FEMR is a Python package for manipulating longitudinal EHR data for machine learning, with a focus on supporting the creation of foundation models and verifying their presumed benefits in healthcare. Such a framework is needed given the current state of large language models in healthcare and the need for better evaluation frameworks.
The currently supported foundation models are CLMBR and MOTOR.
FEMR by default supports the OMOP Common Data Model developed by the OHDSI community, but can also be used with other forms of EHR / claims data with minimal processing. FEMR has been used to process data from a variety of sources, including MIMIC-IV, Optum, Truven, STARR-OMOP, and SickKids-OMOP.
For details, see the github repo – https://github.com/som-shahlab/femr