直线加速器日志文件对容积调强三维剂量验证研究

3D dosimetric verification of volumetric-modulated arc therapy using Linac log files

目的：采用直线加速器日志文件进行容积调强三维剂量验证的研究，分析射野参数误差对临床剂量的影响。方法：对佛山市第一人民医院2013-01—2013-12收治的10例直肠癌患者设计容积调强计划。利用自编程序从日志文件中读取实际治疗时的机架角度、多叶准直器（multileaf collimator，MLC）叶片位置以及机器跳数（monitorunit，MU），比较射野参数的误差。使用实际射野参数替代原治疗计划的射野参数，在cT图像上重新进行三维剂量重建。比较重建计划与治疗计划，分析射野参数误差对靶区和危及器官三维剂量分布的影响。结果：实际治疗与治疗计划的机架角偏差〈1。，最大机器跳数的偏差〈0．2MU，叶片位置最大误差〈2mm，大部分误差为0．05～1mm。叶片位置误差对计划靶区（planning target volume，PTV）处方剂量所包含的靶区（V100）影响较大，差异有统计学意义，P=0．006；机架角偏差和机器跳数偏差对PTV（V1。）影响较小，差异无统计学意义，P〉0．05；各个射野参数对小肠V40（40Gy剂量所包含的体积）、膀胱V40和股骨头V30（30Gy剂量所包含的体积）的影响较小，差异无统计学意义，P〉0．05。结论：利用加速器日志文件可以进行容积调强三维剂量验证方面的研究，叶片位置误差对计划靶区的剂量影响较大，机架角和机器跳数误差对计划靶区剂量的影响较小，各个射野参数对危及器官剂量的影响不大。

OBJECTIVE： To verify the 3D dose of volumetric-modulated arc therapy （VMAT） and evaluate the im- pact of delivery parameter errors on clinical dose variation based on Linac log files. METHODS： Ten clinically acceptable RapidArc treatment plans were generated for rectum cancers. Using the program developed inhouse converted the data in the log files to the actual delivery parameters,multileaf collimator （MLC） positions, gantry angle, and monitor unit（MU）. The errors of the delivery parameters were compared respectively. The extracted parameters were used to replace the cor- responding parameters in the original treatment plan file. The 3D dose distribution was reconstructed on the original CT. The effect of the 3D dose distribution difference that caused by the delivery parameters between the reconstructed plan and the the origin plan was analyzed. RESULTS： The gantry angle, MU and MLC leaf position errors for all patients were 〈1°,〈 0.2 MU and 〈2 mm（mostly between 0.05 mm and 1 ram） ,respectively. The impact of MLC leaf position errors on PTV（planning target volume） V100 （the volume involved by prescription dose）was noticeable and the difference had sta- tistical significance （P=0. 006） ;The impact of gantry angle and MU errors on PTV V100 was little and the difference was not significant （P〉0. 05） ;The delivery parameter errors had no obvious dosimetric impact on the accuracy of organs at risk including the small intestine V40 （the volume involved by 40 Gy dose） ,bladder V40 ,femoral head V30 （ the volume in- volved by 30 Gy dose） and the differences were not significant（P〉0.05）. CONCLUSIONS： It was feasible to verify the 3D dose of volumetric-modulated arc therapy （VMAT） using Linac log files. The impact of MLC leaf position errors on the dose of PTV was noticeable but the impact of gantry angle and MU errors on PTV was little;The delivery parameter er- rors had no obvious dosimetric impact on the accuracy of organs at risk.