Home
Common data models have solved many challenges of utilizing electronic health records, but have not yet meaningfully integrated clinical and molecular data. Aligning clinical data to open biological ontologies (OBOs), which provide semantically computable representations of biological knowledge, requires extensive manual curation and expertise. To address these limitations, we introduce OMOP2OBO, a health system-scale, disease-agnostic methodology to create interoperability between standardized clinical terminologies and semantically encoded OBOs and present results demonstrating the utility within two health systems.
- We presented preliminary results from OMOP2OBO at the 2020 OHDSI Symposium
- We presented the results of our validation for mapping LOINC lab result to the Human Phenotype Ontology group.
- Vasilevsky N, Zhang A, Yates A et al. LOINC2HPO: Curation of Phenotype Data from the Electronic Health Records using the Human Phenotype Ontology [version 1; not peer reviewed]. F1000Research 2019, 8:383 (slides) (doi: 10.7490/f1000research.1116524.1)
- Vasilevsky N, Zhang A, Gourdine J et al. LOINC2HPO: Curation of Phenotype Data from the Electronic Health Records using the Human Phenotype Ontology [version 1; not peer reviewed]. F1000Research 2019, 8:382 (poster) (doi: 10.7490/f1000research.1116517.1)