Leveraging Provenance to Maintain Traceability while Achieving Interoperability
summary
Diameter Health converts messy inbound CCD/CCDA data into normalized FHIR resources, then preserves a byte-for-byte copy of the original as a FHIR resource and links them with a Provenance resource so systems can fetch the original via _revinclude and a second read by id. Extensions were rejected because they add complication and bloat and force storage alongside normalized records. The approach is deterministic, supports auditing, and lets originals be stored cheaply or offloaded (for example originals offloaded to S3) while keeping a Provenance + revinclude workflow.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:07Structure vs. Semantic Interoperability
- 2:21The History of HL7 Standards - Simplified
- 4:04Is semantics a problem?
- 10:03Do Extensions solve the problem?
- 10:56What is Provenance?
- 13:06Does Provenance pointing to originals resources solve the problem?
related talks
25mLesion tracking DICOM SR to FHIR mapping
41mUsing Graph-theory to empower patients
37mClinical Quality Improvement with FHIR
42mMeasuring Data FAIRness: Lessons From the NHS, ONS and IPCC
17mDecentralization and Human Rights
46mEsther Huyer & Raymonde Weyzen
23m3-Step Path for Fixing Data Discoverability
43mFollow the Yellow-Brick Code (A Terminology journey)