The Consortium for Verification Technology Is Working On An Algorithm To Discriminate Radiation Sources in Shipping Containers And Vehicles

The Consortium for Verification Technology Is Working On An Algorithm To Discriminate Radiation Sources in Shipping Containers And Vehicles

One major concern about nuclear power is the fear that terrorists or rebels could gain access to nuclear materials that they could use construct an actual fission bomb or just a dirty bomb that would disperse radioactive materials over a wide area. A number of technologies have been developed to scan containers and vehicles for radioactive materials, but they have problems. Now a new algorithm has been developed that should improve both detection and discrimination with respect to scanning for radioactive materials.
     The team of researchers is a collaboration of University of Michigan, University of Illinois, Los Alamos National Laboratory, Heriot-Watt University (Edinburgh, Scotland) and University of Edinburgh staff. Angela DiFulvio is an assistant professor of nuclear, plasma and radiological engineering at the University of Illinois and corresponding author of the study recently published in Nature Scientific Reports. She was a former postdoctoral researcher in the Detection for Nuclear Nonproliferation Group at U-M led by Sara Pozzi, a professor of nuclear engineering and radiological science. Pozzi is also a senior author on the paper.
     DiFulvio said, “We hope that the findings will be helpful in reducing the false positive alarms at radiation portal monitors, even in scenarios with multiple sources present, and enable the use of cost-effective detectors, such as organic scintillators.” 
       Individual nations have to protect their citizens from the possibility of nuclear terrorism. Nuclear security equipment and protocols detect and deter the smuggling of nuclear materials across national borders. These materials include enriched uranium, weapons grade plutonium or materials that produce a lot of radiation
    The research team has developed an algorithm that is able to identify weak radiation signals such as those that might be from plutonium encased in radiation absorbent materials. The new algorithm can even work when confronted by a high radiation background from everyday sources such as cosmic rays raining down from space and radon released from uranium in rock and soil.
    The research team believes that their algorithm could be used to improve the capability of radiation portal monitors at national borders to discriminated between possible smuggling activity and benign radiation sources.
    There are naturally occurring radioactive materials such as some ceramics and fertilizers or radionuclides in recently treated nuclear medicine patients. These can all set off false positive alarms at radiation scanning facilities. Sara Pozzi said, “There’s also the concern that somebody might want to mask a radioactive source, or special nuclear material, by using naturally occurring radioactive materials such as granite or kitty litter.”
“As vehicles or boxes are scanned, the data from the detector can be put through these algorithms that unmix the different sources. The algorithms can quickly identify whether special nuclear materials are present.”
    If someone wants to hide radioactive materials being smuggled by mixing it with benign radiation sources, it is difficult to quickly differentiate these two different sources of radiation. Specialists in machine learning used the data collected by Pozzi’s team to train algorithms to look for the signatures of radioactive materials that could be used to make a fission bomb.
     Alfred Hero is the John H. Holland Distinguished University Professor of Electrical Engineering and Computer Science and R. Jamison and Betty Williams Professor of Engineering at U-M and a co-author on the paper. He said, “We crafted an unmixing model that both reflects the basic physics of the problem and was also amenable to fast computation. This work is a powerful example of the benefit of close and sustained collaboration between computational data scientists and nuclear engineers, resulting in major improvement of nuclear radiation detection and identification.”
     This research began at U-M as part of the Consortium for Verification Technology, a five-year $25 million nuclear nonproliferation research program funded by the U.S. Nuclear National Security Administration and led by Pozzi.