Skip to content

Synthmas demo

End-to-end walk-through: get a running ETL Orchestrator with the Synthmas pipeline loading the Synthea synthetic-patient dataset into an OMOP CDM.

You'll do this twice — once with a CSV CDM target (self-contained, no database required) and once with a PostgreSQL CDM (closer to production: FK enforcement, downstream OHDSI tool compatibility). Both runs produce row-for-row identical CDM tables on engine 1.4.1+ — the parity isn't approximate, it's exact.

What you'll build

              ┌─────────────────────────────┐
              │  Synthea CSVs (1,171         │
              │  patients, ~2.1M event rows) │
              └──────────────┬──────────────┘
                             │
                             ▼
              ┌─────────────────────────────┐
              │  Synthmas pipeline           │
              │  14 etl_map steps            │
              │  + pre-/post-automation      │
              └──────┬───────────────┬──────┘
                     │               │
                     ▼               ▼
              ┌────────────┐   ┌────────────┐
              │ CSV CDM    │   │ Postgres   │
              │ (per-table │   │ schema     │
              │  files)    │   │ synthmas   │
              └────────────┘   └────────────┘

The demo covers all of:

  • License upload + Settings tour.
  • Athena vocab upload + what tables it must contain.
  • Source CSVs (Synthea generation or download).
  • CSV environment + run.
  • PostgreSQL environment + run (same output, RDBMS-shaped for OHDSI tools + FK enforcement).
  • Reading the DAG, the ribbon bar, the per-step modal.

Prerequisites

  • A Linux host with docker or podman installed.
  • Internet egress (initial image pull, Athena download, optional Synthea generation).
  • For the PostgreSQL run: a reachable PostgreSQL ≥ 14 instance you can create a schema in (the demo creates synthmas).

Steps

  1. Setup — request a temp license, install the orchestrator, log in, import the Synthmas pipeline.
  2. Vocabulary — what Athena bundle you need + how to upload it.
  3. Source data — get the Synthea CSVs.
  4. Run on CSV — create the CSV env, kick off the pipeline, watch it complete.
  5. Run on PostgreSQL — create the schema + the RDBMS env, run again.
  6. Reading the run-mode UI — DAG colors, ribbon bar, per-step modal, what the numbers mean.