PRIMED-AI: Multi-use Framework Playbooks (UG3 Clinical Trial Not Allowed)
NIH will soon fund projects creating frameworks and best practices for developing and managing AI tools that integrate clinical imaging with other health data to improve personalized medicine.
Funder: National Institutes of Health
Due Dates (Anticipated): June 2027 (full application, projected) | June 2026 (application deadline, projected)
Funding Amounts: Estimated total program funding: $650,000; award size and duration TBD in final solicitation.
Summary: Supports development and initial testing of frameworks (“playbooks”) for responsible use of AI-based tools integrating clinical imaging with multimodal health data to advance precision medicine.
Key Information: This is a forecasted opportunity; application instructions and eligibility may change in the final NOFO.
Description
The NIH Common Fund, with participation from other NIH Institutes and Centers, anticipates releasing a Notice of Funding Opportunity (NOFO) for the Precision Medicine with Artificial Intelligence – Integrating Imaging with Multimodal Data (PRIMED-AI) program. This initiative aims to support projects that develop, document, and test “playbooks” of frameworks for the responsible development and use of AI models that integrate clinical imaging with other health data types. The goal is to enhance personalized medicine approaches for patients with chronic and other health conditions.
Funded projects will focus on:
- Developing frameworks for responsible AI use, error mitigation, and technical management of data and algorithms.
- Advancing data ontology, linkage, and harmonization across multimodal health datasets.
- Preparing tools and processes for regulatory approval (e.g., FDA).
- Promoting collaborative, multidisciplinary research that combines expertise in AI, imaging, data science, and biomedical informatics.
The program will use the UG3 cooperative agreement mechanism and is designed for teams with expertise in AI-driven integration of clinical imaging and complex health data.
