The Defense Advanced Research Projects Agency (DARPA), through its Biological Technologies Office (BTO), has issued Special Notice DARPA-SN-26-50 announcing additional research areas of interest for the forthcoming Virtual-Integrated Twin for Autonomous Lifesaving (VITAL) program. This Special Notice serves as public notification ahead of a formal solicitation that will be released on SAM.gov.
DARPA’s overarching goal for VITAL is ambitious and highly consequential: to develop continuously updating computational models of the cardiovascular system that integrate patient data with biological physics to predict outcomes in real time. In practical terms, the program seeks to enable clinicians and future providers to assess treatment options for acute and chronic pathologies using predictive, physics-grounded digital twins of human cardiovascular physiology.
Solicitation Overview and Technical Objectives
At its core, VITAL aims to establish a foundation for causal, prediction-driven decision support using high-fidelity (HF) digital twins that explicitly represent the physical, biochemical, and anatomical dynamics governing cardiovascular physiology. These HF models are intended to serve as a baseline predictive reference across static, chronic, and acute physiological regimes. For each regime, performers will be expected to rigorously quantify computational speed, forecast horizon, predictive accuracy, uncertainty bounds, data requirements, sensitivity to parameter estimation, segmentation error, and measurement sparsity. A central outcome of the program is a rigorous characterization of HF model capability limits, providing evidence-based guidance on when such models are suitable and where fundamental limitations remain.
Recognizing the tradeoff between mechanistic fidelity and real-time responsiveness, DARPA also intends to explore reduced-order models (ROMs) derived from HF dynamics. While HF models provide strong causal interpretability, ROMs are better suited for real-time execution and continuous updating, particularly when coupled to streaming sensor data. Through systematic verification, validation, and uncertainty quantification, VITAL will determine when and where HF models, ROMs, and measurement-driven inference approaches are reliable and appropriate.
A defining feature of the program is the establishment of an Image-to-Physics-to-Twin pipeline. This pipeline will automatically integrate multimodal clinical imaging such as magnetic resonance imaging (MRI), computed tomography (CT), computed tomography angiography (CTA), and ultrasound with sparse physiological and biochemical measurements to construct patient- or archetype-specific high-fidelity digital twins. Segmented anatomy will be translated into vascular networks, organs, and injury geometries spanning chronic and acute conditions, with localized three-dimensional injury-site solvers capturing bleeding and impedance dynamics and whole-body zero-dimensional and one-dimensional physiology models propagating global responses such as shock.
VITAL is structured as a two-phase program:
- Phase 1 will establish technical credibility through the development of HF models and rigorous quantification of their performance limits.
- Phase 2 will evaluate performance tradeoffs associated with transitions from HF models to ROMs, including scalability, real-time execution, and intervention-forecasting accuracy.
DARPA has also explicitly encouraged teaming to ensure the breadth of expertise required to achieve VITAL’s objectives.
Importantly, this Special Notice does not constitute a formal solicitation for proposals and is not a Request for Information (RFI); DARPA will not accept submissions in response to this notice.
Key Dates
At this time, the Special Notice does not provide specific proposal deadlines. However, the following timing and administrative notes are relevant:
- Future Solicitation Release: The formal solicitation will be posted to SAM.gov when released.
- Blast List Sign-Up: Interested parties may email [email protected] to be added to the distribution list for future updates, including any Proposers Day announcement and program solicitation publication.
- DARPAConnect Resources: DARPAConnect offers preparatory modules such as “Making the Most of a Proposers Day” and “Understanding DARPA Broad Agency Announcements,” which may be valuable as teams prepare for a future release.
No specific submission deadlines, Proposers Day dates, or award timelines are included in this Special Notice.
Funding Expectations
The Special Notice does not provide information regarding the number of anticipated awards, the expected size of awards, or the period of performance. As this is not yet a formal solicitation, funding details have not been published. These elements are expected to be addressed in the forthcoming program solicitation.
Advancing Human Health Through Predictive Cardiovascular Modeling
VITAL represents a forward-looking investment in digital medicine and lifesaving intervention. By developing high-fidelity, physics-informed digital twins of the cardiovascular system that can predict outcomes in real time, DARPA is seeking to transform how clinicians assess and manage acute injuries, bleeding events, shock, and chronic cardiovascular pathologies. The ability to forecast physiological response to intervention under sparse and noisy sensing conditions has clear implications for battlefield medicine, emergency response, and potentially civilian trauma and critical care.
For organizations with strengths in computational modeling, cardiovascular physiology, medical imaging integration, artificial intelligence, physics-informed machine learning, and real-time systems, VITAL presents a compelling opportunity to shape the future of predictive clinical decision support.
If your company has considered applying for federal funding, your federal funding journey starts here. EverGlade is a national firm that helps organizations win and manage federal awards. We offer services ranging from Pursuit, Proposal and Post-Award support to comply with federal regulations at agencies including BARDA, ARPA-H, NIH, DTRA, JPEO, DOD, DIU, DOE, and DARPA.
