FHIR Projects
summary
Builds production SMART on FHIR apps using Node.js libraries and a small open-source client to test server capabilities before runtime, then normalizes multiple FHIR versions via an adapter + serializer pattern so a single domain model drives UI logic. Deployments at McKesson and Intermountain include a Boston Children’s growth‑chart integrated into Cerner (converted FHIR v2↔v1) and running in ICUs. Solves practical surprises: LOINC length vs height, missing gestational‑age‑at‑birth (added a field), printing quirks, and caches the capabilities statement rather than refetching.
AI-written summary · about
chapters (21)
- 0:00Introduction and background
- 1:18Why FHIR matters for developers
- 2:22Building SMART apps with FHIR
- 5:10Responsive design patterns for FHIR
- 8:00Live coding FHIR capabilities
- 9:23Domain models vs FHIR models
- 11:16Adapter layer for FHIR versions
- 12:21Working with FHIR profiles
- 13:47Cross-platform FHIR tools
- 14:31Localization in FHIR apps
- 16:18Caching and testing FHIR capabilities
- 17:04FHIR polyfills and alternatives
- 18:32Open source FHIR libraries
- 19:14Intermountain Healthcare overview
- 20:39Integrating growth chart app
- 23:50Terminology and coding challenges
- 26:21Gestational age at birth issues
- 28:58FHIR version compatibility
- 30:01Dynamic knowledge in apps
- 31:46Printing and UI synchronization
- 32:49Project outcomes and lessons
