How to integrate FHIR with your analytics pipeline
summary
Shows a reproducible pipeline that exports FHIR Bulk Export JSON, encodes it with Apache Spark and Pathling into Delta Lake tables, and merges incremental changes by resource ID so the replica stays up to date. Demonstrates notebooks on Databricks that use FHIRPath expressions and Pathling UDFs to flatten extensions, questionnaire items and coded membership into use-case rows, then writes views to an SQL schema and connects live to Power BI; includes orchestration via Databricks Workflows.
AI-written summary · about
chapters (9)
- 0:00Session introduction
- 2:05Defining the challenge
- 4:00Pipeline architecture
- 7:24Export and loading
- 11:50Transformation methods
- 17:48Pipeline orchestration
- 20:02Visualization and analytics
- 22:53Handling complex structures
- 29:31Q&A session
related talks
35mScalable Data Science with FHIR
19mAnalyzing FHIR Data Using ML and Analytical Tools
31mMapping FHIR data to parquet for analytics and machine learning
33mBridging large-scale FHIR to AI in production
21mClinical and Claims Data Transformation utilizing FHIR Bulk Data
52mElevate your data and AI governance with Databricks Data Intelligence Platform
27mStorage and analytics
20mExchange of analytics data using FHIR resources