An open source project to generate more realistic medication data in Synthea™.
- 🏆 First place winners of the ONC Synthetic Health Data Challenge
- Getting started
- Example diagram
- About the challenge
Team CodeRx for the Medication Diversification Tool (MDT) submission to the Synthetic Health Data Challenge consists of Joseph LeGrand (team leader), Kent Bridgeman, Kristen Tokunaga, Robert Hodges, Dalton Fabian, and Yevgeny (Eugene) Bulochnik.
Medication Diversification Tool (MDT) links:
- Final presentation video - START HERE
- Final presentation paper
- CodeRx MDT GitHub repo
- Final presentation Miro board
Challenge winner announcements:
Synthea™ generates open source synthetic patient records from birth to death. The problem we are trying to solve is how to make the medication selection in these populations more diverse and realistic. For instance, 100% of patients who have hypothyroidism are prescribed the same exact strength of generic levothyroxine. We hope to show how we can use open source tools and data to increase the diversity of medication selection in this population and to better represent a realistic patient population in the US.
Current state versus with the medication diversification tool for a hypothyroidism use case.
The U.S. Department of Health and Human Services’ (HHS) Office of the National Coordinator for Health Information Technology (ONC) today announced $100,000 in total awards to six winners of the Synthetic Health Data Challenge (Challenge). Synthetic health data (i.e., data that is artificially created to mimic real-world data), is important to researchers, health IT developers, and informaticians, among others, who need data to test new ideas until access to secure and actual clinical data is available.
The Challenge was conducted under ONC’s Synthetic Health Data Generation to Accelerate Patient-Centered Outcomes Research (PCOR) project, which is supported by HHS’ Office of the Secretary Patient-Centered Outcomes Research Trust Fund. Challenge winners created and tested innovative and novel solutions that will further augment the capabilities of Synthea™, an open-source synthetic health data generator that models the medical histories of synthetic patients. The availability of reliable and robust synthetic data generation tools safeguard patient privacy because they support appropriate stewardship practices in which real patient data is only accessed and used when necessary.
“The availability of realistic, synthetic data is a vital part of supporting iterative testing models and early stage research and product development,” said Steve Posnack, deputy national coordinator for health information technology. “We received a lot of inspired submissions that took this work to the next level and hope that each winner can serve as a foundation to further enhance tools that create synthetic data.”