FHIRCat: Enabling the Semantics of FHIR and Terminologies for Clinical and Translational Research

点击翻译按钮获取中文摘要

Project Summary HL7 Fast Healthcare Interoperability Resources (FHIR) is an emerging next generation standard framework for the exchange of electronic health record (EHR) data. The FHIR specification defines a common vocabulary and mechanism for sharing EHR data independent of how it is actually stored. All of the major EHR vendors are developing standardized FHIR interfaces to their clinical data. One of the exciting aspects of FHIR is that it is among the first clinical data standards to incorporate the vision of the “Semantic Web” described by Tim Berners-Lee in his oft cited article from 2001. In the article, he asserts that the Semantic Web is the next "killer app" and outlines a formula for success: 1) Represent information and meaning in a single, universal format - Resource Description Framework (RDF); 2) Create a catalog of URIs that serve as shared identifiers for the RDF representation in a given context of use; and 3) Create tools for transforming existing data and knowledge resources into the universal language of RDF and URIs. This allows information from disparate sources and structures to be combined into a single, federated collection of “linked open data”. This linked data model allows clinical data to be combined with other linked data such as environmental, genetic, geographic, and any other relevant data and to be analyzed as an integrated whole. We have previously demonstrated that the combination of RDF, OWL semantics, and the Shape Expressions (ShEx) schema language could theoretically provide a robust platform for integrating disparate data models and paradigms under a single, resource oriented collection of interfaces. The objective of this project is to design, develop and evaluate a novel informatics platform known as FHIRCat that leverages Semantic Web technologies, FHIR models/profiles, and ontologies for effective standards-based data integration and distributed analytics, enabling high-quality reproducible clinical and translational research. Our specific aims are: 1) Develop FHIR ShEx/RDF specification and tools for FHIR data validation; 2) Standardize and deploy the RDF-based data transformation between external models and FHIR; and 3) Build and evaluate a FHIR ontology-based data access (OBDA) system interoperable with clinical data repositories across institutions in partnership with the CTSA National Center for Data to Health (CD2H). The outcome of the project makes it possible to realize Berners-Lee’s vision for healthcare in the very near future, in turn creating new opportunities for advancing the automation of FHIR- based clinical data integration and sharing, and accelerating standards-based data discovery and distributed analytics in clinical and translational research. 1