Stony Brook, Redshred Awarded Contract for AI-Powered Radiation Safety Technology

Stony Brook University, in partnership with Redshred, has been awarded a Phase I Small Business Technology Transfer (STTR) contract from the Defense Threat Reduction Agency (DTRA) to develop the Radiation AI Decision and Information Assistant for Nuclear Tasks (RADIANT).

This innovative AI-powered platform is designed to support military operations and nuclear safety through real-time, data-driven radiation guidance.

Led by Principal Investigator Manoj Mahajan of Stony Brook, the RADIANT project will integrate advanced artificial intelligence with health physics expertise to enhance decision-making in environments with radiological hazards. The technology is intended to act as an AI health physicist companion, delivering user-friendly, on-demand insights for tactical units, strategic mission planning and nuclear inspection tasks.

​​"RADIANT exemplifies the kind of high-impact innovation that emerges when academic research aligns with real-world operational needs," said Michael Kinch, chief innovation officer at Stony Brook University. "Projects like this reflect our commitment to expanding both the breadth and depth of industry partnerships, particularly with Long Island and New York-based companies, to help accelerate their success while applying trusted AI and deep scientific expertise to complex, high-risk challenges."

This award represents the university's first with DTRA in over a decade and builds on a growing record of success between Redshred and the university. Redshred is based in Stony Brook's Center of Excellence in Wireless Information and Technology (CEWIT) and the company is CEWIT's first expansion into nuclear and radiation health research.

Last year, Mahajan and Redshred were awarded a $1.8 million STTR Phase II contract from the U.S. Air Force for their AIRworthiness Model Automation for cerTification Evaluations (AIRMATE) system. This project helps expand two of CEWIT's focus areas: Medical Technology and Defense Research.

RADIANT's key innovation is its ability to extract and interpret complex data from nuclear and health physics documentation. This enables the system to see and understand information the way a trained human specialist would, drastically improving response time and accuracy in high-stakes situations.

Building on this capability, the project advances several of the DTRA's strategic initiatives by integrating advanced computing, human-machine interfaces and trusted AI to deliver clear, real-time radiation safety guidance. Designed to operate on small local devices, RADIANT aligns with the DTRA's goal of creating accessible, field-ready technologies that translate complex technical data into actionable insights for military operations and a range of civilian applications.

"CEWIT is looking forward to collaborating with Redshred on RADIANT and providing our expertise in health physics, large language models and AI to contribute to the safety of military personnel and civilians" said Mahajan, CEWIT director of special programs. "Our relationship and success with Redshred continues to grow with this additional project for DTRA and they have been an outstanding and model partner to the university."

DTRA is the first end-user for RADIANT, which could ultimately serve a broad array of military and civilian applications. The system is designed to guide safe radiation exposure practices in real time, offering tactical recommendations, such as time limits based on radiation dose thresholds. On a strategic level, it can help forecast long-term impacts on veteran health, which is an expanding area of CEWIT and Stony Brook Medicine and combat force capabilities due to radiation exposure, thereby assisting in force protection and mission execution.

"This touches on three of our strategic focus areas: Defense, Health, and Energy and leverages AI expertise and resources at Stony Brook University." said CEWIT Director Rong Zhao. "We're continuing to grow this partnership to create more commercialization opportunities with defense technology into both military and civilian sectors in the region."

Beyond its defense applications, RADIANT has strong commercial potential in sectors such as nuclear power, emergency response, special operations and international treaty monitoring. The platform could assist civilian operators, NGOs, and global organizations, such as the United Nations' International Atomic Energy Agency (IAEA) in radiation safety, compliance, and inspections.

Stony Brook's collaboration with Redshred underscores the university's commitment to fostering innovative technologies that address global challenges. The RADIANT project represents a bold step forward in the use of AI, and it demonstrates the fundamental relationship between research and innovation in developing scalable and impactful solutions to society's most pressing challenges.

"We're creating an innovative solution that transforms complex radiation data into actionable insights for non-experts in the field with our partners at Stony Brook University," said Jim Kukla of Redshred.