Open-Source FHIR –> OMOP Pipeline

Name: Open-Source FHIR –> OMOP Pipeline
Availability: InStock
Author: brady_bastian

by brady_bastian·May 2, 2026·2 points·0 comments

Visit Project View on HN

AI Analysis

●●SolidNiche GemBig Brain

Automated FHIR-to-OMOP translation beats manual dbt modeling, but lacks Fivetran's broader connector library.

Strengths

•Schema inference generates dbt models on the fly without manual mapping configuration.
•Synthetic hospital demo with 88k patients proves the pipeline handles complex nested structures.
•Built-in PII detection during transformation eliminates separate scanning steps.

Weaknesses

•FHIR-to-OMOP is a narrow niche compared to general ELT tools like Airbyte.
•Relies on dbt ecosystem; teams not using dbt must adopt it to use this tool.

Post Description

Parsing FHIR data and converting that to OMOP standard is a notoriously difficult problem. I think we've cracked it.

I've open sourced two GitHub repos:

1. Forge-Core. This is the secret sauce. Automatically parse ANY json structure inside your own data warehouse by mapping your schema and building dbt models on the fly. Easy to set up and use. https://github.com/foxtrotcommunications/foxtrotcommunicatio...

2. Avalon-Public. Following the forge parse, use these dbt models to correctly and efficiently build your needed omop models. https://github.com/foxtrotcommunications/foxtrotcommunicatio...