Effect of variability of human bone morphometric parameters on the uncertainty of internal bone marrow doses due to <sup>90</sup>Sr

Abstract

Computational phantoms are used to calculate the doses of internal exposure of active bone marrow. The computational phantoms of ICRP were created for a reference man with anatomical characteristics typical of an average individual. The doses calculated with such phantoms correspond to population-average values. Individual variability introduces a stochastic component of uncertainty into the dose estimation. The objective of this study is to assess the effect of individual variability of bone structure dimensions on the results of dosimetric modeling. The phantoms are represented by simple geometry figures filled with trabecular structures and bone marrow (spongiosa), covered externally with a cortical layer. The models of bone geometry are described by parameters characterizing the linear dimensions, the microarchitecture of the spongiosa (trabecular thickness, trabecular separation, bone volume fraction), as well as the cortical layer thickness. By varying these parameters, sets of phantoms were generated to simulate the individual variability of bone geometry. The mean absorbed dose rate in active bone marrow from a single decay of 90Sr/90Y was calculated assuming isotope distribution either in the volume of the trabecular or cortical bone. All estimates are on the example of the phantom of an adult male skeleton. The individual variability of the main parameters of segment computational phantoms depends on size and equal to: a) for linear dimensions – 12-15%; b) for bone volume fraction – 22-24%; c) for cortical thickness – 21-23%. This leads to uncertainties of dose rate estimation equal to 21% – 25%.