Meitong Wei, Yuxin Wang, Zirui Ye, Hui Zhang, Xie George Xu
{"title":"Evaluation of the KERMA approximation method in Monte Carlo dose calculations for BNCT involving a realistic 3D patient phantom.","authors":"Meitong Wei, Yuxin Wang, Zirui Ye, Hui Zhang, Xie George Xu","doi":"10.1093/rpd/ncag111","DOIUrl":null,"url":null,"abstract":"<p><p>Monte Carlo (MC) methods are generally recognized as the gold standard for dose calculations. However, achieving high statistical confidence is computationally expensive, particularly when complex secondary particles are involved. Kinetic Energy Released in Matter (KERMA) approximation method is a variance reduction technique widely used in MC simulations. This study evaluated the accuracy and efficiency of the KERMA approximation for Boron Neutron Capture Therapy (BNCT) dose calculations using the TOol for PArticle Simulation MC code. Simulations were performed on a CT phantom of a glioblastoma patient using an accelerator-based BNCT source. It was found that neutron KERMA results agreed with Analog MC results within 2%, while providing a 157-fold computational speed-up. However, the photon KERMA approximation overestimated skin photon dose by ~12%. The results suggest that the KERMA approximation for BNCT offers clinically acceptable accuracy and efficiency, while using Analog MC method for skin photon dose evaluation can further improve overall accuracy.</p>","PeriodicalId":20795,"journal":{"name":"Radiation protection dosimetry","volume":" ","pages":""},"PeriodicalIF":0.8000,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Radiation protection dosimetry","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1093/rpd/ncag111","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Monte Carlo (MC) methods are generally recognized as the gold standard for dose calculations. However, achieving high statistical confidence is computationally expensive, particularly when complex secondary particles are involved. Kinetic Energy Released in Matter (KERMA) approximation method is a variance reduction technique widely used in MC simulations. This study evaluated the accuracy and efficiency of the KERMA approximation for Boron Neutron Capture Therapy (BNCT) dose calculations using the TOol for PArticle Simulation MC code. Simulations were performed on a CT phantom of a glioblastoma patient using an accelerator-based BNCT source. It was found that neutron KERMA results agreed with Analog MC results within 2%, while providing a 157-fold computational speed-up. However, the photon KERMA approximation overestimated skin photon dose by ~12%. The results suggest that the KERMA approximation for BNCT offers clinically acceptable accuracy and efficiency, while using Analog MC method for skin photon dose evaluation can further improve overall accuracy.
期刊介绍:
Radiation Protection Dosimetry covers all aspects of personal and environmental dosimetry and monitoring, for both ionising and non-ionising radiations. This includes biological aspects, physical concepts, biophysical dosimetry, external and internal personal dosimetry and monitoring, environmental and workplace monitoring, accident dosimetry, and dosimetry related to the protection of patients. Particular emphasis is placed on papers covering the fundamentals of dosimetry; units, radiation quantities and conversion factors. Papers covering archaeological dating are included only if the fundamental measurement method or technique, such as thermoluminescence, has direct application to personal dosimetry measurements. Papers covering the dosimetric aspects of radon or other naturally occurring radioactive materials and low level radiation are included. Animal experiments and ecological sample measurements are not included unless there is a significant relevant content reason.