Overview
The Scribe x EMR Integration track is a focused challenge for teams to build real-world, deployable integrations between AI Medical Scribes and Electronic Medical Records (EMR) systems.
This track aligns with the V2DD Consortium (Medical Scribe Alliance) objectives, emphasizing interoperability, open standards, and the creation of Digital Public Goods (DPGs).
Problem Statement
The Goal: Build a working, end-to-end integration that moves clinical data from a Scribe solution into an EMR using the predefined interoperability stack.
Participants must implement three distinct layers:
- The Handshake (The Protocol): Implement the pre-defined exchange protocol that manages the connection state between the Scribe and the EMR (similar to MCP).
- The Payload (FHIR Standard): Use or build an SDK (Scribe2FHIR) to structure the clinical data into the standardized FHIR format. This layer ensures the data is strictly typed and interoperable.
- The Bridge (The Integration Adapter): Build a reusable adapter that maps the standardized FHIR output (from step 2) into the specific fields of your target EMR.
- Requirement: This adapter must be open-sourced as a Digital Public Good to allow any Scribe using the standard to connect to this EMR.
System Architecture
Scribe Output >> [Protocol/Handshake] >> [FHIR SDK] >> Standardized JSON >> [Adapter] >> Target EMR
Qualification Checklist
To qualify for evaluation and prizes, teams must:
- Demonstrate End-to-End Flow: Show a live session where a Scribe generates output, passing it to the EMR to "seed" a doctor’s visit note for further editing.
- Open Source Repository: The integration code (specifically the Adapter) must be public.
- Adherence to Specs: The solution must strictly follow the provided Protocol and Payload specifications.
Submissions without a working demo or public code will not be eligible for prizes. Maximum of 4 members may present to the jury.