In an effort to create meaningful data from EHRs, Atlanta-based Georgia Tech is engaged in creating record "phenotypes," or categories of records to identify candidates for clinical research, group patients with comorbidities and develop predictive models for health studies.
The project is part of a four-year, $2.1 million National Science Foundation research project between Georgia Tech and the University of Texas, Austin to build scalable algorithms that can sort records based on clinical data.
Vanderbilt University in Nashville, Tenn., will provide the initial EHR records and algorithm validation, with input from Evanston, Ill.-based Northwestern University as the team works to develop algorithms that will function at any health system.
Researchers hope that by being able to work with multiple record "phenotypes," they will be able to assess risk factors for development of chronic health conditions across many different groups of patients, according to a release.