Objective To study the differences between Acuros XB ( AXB) and anisotropic analytical algorithm ( AAA) in the verifi-cation of volumetric modulated arc therapy ( VMAT) plans for nasopharyngeal carcinoma .Methods Nineteen nasopharyngeal carcinoma trea-ted with VMAT plans were transplanted to a Delta 4 phantom and I′mRT head phantom to generate verification plans and were then calculated by AXB and AAA,respectively,utilizing distributed calculation framework (DCF),with the calculation time being recorded .The verification plans were performed by a Varian TrueBeam linear accelerator ,and were measured by the Delta4 and ionization chamber,respectively.The 3D Gamma index and point dose deviation were compared for AXB and AAA .Results In terms of Gamma Index ,no significant difference was found between the two algorithms .According to 3%/3mm criterion,the Gamma indices were 99.8%±0.2%and 99.9%±0.2%for AXB and AAA,respectively.The difference was -0.1%±0.3%(95%confidence interval,-0.3%~0.1%,P=0.201).According to 2%/2 mm cri-terion,the Gamma indices were 95.7%±1.8%and 95.5%±2.3%for AXB and AAA,respectively.The difference was(0.2±1.7)%(95%confidence interval,-0.6%~1.0%,P=0.611).In terms of the point dose,AXB was slightly closer to the measured dose than AAA:the de-viation from AXB to measured dose was (-0.2±0.6)%while the deviation for AAA was -0.5%±0.6%,and the difference between them was 0.3%±0.6%(95% confidence interval,0.0%~0.6%,P=0.042).Moreover,AXB calculation consumed comparable time (P=0. 078)when the DCF was not busy,and significantly less time by(9.5±9.9)min(P=0.029)when the DCF was busy.Conclusions The difference between AXB and AAA is small ,while AXB is slightly closer to measured dose ,when homogeneous phantoms are used for verifica-tion of VMAT plans.Moreover,AXB has higher calculation efficiency than AAA .In conclusion,AXB algorithm might be recommended for rou-tine calculation of verification plans for nasopharyngeal carcinoma .%目的：探讨Acuros XB（ AXB）算法与各向异性分析算法（ AAA）在鼻咽癌容积调强放疗（ VMAT）计划验证中的差异。方法选取19例鼻咽癌VMAT计划，移植到Delta4模体和I′mRT头部模体上制作验证计划。使用 AXB、AAA 两种算法分别计算剂量分布，利用分布式计算框架（ DCF）加快计算速度，同时记录计算耗时。在 Varian TueBeam直线加速器实施验证计划，使用 Delta4和电离室行实际剂量测量，比较两种算法所得剂量的三维Gamma通过率和点剂量误差。结果使用 Delta4模体验证，两种算法Gamma通过率无明显差异：按3 mm／3％标准，AXB 的通过率为（99．8±0．2）％，AAA的通过率为（99．9±0．2）％，两者差值为（－0．1±0．3）％，95％置信区间（－0．3％，0．1％），P＝0．201；按2 mm／2％标准，AXB的通过率为（95．7±1．8）％，AAA的通过率为（95．5±2．3）％，两者差值为（0．2±1．7）％，95％置信区间（－0．6％，1．0％），P＝0．611。点剂量误差方面，AXB比AAA更接近实测剂量：AXB剂量与实测剂量偏差为（－0．2±0．6）％，AAA剂量与实测剂量偏差为（－0．5±0．6）％，两者差值为（0．3±0．6）％，95％置信区间（0．0％，0．6％），P＝0．042。计算耗时方面，DCF空闲时两者统计学差异不显著（P＝0．078），DCF繁忙时AXB比 AAA节省（9．5±9．9）min， P＝0．029。结论鼻咽癌VMAT计划使用均匀模体验证时，AXB与AAA算法的差异很小，其中AXB略比 AAA接近实测剂量，而 AXB比 AAA计算效率高。综合计算精度与效率，推荐使用AXB算法常规计算验证计划。
展开▼
