Skip to main content

FHIR Export & Import

VADRR supports HL7 FHIR R4 as its primary interoperability format for exchanging systematic review data.

What is FHIR?​

FHIR (Fast Healthcare Interoperability Resources) is a standard for exchanging healthcare information electronically. VADRR uses FHIR R4 Evidence resources to represent systematic review data in a machine-readable, standards-compliant format.

FHIR Export​

Exporting a Project​

  1. Go to the project Settings.
  2. Click Export (FHIR).
  3. VADRR generates a FHIR R4 Bundle containing:
    • Citations (as FHIR Citation resources)
    • Extractions (as FHIR ArtifactAssessment resources)
    • Key questions, populations, outcomes (as FHIR EvidenceVariable resources)
    • SR360 metadata (as FHIR Evidence and EvidenceReport resources)
  4. Download the bundle JSON.

API-based Export​

GET /api/v3/projects/:id          # Full project FHIR bundle
GET /api/v3/extractions/:id       # Single extraction as FHIR
GET /api/v3/sd_meta_data/:id      # SR360 as FHIR EvidenceReport

Include your API key as: Authorization: Bearer YOUR_API_KEY

FHIR Import (LLM-Assisted)​

VADRR can import citations and extraction data from FHIR bundles, including those generated by LLMs (large language models) using VADRR's schema.

LLM Import Workflow​

  1. Go to Imports → New Import → LLM / FHIR.
  2. Upload a FHIR JSON bundle (generated by an LLM from paper PDFs or structured data).
  3. VADRR validates the bundle against its FHIR schema.
  4. Valid resources are imported as citations and/or extractions.

Generating an LLM Package​

VADRR can generate a structured prompt package for LLMs:

  1. Go to Imports → New Import → LLM Package.
  2. VADRR exports your extraction form as a structured schema an LLM can use to guide data extraction from papers.
  3. Use this schema with your LLM of choice (GPT-4, Claude, etc.) to generate FHIR-formatted extraction output.
  4. Import the LLM output back into VADRR via FHIR Import.