defining-cohort-phenotypes

To build a computable OMOP cohort definition, you author concept sets, entry events, and inclusion rules using CIRCE-compatible JSON, then augment eligibility criteria with NLP-derived features extracted locally from clinical notes.

A computable phenotype in the OMOP Common Data Model is a reproducible patient cohort definition grounded in standard vocabularies, combining structured codes with clinical NLP signals for validated research eligibility.

You create CIRCE cohort expression JSON for OHDSI ATLAS by defining concept sets, entry events, and inclusion rules that are grounded in standard vocabularies and validated against OMOP CDM data.

Yes, you can reuse PheKB and OHDSI Phenotype Library logic by integrating existing phenotype definitions into new CIRCE-compatible cohort expressions while maintaining standard vocabulary grounding and provenance tracking.

You add NLP-derived smoking status to an OMOP diabetes cohort by extracting social and behavioral features from clinical notes and incorporating them as custom inclusion rules within the CIRCE cohort expression.

CIRCE NLP-augmented phenotyping requires local processing of clinical text and standard vocabulary grounding, meaning definitions depend heavily on local NLP extraction quality and provenance documentation for reproducibility.