Grant: Engage Grants Program

In a large scale release of radionuclides from a nuclear accident or malicious event, there may be masscasualties requiring rapid triage and treatment to prevent or mitigate illness from the radiation exposure. The""Gold Standard"" for determining the magnitude of radiation exposure post-event is a cytogenetic techniquescalled chromosome aberration analysis (CAA). The most useful CAA technique is chromosome dicentricanalysis. One of the issues for using dicentric analysis in a mass casualty event is that it is extremely timeconsuming, as typically it is personnel intensive to look at metaphase spreads for aberrations. To speed up theprocess, automated software is desirable to examine the spreads and determine the number of aberrations.Since aberration count is proportional to dose, it is then possible to estimate dose in exposure persons.CytoGnomics has developed a state-of-the art image analyzer for estimating dicentrics, called ADCI.However, to demonstrate the efficacy of this software for mass casualty events, the image analysis softwareneeds to be benchmarked. To this end, a simulated biodosimetry analysis on a population, incorporatinggeographic information system (GIS)-calibrated population dose estimates into these analyses will beperformed. Patient datasets will be sampled from radiation calibrated metaphase cell images fromCytoGnomix. HPAC software will be used to model physical dosimetry exposures in an actual radiologicaldispersal scenario. HPAC estimates of physical doses will be used to calibrate the cytogenetic biodosimetrydata. Key issues to be addressed from this analysis are: 1) To interpolate biological radiation exposures forindividuals who were not tested using dose estimates from those who were tested, 2) To determine the densityof individuals affected, 3) To determine the minimum number of individuals that need to be sampled from thetotal population exposed, 4) To determine the variance on dose estimates based on population locationclustering and sampling of clusters. We will further determine 5) the compute time required to complete theseanalyses. The overall goal of this project will be to provide a fully-operational prototype system.