Introduction: Biokinetic and dosimetry studies in small animals often precede clinical radionuclide therapies. As in human studies, a reliable evaluation of therapeutic efficacy is essential and must be based on accurate dosimetry, which must be based on a realistic dosimetry model. The aim of this study was to evaluate the differences in the results when using a more anatomic realistic mouse phantom, as compared to previously mathematically described phantoms, based mainly on ellipsoids and cylinders. The difference in results from the two Monte Carlo codes, EGS4 and MCNPX 2.6a, was also evaluated. Methods: An anatomical correct mouse phantom (Moby) was developed by Segars et al. for the evaluation and optimization of the in vivo imaging of mice. The Moby phantom is based on surfaces, which allows for an easy and flexible definition of organ sizes. It includes respiratory movements and a beating heart. It also allows for a redefinition of the location of several internal organs. The execution of the Moby program generates a three-dimensional voxel-based phantom of a specified size, which was modified and used as input for Monte Carlo simulations of absorbed fractions and S-factors. The radiation transport was simulated both with the EGS4 system and the MCNPX 2.6a code. Calculations were done for the radionuclides 18F, 124I, 131I, 111In, 177Lu, and 90Y. S-factors were calculated using in-house-developed IDL programs and compared with results from previously published models. Results: The comparison of S-factors obtained by the Moby model and mathematical phantoms showed that these, in many cases, were within the same range, whereas for some organs, they were underestimated in the mathematical phantoms. The results were closer to the more anatomically realistic phantom than to the mathematical phantoms, with some exceptions. When investing differences between MCNPX 2.6a and EGS4 using the Moby phantom, results indicated some differences in absorbed fractions for electrons. This reason may be owing to differences in the codes regarding the theory for which electron transport are simulated. Conclusions: It is possible to calculate S-factors that are specific for small animals, such as mice. The Moby phantom is useful as a dosimetry model because it is anatomically realistic, but still very flexible, with 35 accurately segmented regions.
