PRIMED-AI: Data-to-Model Academic-Industrial Partnerships (D2M-AIP) for Precision Medicine with AI: Integrating Imaging with Multimodal Data (UG3/UH3 Clinical Trial Optional)

The PRIMED-AI Data-to-Model Academic-Industrial Partnerships for Precision Medicine with AI program is a federal funding opportunity administered by the National Institutes of Health through the Office of Strategic Coordination within the Office of the NIH Director. This initiative is part of the NIH Common Fund, which supports cross-cutting programs designed to accelerate biomedical discovery and translation. The program is administered by the National Cancer Institute on behalf of NIH and reflects a broad, multi-institute effort to advance precision medicine through artificial intelligence and data integration. The primary purpose of this funding opportunity is to catalyze the development and validation of innovative clinical decision support tools that integrate clinical imaging with multimodal health data. These tools are intended to function as software as a medical device and are expected to demonstrate measurable improvements in patient outcomes or healthcare delivery processes. The program specifically targets the translational gap between early-stage AI prototypes and clinically viable, deployable tools by supporting multidisciplinary academic-industrial partnerships that can combine technical innovation with commercialization pathways. Funding is structured as a phased cooperative agreement with two linked components: an initial UG3 phase focused on exploratory development and pilot validation, followed by a UH3 phase dedicated to refinement and performance validation. Applicants may request up to 450000 dollars per year in direct costs during the UG3 phase and up to 800000 dollars per year during the UH3 phase, with a total project period not exceeding five years. Funds may be used for personnel, data integration, computing resources, validation activities, regulatory engagement, and related research costs. Cost sharing is not required under this program. Eligible applicants include a wide range of domestic and international entities such as higher education institutions, nonprofit organizations, for-profit companies, and government entities. However, a core requirement is the formation of a partnership that includes at least one academic institution and one industry partner. These partnerships must demonstrate access to high-quality multimodal datasets and the capability to develop and validate AI models anchored in clinical imaging. Foreign collaborations are allowed under specific conditions, though foreign subawards are prohibited. Applications must be submitted through NIH-approved systems such as Grants.gov or ASSIST and must comply with all requirements outlined in the NIH application guide. Required components include detailed research plans structured around five core pillars, milestone-driven timelines, data management plans, and an error mitigation and technical management plan. Applications will undergo rigorous peer review based on criteria including scientific significance, innovation, methodological rigor, feasibility, and the strength of the interdisciplinary team. The application timeline includes an open date of September 19, 2026, and a submission deadline of October 19, 2026. Applications must be submitted by 5:00 PM local time of the applicant organization. Review is expected to occur in March 2027, followed by advisory council review in May 2027 and an earliest project start date in July 2027. This funding opportunity does not indicate recurring cycles and is considered a one-time announcement unless reissued in the future. Overall, the PRIMED-AI D2M-AIP program represents a significant federal investment in advancing AI-driven precision medicine by fostering collaboration between academia and industry. It emphasizes robust data integration, regulatory readiness, and real-world applicability, positioning funded projects for eventual clinical adoption and commercialization.

Ensure strong academic industry collaboration demonstrate access to high quality multimodal data and align project with clinical impact and commercialization pathway