An update on AI and FHIR
summary
Update summarizes community work: an AI Evaluation Focus Team plus experiments that wire ChatGPT plugins to a local API which runs FHIR validator + terminology lookups loop three times to refine outputs, yielding fewer but still imperfect results; GPT‑4 performs much better than GPT‑3.5. Covers governance work: proposals for provenance, explainability, Nutrition‑label style model metadata, an Analytics on FHIR prototype IG, and the US ONC HTI1 proposed rule with FAVES evaluation criteria.
AI-written summary · about
chapters (7)
- 0:00Introduction and setup
- 1:57Gen AI demo with FHIR
- 9:01Transparency in AI models
- 13:55Standardizing AI metadata
- 17:02HTI-1 regulation overview
- 22:01Future of AI in healthcare
- 23:35Q&A session
related talks
25mYou put what, where? What FHIR should be doing about AI/ML
15mAI as human augmentation
43mAI Has Entered The Chat: Humans, AI, and the Art of Collaboration
44mAI Foundations: Mitigate Risks and Boost SOC Efficiency
36mAI for Government Efficiency with ChatGPT: From Hype to Reality
26mHumanity and AI
46mUsing AI in Software Design: How ChatGPT Can Help With Creating a Solution Arch.
31mAI For Good: Fighting Health Insurance with AI