目的：评价开放10 x 10 $cm^2$和小视野的TrueBeam光子相空间文件（PSF）的潜在方差（LV）。方法：利用BEAMnrc/DOSXYZnrc将Varian相空间文件（PSFA）中的粒子通过二级准直器进行传输。将输送的颗粒刻划到位于钳口下方的另一个相空间（PSFB）中，或进一步输送通过锥形准直器并在正下方刻划，形成PSFC。在直径为0.13、0.25、0.35、1.0、1.2、1.5和4cm的锥体下，对6MV-FFF、6MV、10MV-FFF、10MV和15MV射束的PFSB进行评分，并对6MV射束的PSFC进行评分。PSFB和PSFC被输送到水模体中，颗粒循环数范围为10至1000。对于10 × 10 $cm^2$射野，使用0.5 × 0.5 × 0.5 cm ^3体素对剂量进行评分，而对于用小锥准直的射束，使用0.1 × 0.1 × 0.5 cm ^3体素对剂量进行评分。对于直径为0.25 cm的小型锥形准直6MV射束，体模体素尺寸变化为0.02x0.02x0.5cm^3、0.05x0.05x0.5cm^3和0.1x0.1x0.5cm^3。根据Sempau等人对所有病例的剂量差异进行评分，并对LV进行评价。结果：对于10 x 10 cm ^2射野，计算的LV在体模表面最大，并随深度降低，直至在5 cm深度达到平台。在10cm深度处，6MV-FFF、6MV、10MV-FFF、10MV和15MV能量的LV分别为0.54%、0.96%、0.35%、0.69%和0.57%。对于6MV相空间准直器，在1.5cm深度处，用直径为0.13、0.25、0.35、1.0cm的锥进行准直，计算的LV分别为75.6%、25.4%、17.6%和8.0%。在0.02x0.02x0.5cm^3、0.05x0.05x0.5cm^3和0.1x0.1x0.5cm^3体素中，0.25 cm锥形准直6MV射束的计算LV分别为61.2%、40.7%和22.5%。结论：单个PSF可用于在开放的10 × 10cm^2场MC模拟中实现亚百分比的潜在方差，而对于小SRS场，必须对更多的PSF求和。
英文摘要：
Purpose: To evaluate the latent variance (LV) of TrueBeam photon phase-space files (PSF) for open 10x10$cm^2$ and small fields.
Methods: BEAMnrc/DOSXYZnrc was used to transport particles from Varian phase-space files (PSFA) through the secondary collimators. Transported particles were scored into another phase-space located under the jaws (PSFB), or transported further through the cone collimators and scored straight below, forming PSFC. PFSB were scored for 6MV-FFF, 6MV, 10MV-FFF, 10MV and 15MV beams with 10x10$cm^2$ field size, and PSFC were scored for 6MV beam under cones of 0.13, 0.25, 0.35, 1.0, 1.2, 1.5 and 4cm diameter. PSFB and PSFC were transported into a water phantom with particle recycling number ranging from 10 to 1000. For 10x10$cm^2$ fields 0.5x0.5x0.5cm^3 voxels were used to score the dose, whereas the dose was scored in 0.1x0.1x0.5cm^3 voxels for beams collimated with small cones. For small 0.25cm diameter cone-collimated 6MV beam, phantom voxel size varied as 0.02x0.02x0.5cm^3, 0.05x0.05x0.5cm^3 and 0.1x0.1x0.5cm^3. Dose variances were scored in all cases and LV evaluated as per Sempau et al.
Results: For the 10x10cm^2 fields calculated LVs were greatest at the phantom surface and decreased with depth until reached a plateau at 5cm depth. LVs were 0.54%, 0.96%, 0.35%, 0.69% and 0.57% for the 6MV-FFF, 6MV, 10MV-FFF, 10MV and 15MV energies, respectively at 10cm depth. For the 6 MV phase-space collimated with cones of 0.13, 0.25, 0.35, 1.0cm diameter, the LVs calculated at 1.5cm depth were 75.6%, 25.4%, 17.6% and 8.0% respectively. Calculated LV for the 0.25cm cone-collimated 6MV beam were 61.2%, 40.7%, 22.5% in 0.02x0.02x0.5cm^3, 0.05x0.05x0.5cm^3 and 0.1x0.1x0.5cm^3 voxels respectively.
