Using the Bayesian Approach for the Case of Acute Inhalation of Pu-239 Industrial Compounds

Introduction: The Bayesian approach has found wide application for the tasks of estimating doses of internal exposure under various intake scenarios. The South Ural Institute of Biophysics has accumulated considerable experience in using the Bayesian approach to estimate the expected effective doses of internal exposure for current individual dosimetry control when radionuclides intake into body by various ways. Attributing the type of compounds Moderate or Slow according to the NRB-99/2009 classification for acute inhalation of industrial Pu-239 compounds bias dose estimates to the lungs, which leads to the need to develop a new methodology. Purpose: The development of a calculation methodology using the Bayesian approach for the case of acute inhalation of industrial Pu-239 compounds into the human body and its testing in artificial cases. Material and methods: A technique is presented for interpreting a series of measurement results of Pu-239 activity in daily urine and/or daily feces to assess the distribution of intake, two key parameters of the biokinetic model of ICRP Publication 66 (the proportion of rapid absorption fr and the rate of slow absorption into the blood ss), as well as annual weighted equivalent doses to the lungs. The technique allows using a prior information about the estimated parameters and correctly processing measurement results below the detection limit. Results: A jDose program has been created that implements the technique in ~ 20 minutes on a modern office computer. The program was tested on artificial cases with 10 measurements of Pu-239 activity in daily urine and 10 in daily feces during the first 10 days after acute inhalation intake at AMAD = 1 µm. Testing showed reproducibility of the set «true» parameter values in the range (average ± 2 standard deviations) with an increase in the proportion of unreliable measurement results. The increase in the proportion of unreliable measurement results had the greatest impact on the estimation of the coefficient of variation of the slow absorption rate parameter into the blood ss.