Study On The Influencing Factors Of Low-energy Electronic Microdosimetry Parameters

Microdosimetry studies the biological effects of ionizing radiation at the cellular level,especially at low doses or at very small volumes of concern,a quantitative description of the randomness of energy deposition in a radiating medium is provided.Microdosimetry has been increasingly used in the evaluation of radiation quality for low-dose radiation protection and in the study of radiation effects at DNA level.The Monte Carlo track structure code is currently the main method for accurately obtaining nano-level microdosimetry parameters.The track structure of radiation particles in the medium is obtained through Monte Carlo simulation.And then the microdosimetry distribution of energy deposition of particles in the medium in the concerned tiny volume is calculated.However,due to the differences in cross-sectional databases used by different software and different packages of electromagnetic physical models for the same software,there is no universally accepted value for the results of microdosimetry parameters so far.As a result,it is not easy to verify the accuracy of the results after calculating the microdosimetric parameters.How to ensure the accuracy of the calculation results is a problem to be solved.In other words,it is necessary to explore the factors that influence the results of microdosimetry parameters.The tracks of monoenergetic electron(0.1keV-10keV)in infinite liquid water was simulated using the Geant4-DNA extension included in Geant4 toolkit version 10.4(patch 01).And then in Matlab platform by self-programming,through three different sample method,uniform random sampling,weighted random sampling and point weighted random sampling to obtain the microdosimetric distributions.At the same time,the differences of the three methods were compared.Finally,the dose mean lineal energy results are compared with the published data using KURBUC and Geant4-DNA(version 10.2).The effects of different software versions and different physical models on the results of microdosimetric parameters between different softwares are given.The main results are as follows:(1)In this paper,three common random sampling methods are compared in detail in the calculation of microdosimetry parameters,and it is found that the weighted random sampling method has good stability and the highest efficiency under the same conditions.The results of weighted sampling are consistent with those of weighted random sampling,but the computational efficiency is not as good as the former.Uniform random sampling results are generally higher than the other two methods,and the lowest efficiency.This work can provide reference for other staff to select the sampling method when carrying out relevant calculation.(2)In this paper,the effects of different sample sizes on the calculation results are compared in detail.With the increase of electron energy,the accuracy of the results can also be guaranteed by selecting fewer tracks.The total number of samples required to achieve the standard deviation of the calculated results decreases with the increase of site diameter.When the site diameter remains unchanged,the number of samples required increases as the electron energy increases.This work can provide a reference for the selection of sample size when similar calculations are carried out in the future.(3)The dose mean lineal energy results are compared with the published data using KURBUC,Geant4-DNA(10.2)(opt2)and Geant4-DNA(10.2)(opt5).The effects of different software versions and different physical models on the results of microdosimetric parameters between different softwares are given.