Implementing Provider Directory APIs
summary
Reference implementation converts public CMS/NPPES CSVs into a validated FHIR store: files land in Google Cloud Storage, run through Dataflow (Apache Beam) jobs to NDJSON, then transform via VESSEL mappings into FHIR transaction bundles and write 74 validated resource types into Firestore with Firestore validation against DaVinci PDEX PlanNet / US Drug Formulary profiles. It includes reconciliation/deduplication rules, partner NPI enrichment, and app registration + OAuth via Health API X portal.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:15Overview
- 3:10Reference Implementation
- 8:30Conversion
- 14:15Questions
- 14:47Format of CSV files
- 16:10Source data
- 17:05Different patients
- 17:55Resolving duplicate providers
related talks
36mHealthcare Directories
20mPrototyping data ingestion pipelines
46mThe FHIR Bulk Data API and what's new in v2!
37mA novel way to manage and represent unstructured clinical data
46mC-CDA on FHIR
18mRelevant technologies for CMS-0057-P and CMS-0053-P
16mThe story of FHIR data exchange and storage combined
43mLet's Build! Patient Mediated Data Connectivity
