Clinical guidelines with FHIR and CQL
summary
Describes how the FHIR Clinical Guidelines IG encodes WHO Antenatal Care and CDC Opioid Prescribing recommendations as L3 computable artifacts using CQL, PlanDefinition and value sets so a clinical reasoning server can run them via CDS Hooks. Uses the CQL translator, language server, VS Code plugin, cqf tooling and HAPI stacks to build an opioid prescribing IG, extract an MME calculation and power an ARC pain app. Packaging and value set expansion prove painful and tend to go stale despite successful end-to-end evaluations.
AI-written summary · about
chapters (22)
- 0:00Intro
- 0:32Learning Objectives
- 4:12Types of Interoperability
- 7:14Components of Sharing Logic
- 9:21Requirements to Running Code
- 10:17Key Interoperability Resources
- 11:02Opioid-related Projects
- 12:16Recommendation #11 - L2
- 15:32Value Set Definition
- 16:36Challenges
- 19:52Example Data Element
- 20:24Data Element Contextualization
- 20:59For example...
- 21:36Complex Data Element
- 22:01L3- Profiles (Data Elements)
- 23:45L3 Recommendation
- 26:50Service-based Integration via CDS Hooks (L4)
- 28:00Terminology Requirements
- 29:39ValueSet Expansion
- 30:13Artifact ValueSet Usage
- 38:24What did you learn?
- 41:17Clinical Quality Framework - Open Source Stack
related talks
30mFHIR in the W.H.O.L.E.: Graphing Clinical Data for Smarter AI Systems
07mOCI Language Services
29mAI, SQL, and GraphQL Walk into a Fertility Clinic… LLM-based Medical feature development
29mRAG for a medical company: the technical and product challenges
34mFrom Documents to Dynamic Graphs: Ontology-Driven Clinical Graph Construction
44mRAG for a Medical Company: Technical and Product Challenges AI & Data
27mAI, SQL, and GraphQL Walk into a Fertility Clinic
09mOracle Health Clinical AI Agent