Observational Health Data Sciences and Informatics

An open-science collaborative turning observational health records into evidence through the OMOP common data model and federated analytics.

Location: New York, NY

Website

1 model

Models (1)

CEHR-GPT

Columbia University / Washington University in St. Louis / NewYork-Presbyterian Hospital / Observational Health Data Sciences and Informatics / New York Genome Center

Released February 6, 2024

———

EHR foundation model that writes patient histories as token sequences carrying explicit visit and day-interval tokens, invertible back to OMOP tables.

Language model