(By Alongkorn for AdobeStock)
Key Points
FAMU-FSU College of Engineering Dean Suvranu De is leading the FSU portion of a national ARPA-H project to build a virtual testing platform for robotic stroke interventions.
FSU’s share of the award is approximately $2.7 million, part of a total award of up to $17.5 million from ARPA-H to Kitware.
De’s team will build real-time biomechanical models of blood vessel and clot mechanics to power the platform’s core simulation engine.
The project, called TAIR, brings together Kitware, Johns Hopkins, UNC, the University of Pittsburgh, FSU and Lightside Surgical Inc.
De’s group has a prior open-source collaboration with Kitware on iMSTK, a surgical simulation toolkit. All software and imaging data produced through the project will be released as open source.
FSU’s roughly $2.7 million share of the award will fund Dean Suvranu De’s team in building real-time biomechanical models for a national testbed evaluating robotic stroke-treatment devices.
Suvranu De, Google Endowed Dean of the FAMU-FSU College of Engineering and professor in the Department of Mechanical & Aerospace Engineering, is leading the Florida State University portion of a new national effort to build a virtual testing platform for robotic stroke interventions. This project is supported by up to $17.5 million from the Advanced Research Projects Agency for Health (ARPA-H)’s Autonomous Interventions and Robotics (AIR) program, led by Dr. Ileana Hancu.
De’s team will develop the real-time biomechanical simulation capabilities at the core of the platform, known as the Testbed for Autonomous Interventions and Robotics (TAIR). The FSU award, approximately $2.7 million, will support research faculty, postdoctoral researchers, support staff, software development and computational infrastructure over the life of the project.
“My team will lead the development of the real-time biomechanical simulation capabilities that underpin a next-generation virtual testbed for autonomous endovascular stroke interventions,” De said. His group will build high-fidelity models of blood vessel and clot mechanics—including vessel deformation, clot fragmentation, device-tissue interaction and fluid-structure coupling—and integrate them into an open-source simulation platform to support development, testing, verification and FDA evaluation of robotic stroke-treatment systems. The team will also contribute to the project’s software architecture, validation and performance metrics, open-source releases and technical coordination with ARPA-H and the FDA.
What is the TAIR Project?
TAIR is a four-year effort to build an open-source virtual environment that lets researchers simulate stroke procedures and evaluate robotic medical devices across a range of anatomies and interventions before clinical studies begin. It is part of ARPA-H’s Autonomous Interventions and Robotics (AIR) program, led by ARPA-H Program Manager Ileana Hancu.
The project is directed by Principal Investigator Rachel Clipp and Program Manager Jeff Webb of Kitware, a software research and development company based in Clifton Park, New York, that builds custom scientific software on open-source technology. Kitware leads the project and brings together researchers from Johns Hopkins University, the University of North Carolina, the University of Pittsburgh, FSU and Lightside Surgical Inc.
De’s group brings a long-standing collaboration with Kitware to the project: the two have jointly developed iMSTK, the open-source Interactive Medical Simulation Toolkit used for interactive surgical simulation.
How Will the Simulation Platform Work?
Kitware and the clinical team will collect and curate a large-scale, open-access medical imaging dataset, including CT/MR angiography and video fluoroscopy data, for use by the broader research community. UNC will serve as the project’s clinical lead, providing imaging data, annotation and neurovascular interventional expertise to ensure the simulations are clinically realistic.
The platform will use segmentation and shape-analysis algorithms to generate anatomical reconstructions, paired with multiphysics models—largely developed by De’s FSU team—to simulate robotic interventions. Kitware’s open-source Pulse Physiology Engine and open-source extended position-based dynamics will model blood flow, patient physiology, vessel mechanics and medical device-to-tissue interactions in real time. Johns Hopkins’ open-source DeepDRR tool will be extended to improve the realism and speed of virtual fluoroscopy simulations.
TAIR will also provide an open interface that allows robotic systems to connect to the simulator and be evaluated against FDA-defined performance metrics.
“By providing an open interface, we can enable robotic systems to communicate with the TAIR platform for navigation and performance evaluation using FDA-defined metrics, while supporting human-in-the-loop testing throughout the development process,” Clipp said.
What Happens After the Platform is Built?
All software developed through TAIR will be released as open source, and the medical imaging dataset will be made available through open-access repositories. The team also plans to submit TAIR to the FDA’s Medical Device Development Tools (MDDT) Program, building a foundation to support future robotic and image-guided interventions beyond stroke care.
Acknowledgment and disclaimer: “This research is funded, in part, by the Advanced Research Projects Agency for Health (ARPA-H). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Government.”
Editor’s Note: This article was adapted from a Kitware press release (Aug. 6, 2026) and additional source material provided by the researchers, with the assistance of Claude (Sonnet 5), an AI model developed by Anthropic. AI-assisted editing was used for structural reorganization, AP style compliance, and SEO/AEO optimization. All facts and figures are sourced directly from the original press release and provided materials; no information was independently verified or added by the AI. This content was fact-checked and edited by FAMU-FSU College of Engineering communications staff.
RELATED ARTICLES
Using AI-Powered Robotics to Revolutionize Stroke Recovery and Balance Rehabilitation
Engineering Researchers Take Robotic Touch to a New Level with Novel Haptics
Building a Better Surgeon: Researchers Develop Artificial Intelligence Tool for Surgical Training
FAQ
A four-year, ARPA-H-funded effort to build an open-source virtual testing platform that simulates stroke procedures and evaluates robotic medical devices before clinical studies.
FSU’s share of the award is approximately $2.7 million, part of a total award of up to $17.5 million to Kitware from ARPA-H.
Suvranu De, Google Endowed Dean of the FAMU-FSU College of Engineering and professor in the Department of Mechanical & Aerospace Engineering.
Real-time biomechanical simulation of blood vessel and clot mechanics, device-tissue interaction and fluid-structure coupling, integrated into TAIR’s open-source simulation platform.
Kitware (lead), Johns Hopkins University, the University of North Carolina, the University of Pittsburgh, Florida State University and Lightside Surgical Inc., under ARPA-H’s Autonomous Interventions and Robotics program.
De’s team and Kitware jointly developed iMSTK, an open-source toolkit for interactive medical/surgical simulation.
Yes. All software developed through the project will be released as open source, and the medical imaging dataset will be made available through open-access repositories.