The International Commission on Radiological Protection (ICRP) has developed the Human Alimentary Tract Model (HATM) to calculate radiation doses from the ingestion of radionuclides for the protection of workers and the public. In parallel, the ICRP's Occupational Intake of Radionuclides (OIR) series provides biokinetic models and dose coefficients based on a reference human, primarily for regulatory purposes. Although these coefficients are not usually checked for uncertainties, the investigation of such uncertainties is crucial to ensure their reliability in radiation protection. This study uses INTDOSKIT, a software tool developed using the R programming language and RStudio as the Integrated Development Environment (IDE), to calculate doses and explore uncertainties following ingestion of U-238 by workers. Two scenarios were investigated: Case I, using the latest HATM model and the systemic uranium model from ICRP publications 100 and 137 respectively, and Case II, based on data from literature, using older ICRP models. In both cases, intake of U-238 in its soluble form (Type F) was modeled and the results were validated using the ICRP OIR dataviewer software. Following validation, uncertainty and sensitivity analyses were performed. In Case I, uncertainties were assigned to the particle transport parameters in both the systemic model model and HATM as well as the uranium uptake fraction into the blood. While in Case II they were limited to the uptake fraction (f1) and the systemic uranium model. A Monte Carlo simulation of 60,000 runs was performed for both cases, sampling model parameters from their respective probability distributions to generate dose distributions. The influence of each parameter on these distributions was also analyzed. Probability distributions were inferred to the calculated dose values using the maximum likelihood estimation method and the Kolmogorov-Smirnov goodness-of-fit. The results showed that for both cases the committed effective dose coefficient, e (50), followed a lognormal distribution. Case I had a geometric mean of 3.2E-08 Sv/Bq and GSD of 2.0, while case II had a slightly lower geometric mean of 3.1E-08 Sv/Bq and GSD of 1.9. Sensitivity analysis showed that the main contributor to dose uncertainty was the fraction of uranium absorbed from the small intestine into the blood. Both cases showed similar trends, with slightly higher results in case I. Overall, this study demonstrates the effectiveness of INTDOSKIT in calculating dose coefficients and analyzing uncertainties. It suggests that while the ICRP reference values remain useful for protection, the incorporation of additional statistical measures and distribution characteristics could further enhance radiation protection strategies.
Read full abstract