The vast and rich data found in electronic health record systems can offer researchers and scientists important clues that may lead to a new generation of scientific insights. However, the ability to use these “bulk data” at scale often requires expensive and time-consuming manual processes, proprietary software, and data transformations into complex models. Enabling researchers to work with bulk sets of clinical data without cumbersome efforts to reorganize the information could serve as a catalyst to enable precision medicine and new scientific discoveries.
Children’s Hospital Corporation, in collaboration with Yale University and Yale-New Haven Health, aims to develop a platform based on Health Level Seven International® (HL7®) Fast Healthcare Interoperability Resources® (FHIR®) that leverages bulk data to support an ecosystem for research and learning, called Cumulus.
Cumulus will build from existing standards and open health information technology (IT) tools to offer turnkey functions that support rapid learning within a healthcare system. Tools to be developed and tested will allow users to annotate FHIR data for analytics, de-identify data, and query cohorts.
This project addresses the Leading Edge Acceleration Projects (LEAP) in Health IT fiscal year 2020 special area of interest 2: Cutting Edge Health IT Tools for Scaling Health Research.
This project began in 2020 and will end in 2022.
The goals of this project are to:
- Leverage the Substitutable Medical Applications, Reusable Technologies (SMART)/HL7 Bulk Data API—called Flat FHIR—to provide data at scale through a cloud-based platform called Cumulus that will support an apps ecosystem for research and learning.
- Select and adapt an appropriate open-source de-identification toolkit to implement different privacy settings for FHIR data sets.
- Standardize a set of processes and tools as modules on Cumulus supporting bulk data de-identification, basic cohort-building, and FHIR data analytics.