Vanderbilt University Medical Center researchers have secured a funding extension from the Advanced Research Projects Agency for Health (ARPA-H) to expand use of an artificial intelligence-guided platform for extracting and organizing critical information from unstructured clinical notes and reports in medical records.
The Democratized AI-Guided Chart Abstraction Platform (DAGCAP) project has received a one-year extension of its ARPA-H award to three years (2024-2027), adding $2 million in funding for a total of just under $4 million.
Led by Daniel Fabbri, PhD, associate professor of Biomedical Informatics, and Christine Micheel, PhD, research assistant professor of Medicine in the Division of Hematology and Oncology, the DAGCAP technology has since spun out into a commercial venture, Brim Analytics, where it has seen rapid growth and adoption.
Over the past year, more than 120 research and clinical teams at VUMC have leveraged Brim through the Vanderbilt Institute for Clinical and Translational Research (VICTR) core service to perform chart abstraction for a wide range of uses, from surgical planning and cancer registries to orthopedic research.
“Brim has quickly become part of our intelligence system infrastructure at Vanderbilt,” said Neal Patel, MD, MPH, chief informatics officer at VUMC. “It provides a secure and effective way for our faculty and staff to harness large language models for chart abstraction. We’re seeing impact both in advancing research and in improving operational efficiency.”
Brim’s reach now extends well beyond Vanderbilt. The software is being used at health care institutions nationwide to support clinical research, cancer registries, maternal health initiatives and other projects. As part of ARPA-H’s Biomedical Data Fabric Toolbox program, Brim is contributing to a broader effort to make data within clinical notes more accessible to researchers, staff and health care operations teams.
Fabbri said the funding will allow DAGCAP and Brim to expand their focus on clinical trial pre-screening, clinical workflows, registry abstraction and retrospective research across diverse care settings.
“This extension underscores both the promise and the proven impact of our work,” said Fabbri. “We are excited to continue collaboration across Vanderbilt, ARPA-H and other deployment sites to accelerate research and improve patient care through AI-powered chart abstraction.”