Purpose. This study aims to evaluate the performance of a commercial 1.5 T MR-Linac by analyzing its patient-specific quality assurance (QA) data collected during one full year of clinical operation. Methods and Materials. The patient-specific QA system consisted of offline delivery QA (DQA) and online calculation-based QA. Offline DQA was based on ArcCHECK-MR combined with an ionization chamber. Online QA was performed using RadCalc that calculated and compared the point dose calculation with the treatment planning system (TPS). A total of 24 patients with 189 treatment fractions were enrolled in this study. Gamma analysis was performed and the threshold that encompassed 95% of QA results (T95) was reported. The plan complexity metric was calculated for each plan and compared with the dose measurements to determine whether any correlation existed. Results. All point dose measurements were within 5% deviation. The mean gamma passing rates of the group data were found to be 96.84.0% and 99.60.7% with criteria of 2%/2mm and 3%/3mm, respectively. T95 of 87.4% and 98.2% was reported for the overall group with the two passing criteria, respectively. No statistically significant difference was found between adaptive treatments with adapt-to-position (ATP) and adapt-to-shape (ATS), whilst the category of pelvis data showed a better passing rate than other sites. Online QA gave a mean deviation of 0.22.2%. The plan complexity metric was positively correlated with the mean dose difference whilst the complexity of the ATS cohort had larger variations than the ATP cohort. Conclusions. A patient-specific QA system based on ArcCHECK-MR, solid phantom and ionization chamber has been well established and implemented for validation of treatment delivery of a 1.5 T MR-Linac. Our QA data obtained over one year confirms that good agreement between TPS calculation and treatment delivery was achieved.

目的。本研究旨在通过分析在一整年的临床操作期间收集的患者特定质量保证 (QA) 数据来评估商用 1.5 T MR-Linac 的性能。方法和材料. 患者特异性 QA 系统由离线交付 QA (DQA) 和基于在线计算的 QA 组成。离线 DQA 基于结合电离室的 ArcCHECK-MR。使用 RadCalc 进行在线 QA，计算点剂量计算并将其与治疗计划系统 (TPS) 进行比较。本研究共纳入 24 名患者，共 189 次治疗。进行了 Gamma 分析，并报告了包含 95% 的 QA 结果 (T95) 的阈值。计算每个计划的计划复杂性度量并与剂量测量值进行比较以确定是否存在任何相关性。结果. 所有点剂量测量的偏差均在 5% 以内。发现组数据的平均伽马通过率分别为 96.84.0% 和 99.60.7%，标准分别为 2%/2mm 和 3%/3mm。具有两个通过标准的整个组的 T95 分别为 87.4% 和 98.2%。适应位置（ATP）和适应形状（ATS）的适应性治疗之间没有发现统计学上的显着差异，而骨盆数据类别显示出比其他部位更好的通过率。在线 QA 的平均偏差为 0.22.2%。计划复杂性指标与平均剂量差异呈正相关，而 ATS 队列的复杂性比 ATP 队列的变化更大。结论. 基于 ArcCHECK-MR、实体模型和电离室的患者特定 QA 系统已经很好地建立并实施，用于验证 1.5 T MR-Linac 的治疗交付。我们一年多获得的 QA 数据证实，TPS 计算和治疗实施之间取得了良好的一致性。