An improvement in IMRT QA results and beam matching in linacs using statistical process control
An improvement in IMRT QA results and beam matching in linacs using statistical process control
复制标题
DOI：
10.1120/jacmp.v15i5.4927
复制
发表时间：
2014-01-01
影响因子：
2.1
通讯作者：
Ezzell, Gary A.
中科院分区：
文献类型：
--
作者：
Gagneur, Justin D.;Ezzell, Gary A.
本研究的目的是在患者特定调强放射治疗（IMRT）QA背景下应用统计过程控制（SPC）原则，以设定临床特定行动限，并评价多叶准直器（MLC）校准变更对IMRT QA结果的影响。回顾性分析了在三个射束匹配直线加速器上收集的10个月的IMRT QA数据和247名患者QA，重点关注伽马通过率（GPR）和测量剂量与计划剂量之间的平均比率。计算初始控制图和行动限。基于这些数据，对MLC的叶隙参数进行了更改，以提高直线加速器之间的一致性。使用MLC扫描测试每月测试该叶隙参数。使用具有424个独特QA的随访数据集来评估叶隙参数变化的影响。初始数据平均GPR为98.6%，SPC行动限为93.7%。平均剂量比为1.003，行动上限为1.017，行动下限为0.989。直线加速器的扫描测试结果为标称值的-1.8%，0%和+1.2%。调整叶片间隙参数后，所有扫描试验结果均在标称值的0.4%范围内。随后，平均GPR为99.4%，SPC行动限为97.3%。平均剂量比为0.997，行动上限为1.011，行动下限为0.981。将SPC的原理应用于IMRT QA，可以识别和减少紧密匹配的直线加速器之间的微小差异。正在进行的分析将监测该过程，并用于完善IMRT QA的临床行动限。
The purpose of this study is to apply the principles of statistical process control (SPC) in the context of patient specific intensity-modulated radiation therapy (IMRT) QA to set clinic-specific action limits and evaluate the impact of changes to the multileaf collimator (MLC) calibrations on IMRT QA results. Ten months of IMRT QA data with 247 patient QAs collected on three beam-matched linacs were retrospectively analyzed with a focus on the gamma pass rate (GPR) and the average ratio between the measured and planned doses. Initial control charts and action limits were calculated. Based on this data, changes were made to the leaf gap parameter for the MLCs to improve the consistency between linacs. This leaf gap parameter is tested monthly using a MLC sweep test. A follow-up dataset with 424 unique QAs were used to evaluate the impact of the leaf gap parameter change. The initial data average GPR was 98.6% with an SPC action limit of 93.7%. The average ratio of doses was 1.003, with an upper action limit of 1.017 and a lower action limit of 0.989. The sweep test results for the linacs were -1.8%, 0%, and +1.2% from nominal. After the adjustment of the leaf gap parameter, all sweep test results were within 0.4% of nominal. Subsequently, the average GPR was 99.4% with an SPC action limit of 97.3%. The average ratio of doses was 0.997 with an upper action limit of 1.011 and a lower action limit of 0.981. Applying the principles of SPC to IMRT QA allowed small differences between closely matched linacs to be identified and reduced. Ongoing analysis will monitor the process and be used to refine the clinical action limits for IMRT QA.