蒙特卡罗方法
剂量学
核素
计算机科学
统计物理学
一致性(知识库)
趋同(经济学)
物理
不确定性传播
吸收剂量
数学优化
参数统计
计算物理学
采样(信号处理)
算法
应用数学
辐射传输
能量(信号处理)
重要性抽样
辐射
概率分布
辐射测量
数学
作者
Jiawei Zhu,Cao S,Xiaoyu Li,Yaping Qi,Zhi Chen
摘要
Abstract Background Dosimetric studies of targeted alpha therapy (TAT) show that microscopic heterogeneity in geometry and source distribution strongly affects dose‐effect relationships, underscoring the potential of microdosimetry. While Monte Carlo (MC) simulation is the gold standard for dose assessment, its application to TAT microdosimetry remains methodologically challenging and computationally inefficient. This necessitates a systematic and implementable methodology for microdosimetric simulation and analysis. Purpose This study aimed to establish a quantitative MC simulation framework for TAT microdosimetry, in which heterogeneous conditions are analytically processed and Type A standard uncertainty of derived microdosimetric quantities can be estimated. Methods (1) Elementary‐source events were defined under arbitrary microscopic conditions, and the raw moments of single‐event microdosimetric quantities scored via MC simulation were identified for each event. The heterogeneous microdosimetric environments were modeled as weighted mixtures of elementary events. (2) Analytical error propagation was derived to explicitly quantify the standard error (SE) of mixture quantities. This enables, for the first time, a trace‐back mechanism to identify error‐dominating source components for optimized resource allocation. (3) A long‐range radiation correction for ‐emitting nuclides was developed to bridge microscopic energy deposition with macroscopic absorbed dose, enabling the acquisition of reliable results under affordable computational resources. Results A Geant4‐based application implemented the framework in heterogeneous microscopic models. The validation against repeated direct sampling demonstrated consistency between the mixture framework and conventional repeated direct simulation. Subsequently, the MC SE of single‐event dose‐mean specific energy was adopted as the convergence criterion for simulation. Long‐range radiation effects were effectively compensated, relaxing spatial‐extent requirements for TAT MC geometries. The source‐mixture scheme allowed reuse of elementary microdosimetric data, supporting efficient factor analysis for diverse heterogeneous conditions. Coupled with cell‐line parameters and the non‐Poisson‐corrected microdosimetric kinetic model, the analytic results yielded dose‐response characteristics consistent with prior studies of TAT. Conclusions The framework enables trace‐back variance analysis and microdosimetry‐compatible variance reduction. This enables precision‐controlled MC microdosimetry in TAT, and supports flexible analysis of heterogeneity factors and reliable microdosimetric assessment.
科研通智能强力驱动
Strongly Powered by AbleSci AI