Robert Carroll, PhD
Robert Carroll, PhD, is a Research Assistant Professor of Biomedical Informatics. His focus is on research using electronic health record (EHR) data tied to DNA biobanks. In the field of phenotyping individuals through EHR data, he has developed and applied algorithms and machine learning methods to predict disease state and drug response. These works include showing the portability of a logistic regression model for phenotype identification across several sites. He is the primary author and maintainer of a package for R implementing phenome wide association studies (PheWAS), a tool that makes the method more accessible to researchers and has been used throughout the eMERGE Network and elsewhere. He is an investigator in the eMERGE Network and other consortia, implementing and designing methods intended for EHR-based cohorts which have led to new genetic associations. He also works with the Vanderbilt Institute for Clinical and Translational Research (VICTR), assisting in the development of the Synthetic and Research Derivatives.