
- Parent and child elements are linked with the “groups” relation, so the full FHIR hierarchy (for example Patient → contact → contact.telecom) can be browsed in a Collibra diagram.
Use Cases
- A central catalog of FHIR definitions for your health data.
- A foundation for mapping FHIR elements to HL7 v2 codes (for example PID-3, NK1-2) used in Epic and other systems, or to other standards such as LOINC, CDA and RIM.
- Linking FHIR resources to the API endpoints that expose them in Epic, Google Cloud Healthcare API and AWS HealthLake.
- Connecting the ontology to FHIR datasets already catalogued in Collibra, for example in Google BigQuery.
Prerequisites
- Python 3 with the requests library
- A Collibra user with permission to create assets, attributes and relations
- A Collibra domain for the FHIR ontology (its ID goes into the script)
- The JSON definition of the FHIR resource you want to import
Media
More details
Release Notes
Initial release
Compatibility
- HL7 FHIR R4
- Collibra Data Intelligence Cloud
Dependency
- Python 3
- requests library
- Collibra REST API v2


