SAM #DARPA-PS-26-26

DARPA BTO seeks R&D for real-time cardiovascular digital twin models under VITAL program

DARPA's Biological Technologies Office (BTO) is seeking proposals for the VITAL program, focused on developing real-time digital twin models of the cardiovascular system for advanced medical decision support. - Government Buyer: - Defense Advanced Research Projects Agency (DARPA), Biological Technologies Office (BTO) - OEMs and Vendors: - No specific Original Equipment Manufacturers (OEMs) or vendors are named in the solicitation - Products/Services Requested: - Research and development of continuously updating digital twin models for the cardiovascular system - High-fidelity, patient-specific models integrating clinical data and biological physics - Reduced-order models (ROMs) using physics-informed neural operators and differentiable simulations for real-time prediction - Development and validation of models for chronic heart failure and acute hemorrhagic shock - Integration of models for use in both battlefield and clinical care settings - Unique or Notable Requirements: - Models must enable causal, prediction-driven decision support, allowing clinicians and autonomous systems to evaluate treatment options before intervention - Emphasis on shifting medical care from reactive to anticipatory intervention - Program duration is 36 months, split into two phases (high-fidelity model development and validation, then reduced-order model integration) - Multiple awards anticipated under Other Transaction for Prototype agreements - Key milestones include model demonstrations, workshops, and final prototype delivery

Description

The Defense Advanced Research Projects Agency (DARPA) is soliciting innovative approaches to address challenges in developing continuously updating digital twin models of the cardiovascular system that combine patient data with biological physics to simulate in real time possible interventions and to predict their outcomes before actual intervention on a patient. The VITAL program will establish a foundation for causal, prediction-driven decision support in medicine, enabling clinicians and future autonomous systems to evaluate treatment options before they are applied. By defining when mechanistic digital twins, reduced-order models, and data-driven inference are reliable and actionable, VITAL will shift medical care from reactive assessment to anticipatory intervention. This capability has the potential to improve outcomes across acute, chronic, and emergent clinical settings.

View original listing