Abstract：
In the present work, a new Monte Carlo Geant4 based code called InterDosi 1.0, was used to simulate specific absorbed fractions (SAFs) in the six reference pediatric voxel-based phantoms developed by the International Commission on Radiological Protection (ICRP). The aim of this study was to assess the ability of this code to estimate SAFs in a variety of voxel-based phantoms. A large number of photon SAFs were calculated for pairs of organs corresponding to three sources and 170 target organs/regions. A total of 108 initial photons were uniformly emitted from the source organs with eight discrete energies. In order to speed up the calculation of SAFs, Monte Carlo multithreaded simulations were started on a workstation with 12 threads, and a Geant4 tracking optimization technique was applied that consists in skipping the voxel boundaries when two adjacent voxels share the same material, which seems to reduce the simulation time by an average of approximately 36%. The results showed good agreement with the reference data produced through the MCNP 2.7 code, with average and maximum absolute discrepancies of 0.5% and 7.68%, respectively. We concluded that these results confirm the feasibility of InterDosi code to perform photon internal dosimetry calculations at a voxel level.
摘要：
在目前的工作中,一种名为InterDosi1.0的基于MonteCarloGeant4的新代码用于模拟由国际放射防护委员会(ICRP)开发的6种参考儿科基于体素的体模中的特定吸收分数(SAF).这项研究的目的是评估该代码在各种基于体素的体模中估计SAF的能力。针对对应于三个源和170个目标器官/区域的成对器官计算了大量的光子SAF。源器官以八个离散能量均匀地发射了总共108个初始光子。为了加快SAF的计算，MonteCarlo多线程仿真是在12个线程的工作站上开始的，并且应用了Geant4跟踪优化技术，该技术包括在两个相邻体素共享相同材料时跳过体素边界，这似乎将模拟时间平均减少了约36%。结果显示与通过MCNP2.7代码产生的参考数据具有良好的一致性，平均和最大绝对差异为0.5%和7.68%，分别。我们得出的结论是，这些结果证实了InterDosi代码在体素水平上执行光子内部剂量测定计算的可行性。
