Federal Grant Award Summary

The National Institute of Mental Health (NIMH) awarded Massachusetts General Hospital a $790,560 Project Grant under the Mental Health Research Grants program (CFDA 93.242) effective April 1, 2026 through March 31, 2030. The award funds development of advanced artificial intelligence models to predict differential antidepressant treatment responses by integrating electronic health record (EHR) data with genomic information. The research addresses critical limitations in current treatment selection approaches, which largely rely on trial-and-error methodology and underutilize available patient data. The project will leverage natural language processing, predictive modeling, and causal inference techniques to improve treatment prediction accuracy and patient outcomes in depression care.

The award supports three primary research objectives: (1) developing a large language model-based active learning framework to identify patients initiated on antidepressants, assess treatment responses, and extract depression-related clinical information from unstructured notes; (2) creating uncertainty-aware EHR-based prediction models that account for cases outside training data and residual confounding; and (3) integrating EHR data with genome-wide, pathway-specific, and polygenic predictors to enhance response prediction accuracy. The deliverables include computational tools, validated prediction models, and datasets that will advance precision medicine approaches to antidepressant treatment selection and contribute to reducing treatment delays and patient suffering from depression.