用蒙特卡罗模拟评估小野条件下肺介质中AAA算法的精度

摘要

目的:用蒙特卡罗模拟评估放射治疗剂量计算使用的各向异性分析算法(Anisotropic Analytical Algorithm,AAA)在小野条件下肺介质中的计算精度.材料与方法:建立一包含肺介质的水模体,分别用AAA算法、笔形束卷积算法(Pencil Beam Convolution,PBC算法)(作为对比)和蒙特卡罗(Monte Carlo,MC)模拟计算2 cm×2 cm到8 cm×8 cm射野条件下该模体中的深度剂量和离轴比,并以MC模拟为标准比较深度剂量,用一维伽马分析对离轴比进行分析.结果:AAA算法在2 cm×2 cm射野肺介质区域高估了深度剂量,其它情况均低估了深度剂量,剂量偏差范围为-0.24％～2.66％,PBC算法在肺介质区域高估了深度剂量,剂量偏差的范围为1.18％～14.55％.AAA算法计算的离轴比和MC模拟,在射野剂量平坦区相对内收,在剂量跌落区向两侧发散,但AAA算法略高估了射野边缘的剂量,一维伽马分析(与MC相比)通过率为100％(3 mm/3％).PBC算法在射野剂量平坦区相对发散,而在剂量跌落区向两侧内收.一维伽马分析通过率范围为51％～88％.结论:在肺介质中,AAA剂量计算的结果与MC模拟的一致性较好,与PBC算法相比,剂量计算的精度较高.%Objective: In order to evaluate the calculation accuracy of Anisotropic Analytical Algorithm (AAA) for the situation with small fields in a water-lung phantom using Monte Carlo simulation as benchmarks. Materials and Methods: A water phantom with a lung slab was built ,in which the depth dose (DD) and off-axis ratio (OAR) for field 2 cmx2 cm to field 8 cmx8 cm were calculated by AAA algorithms, Pencil Beam Convolution (PBC) algorithms (as comparison),and Monte Carlo (MC) simulation. The evaluation of algorithms by MC simulation was achieved by the comparisons of DD and the 1 dimension gamma analysis of OAR curve. Results: It was shown that AAA algorithm overestimated the DD in lung region for field 2 cmx 2 cm and underestimate the DD in lung region for other fields, and the dose differences ranged from -0.24%~2.66%, while PBC algorithm overestimated the DD in lung region , and the dose differences ranged from 1.18%~14.55%. AAA algorithm showed better agreement with MC simulation for the OARs, which spread to both laterals in field steep dose gradient regions and was adducted in shallow dose gradient as well as AAA overestimated dose in field edge. One dimension gamma analysis pass rate was 100%,while that was not predicted accurately by PBC algorithm. Conclusions: In lung medium, the agreement between dose calculated by AAA and MC simulation was better, the calculation accuracy was higher compared with PBC.