DOD SBIR 24.2 Annual

Release Date
April 17th, 2024
Open Date
May 15th, 2024
Due Date(s)
June 12th, 2024
Close Date
June 12th, 2024
Topic No.


Statistical Analysis of Neutron Relative Biological Effectiveness


Department of DefenseN/A


Type: SBIRPhase: BOTHYear: 2024


The Department of Defense (DOD) is seeking proposals for a research topic titled "Statistical Analysis of Neutron Relative Biological Effectiveness" under the SBIR program. The Defense Threat Reduction Agency is the specific branch responsible for this topic. The objective is to develop a statistical or computational model that accurately calculates neutron relative biological effectiveness (RBE) for lethality based on historical data. Neutron RBE is an important factor in calculating casualties, fatalities, and performance capabilities due to radiation effects in a nuclear detonation. The challenge lies in the difficulty of experimentally measuring neutron RBE, as it is dependent on various biological and physical factors. The goal is to utilize machine learning, statistical, or other computational techniques to analyze the large dataset of historical experimental data and develop a model that can accurately calculate neutron RBE. The project will be conducted in three phases: Phase I involves collecting experimental data and creating a searchable database, Phase II focuses on constructing the model and demonstrating its capabilities, and Phase III involves refining the model based on feedback and updating the data. The ultimate aim is to improve the accuracy of neutron RBE calculations and enhance radiation injury assessments. The solicitation is currently open, with a closing date of June 12, 2024. For more information and to submit proposals, interested parties can visit the DOD SBIR website.


OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Advanced Computing and Software; Biotechnology; Directed Energy


OBJECTIVE: To develop a statistical or computational model to accurately calculate neutron relative biological effectiveness for lethality based on historical data


DESCRIPTION: In a nuclear detonation, gamma and neutron radiation is released. This radiation can have detrimental effects on a human body. Methods to calculate casualties, fatalities, and performance capabilities due to radiation effects require combining the gamma and neutron doses. This is done by multiplying the neutron dose by a factor called the neutron relative biological effectiveness (RBE), and adding that value to the gamma dose. Neutron RBE is part of every radiation injury calculation which is important to many areas of the government from NIAID to COCOMS. It is then important to have an accurate neutron RBE value. Unfortunately, experimentally measuring neutron RBE is very difficult. Neutron RBE is dependent on many biological and physical factors such as tissue type, energy, gamma to neutron ratio, and dose rate. This makes comparing experimental data sets challenging. There have been tens of thousands of experiments probing neutron RBE dating back from the 1950s producing a large dataset. Although these experimental setups differ or their methods are lacking compared to modern abilities, they all are probing the same physical and biological processes. We believe that this data should not be disregarded, and applying modern machine learning or statistical or other computational techniques to this historical data, a model can calculate neutron RBE accurately. Based on time limits of SBIR, we would limit the scope to calculating neutron RBE to lethality endpoint.


PHASE I: Phase I will focus on collecting experimental data in a searchable database that will aid in the model development. Offerors should be able to understand the previous experiments and how they differ from each other. Phase I deliverables will include a final report and a demonstration of the architecture. The report should include statistical analysis of the experimental data.


PHASE II: Phase II effort will focus on the model construction from the collected data. Phase II deliverable will be a prototype demonstration and a final report. The demosntration will showcase calculating lethal neutron RBE value with confidence intervals. The final report should include explanation on the model including advanatages, disadvantages and assumptions made, and it can include suggestions for experimental data that can improve the results. The performer will include details about user interfaces (if applicable), any associated executables, and software requirements.


PHASE III DUAL USE APPLICATIONS: The performer should refine the model based on feedback from the demonstration. The data need to be updated according to the newest research. Maintenance and update will be performed in phase III.



Amy Creel, Tyler Dant, Rachel Jennings, Darren Oldson, Aaron Parks, Kiran Sewsankar, Christina Wagner. HENRE 4.0 Technical Reference Manual (DTRA-TR-23-011). Fort Belvoir: DTRA, 2023.
Bruce A. Carnes, Douglas Grahn. Neutron Issues in the JANUS Mouse Program. Argonne, Illinois: Biological and Medical Research Division, Argonne National Laboratory, 1990.
Daniela L. Stricklin, Jama VanHorne-Sealy, Carmen I. Rios, Lisa A. Scott Carnell, Lanyn P. Taliaferro. "Neutron Radiobiology and Dosimetry." Radiation Research 195 (2021): 480-496.
Daniela Stricklin, Kevin Kramer, Dave Crary, Robert Prins. Review of Deterministic Neutron RBEs for Survivable Personnel Radiation Exposures from Nuclear Detonation Simulations (DTRA-TR-19-001). Fort Belvoir: DTRA, 2019.
Hall, Eric J. Radiobiology for the Radiologist. Hagerston, MD: Harpers & Row, 1978.
James J. Conklin, Richard I. Walker. Military Radiobiology. San Diego, CA: Academic Press, 1987.
Robert E. George, Raymond L. Chaput, David M. Verrelli, Edward L. Barron. "The Relative Effectiveness of Fission Neutrons for Minature Pig Performance Decrement." Radiation Research 48 (1971): 332-345.
Strike, T. A. Acute Mortality of Mice and Rats Exposed to 14 MeV Neutrons. Bethesda, MD: Armed Forces Radiobiology Research Institue, n.d.


KEYWORDS: Relative Biological Effectiveness, lethal dosage, Radiation Exposures, NUDET simulations, radiobiology, deep learning, statistical analysis