71st Annual Meeting of the Health Physics Society, National Harbor, MD, July 6-9, 2026
Three USTUR faculty members gave four presentation at the 2026 Annual Health Physics society meeting.
Validating curium biokinetic models using human tissue data
Sergey Tolmachev (USTUR), George Tabatadze (USTUR), Elizabeth Thomas (USTUR)
The United States Transuranium and Uranium Registries’ (USTUR) partial tissue donor was exposed to airborne 244Cm due to a glove-box failure. He died from hypertensive heart disease 52 years after the intake. A total of 35 soft tissue and 16 bone samples collected at autopsy were radiochemically analyzed for 244Cm. Activity concentration in systemic organs followed the pattern: skeleton>liver>kidney>muscles. The total systemic 244Cm activity was estimated to be 0.82±0.16 Bq with 92.3% retained in the skeleton, followed by 3.5% in muscles, 2.2% in the liver, and 0.13% in the kidney (skeleton>muscles>liver>kidney). A total of 0.018±0.002 Bq was deposited in the respiratory tract tissues. The activities in the respiratory tract, liver, and skeleton were used to estimate 244Cm intake and committed effective dose. The ICRP default biokinetic models, assuming 84% of curium oxide, nitrate, and chloride, and 16% of type S material, described the data well (c2 = 0.615; p = 0.43). The total 244Cm intake was estimated at 230±39 Bq, corresponding to a committed effective dose of 1.62±0.24 mSv. [USTUR-0727-26A]
Bayesian analysis of uncertainty in 239Pu biokinetic predictions: A study of former Hanford workers
Maia Avtandilashvili (USTUR), Martin Šefl (AMEDIS, Czechia), Joey Y. Zhou (DOE), Sergey Y. Tolmachev (USTUR)
The ICRP biokinetic and dosimetry models do not explicitly account for dosimetric uncertainty. Because time dependent radionuclide activity retained in organs and tissues directly determines radiation dose, uncertainties in biokinetic model predictions of organ activity can serve as surrogates for uncertainties in dose estimates. A Bayesian approach provides a distribution of possible dose values rather than a single point estimate and true dose is assumed to lay within the modeled posterior distribution. This study was conducted to validate that assumption. Uncertainties in predicted 239Pu organ activities were evaluated for 14 former Hanford workers, voluntary tissue donors to the USTUR. Urine bioassay data and post mortem 239Pu measurements in the liver, skeleton, and respiratory tract were used. Latin hypercube sampling was used to generate parameter sets for each model realization, varying the rapidly dissolving fraction and the slow dissolution rate. For each realization, intake was estimated by the maximum likelihood fitting of the urine data, and organ activities were predicted from the intake. The modeled activity distributions did not cover the measured post-mortem values for 79% and 86% for the liver and skeleton, respectively. For the respiratory tract, urine data were not informative. For the liver and skeleton, the means of the modeled posterior distributions differed from measured activities by 69% and 57%, respectively. The organ activity predictions were not conservative in 50% cases for the liver and 43% for the skeleton, which can be problematic from a radiation protection perspective. These findings are consistent with published results for Los Alamos and Rocky Flats worksites. [USTUR-0730-26A]
Quantitative digital autoradiography of radionuclides in human tissues
George Tabatadze (USTUR), Maia Avtandilashvili (USTUR), Sergey Y. Tolmachev (USTUR)
An ionizing radiation quantum imaging detector (iQID) is used at the United States Transuranium and Uranium Registries as a non-destructive digital autoradiography technique for visualizing the micro-distribution of alpha-emitting radionuclides in human tissue samples. This work extends iQID imaging beyond qualitative mapping to support quantitative assessment of radionuclide activity concentrations. Plastic-embedded human bone sections from a historical radium exposure case were imaged to resolve spatial heterogeneity associated with cortical and trabecular bone regions. Region-specific segmentation was coupled with a computational modeling of geometric detection efficiencies for alpha particles emitted within the tissue. These sample-specific efficiencies were used to derive estimates of radionuclide activity concentrations. Independent activity concentration estimates were also derived using International Commission on Radiological Protection biokinetic model predictions. iQID-based activity concentration estimates were compared with model predictions, with differences less than 3 percent. These results indicate that digital autoradiography with iQID can support quantitative, non-destructive measurements of alpha-emitting radionuclide activity concentrations while preserving microscale spatial distribution information. This approach is particularly suited to studies involving archived human tissue specimens, where sample preservation constrains destructive analytical methods. [USTUR-0729-26A]
Protecting public health through comprehensive radon monitoring and dosimetry in urban Tbilisi
George Tabatadze, Maia Avtandilashvili, Magda Metskvarishvili, Giorgi Melikadze, Natia Beriashvili, Giorgi Giorganashvili, Kakha Gorgadze, Shorena Dekanosidze, Iamze Kalandadze, Roy Dunker, Samson Pagava, Medea Burjanadze
Radon is a naturally occurring radioactive noble gas and a major contributor to indoor radiation exposure. In Georgia, quantitative data on indoor radon concentrations remain limited, particularly in public and occupational buildings, constraining exposure assessment and mitigation planning. This project implements a structured radon measurement and dosimetric assessment program in urban Tbilisi, combining passive and active radon measurements to characterize time-integrated concentrations and short-term temporal variability. Measurements are conducted across residential, public, and selected occupational buildings using a site selection framework informed by geological age, soil and bedrock composition, proximity to tectonic structures, hydrogeological features, population and building density, and the absence of historical radon data. Measured indoor radon activity concentrations are analyzed spatially to support high-resolution radon mapping, and annual effective dose estimates are derived using internationally accepted dose conversion coefficients and occupancy assumptions, with uncertainty considerations linked to measurement duration. Results are evaluated relative to the Georgian regulatory reference level of 300 Bq m<sup-3 and relevant European and international guidance, providing a technical basis for population-level exposure assessment and identification of areas where mitigation may be warranted.
This work was supported by Shota Rustaveli National Science Foundation of Georgia (SRNSFG), FR-24-2178, Protecting public health through comprehensive radon monitoring and dosimetry in urban Tbilisi. [USTUR-0732-26A]