PHILADELPHIA — In a bold step toward transforming mental health care, Dr. Yong Chen, a senior fellow at the Leonard Davis Institute of Health Economics and professor at Penn Medicine, has received an $8 million grant from the National Institute of Mental Health (NIMH) to lead a data-driven initiative using artificial intelligence (AI) to develop more precise psychiatric phenotypes.
The five-year project is part of the NIMH’s $150 million Individually Measured Phenotypes to Advance Computational Translation in Mental Health (IMPACT-MH) initiative. This ambitious research effort, conducted in partnership with Yale University and the Mayo Clinic, will bring together eight research teams focused on creating novel, computationally defined clinical profiles that support personalized mental health diagnosis and treatment.
Dr. Chen, who directs the Penn Computer, Inference and Learning Lab (PennCIL) and the Center for Health AI and Synthesis of Evidence (CHASE), will lead the Data Coordinating Center for the multi-institutional project.
“The IMPACT-MH project is highly unique because it seeks to apply precision medicine concepts in the mental health field, where heterogeneity is especially challenging,” said Chen, Professor of Biostatistics, Epidemiology, and Informatics at Penn’s Perelman School of Medicine.
Integrating Multimodal Data for Personalized Care
A key goal of the project is to integrate multimodal data—behavioral, clinical, and biological—into advanced computational models to create highly detailed phenotypic profiles of individuals with mental health conditions. These profiles will go far beyond traditional diagnostic categories by incorporating cognitive assessments, brain imaging data, hormone levels, and behavioral patterns.
Chen emphasized the project’s novelty and scope:“This level of integration is not yet common in mental health research. While PennCIL has led computational methods in various clinical disciplines, this project stands out for its scale and focus on mental disorders. It will also help establish new standards for mental health data collection, representation, and analysis.”
Tackling Diagnostic Ambiguity
Mental health care has long grappled with the limitations of broad diagnostic labels like depression, anxiety, or bipolar disorder, which often fail to capture the diversity of patient experiences. Clinicians typically rely on observable symptoms and structured interviews, which may miss subtler biological or cognitive variations that influence treatment outcomes.
By developing richer, data-informed phenotypes, the project aims to improve clinical decision-making and treatment personalization.
“In a clinical situation, the finished IMPACT-MH product could be used to improve diagnosis, prognosis, and treatment personalization,” Chen explained. “Clinicians could use these signatures to predict treatment response, monitor progress, and adjust care based on an individual’s unique clinical signature.”
A Vision for Precision Psychiatry
At its core, the IMPACT-MH initiative represents a shift toward precision psychiatry—an emerging field that mirrors the personalized approaches seen in oncology and cardiology. By moving beyond surface-level diagnoses and embracing the complexity of mental health, Chen and his collaborators hope to deliver more accurate, efficient, and compassionate care.
If successful, the project could pave the way for the next generation of mental health tools and frameworks—ones that offer both clinicians and patients clearer paths to recovery.
