Cumulus ETL

Cumulus ETL is your entry point into the whole Cumulus pipeline.

ETL stands for “extract, transform, load.”

1. Cumulus ETL first extracts data from EHR servers (usually in the form of a bulk FHIR export).
2. Then it transforms that data by de-identifying it and converting clinical notes into lists of symptoms.
3. And finally it loads that data onto the cloud to be consumed by the next phase of the Cumulus pipeline.

Installing

Read the Local Test Setup documentation to learn how to install & run Cumulus ETL on your machine with sample data, as an introduction to the flow.

Then you can move on to the Production Setup instructions to set up the AWS infrastructure that the full Cumulus pipeline will require.

Source Code

Cumulus ETL is open source. If you’d like to browse its code or contribute changes yourself, the code is on GitHub.

---

Table of contents

• Local Test Setup
• Production Setup
• Studies
• NLP
• Reviewing Output
• Bulk Exports
• Performance Tuning
• De-identification Explained
• Chart Review
• Vendor Tips & Tricks