Conclusion: Single PSF can be used to achieve sub-percent latent variance in open 10x10cm^2 field MC simulations, whereas more PSFs would have to be summed for small SRS fields.
Closing the gap: A two-site MLC matching study for Varian TrueBeam Linear Accelerators
Royal Adelaide Hospital（皇家阿德莱德医院）；Adelaide University, School of Physical Sciences（阿得雷德大学物理科学学院）；Calvary Mater Newcastle（卡尔瓦里马特纽卡斯尔）；University of Newcastle（纽卡斯尔大学）；University of Sydney（悉尼大学）
An Analytic Linear Accelerator Source Model for Monte Carlo dose calculations. II. Model Utilization in a GPU-based Monte Carlo Package and Automatic Source Commissioning
Characteristics of a 9 MeV electron beam for total skin electron radiotherapy evaluated in comparison to a 6 MeV electron beam
Cancer Care Services, Royal Brisbane and Women’s Hospital（皇家布里斯班和妇女医院癌症护理服务）；School of Chemistry and Physics, Queensland University of Technology（昆士兰科技大学化学与物理学院）；School of Information Technology and Electrical Engineering, University of Queensland（昆士兰大学信息技术与电气工程学院）
Non-invasive Reversible Software-based Electron FLASH Irradiation Configuration of a Linear Accelerator in Clinical Use
Stanford University School of Medicine（斯坦福大学医学院）；University of California Irvine（加州大学尔湾分校）；Princess Margaret Cancer Centre, University Health Network（公主玛格丽特癌症中心，联合健康网络）
Development of a Full Monte Carlo Therapeutic Dose Calculation Toolkit for Halcyon Using Geant4
目的：在Geant4（10.7版）中开发一个用于最近发布的环形门架直线加速器的蒙特卡罗（MC）治疗剂量计算工具包，用于放射治疗计划的二次剂量验证。研究方法：对于Halcyon（Varian Medical Systems），对DSMLC进行建模，并使用Geant 4模拟DSMLC和患者体模中的辐射传输。辐射源从相空间文件中采样，用于DSMLC上方的直线加速器头。使用Varian提供的基于云的Monte Carlo（MC）模拟器VirtuaLinac（VL）获得相空间文件。不同方野宽度（2x2、4x4、6x6、8x8、10x10、20x20和28x28 cm 2）的剂量曲线，即，百分比深度剂量（PDD）曲线和横向剖面进行了模拟，并与实验剖面进行了比较。还使用开发的工具包计算了两个解剖部位（前列腺和脑）的IMRT（调强放射治疗）计划，并与TPS计算的剂量进行了比较（Acuros，Eclipse 15.6）。使用3D剂量差和3D伽马分析来评价与TPS计算剂量相比的模拟准确度。结果如下：对于所有模拟射野尺寸，水模中模拟的侧向剂量分布和PDD曲线与测量值匹配良好，相对差异为± 2%。对于前列腺和脑调强放射治疗计划，模拟剂量与TPS计算剂量显示出良好的一致性。两个前列腺和脑计划的3D伽马通过率（3%/3mm）分别为98.08%和95.4%。结论：开发的Halcyon全MC剂量计算工具包在水模体…
MAAS-SFRThelper: An Integrated ESAPI Plugin for Structure Generation, Optimization, and Evaluation of Spatially Fractionated Radiation Therapy
Gateway Scripts；Varian Medical Systems（瓦里安医疗系统）；Department of Radiation Oncology, Intermountain Health（Intermountain Health 放射肿瘤科）；Department of Radiation Oncology, Icahn School of Medicine at Mt. Sinai（西奈山伊坎医学院放射肿瘤科）；Biological Sciences Division, University of Chicago（芝加哥大学生物科学部）；Department of Radiation Oncology, University of Freiburg（弗赖堡大学放射肿瘤科）；Department of Biomedical Engineering, University of Southern California（南加州大学生物医学工程系）；Radiation Therapy Program, BC Cancer（BC Cancer 放疗项目）；Clinic Hirslanden（Hirslanden诊所）；Foshan Fosun Chancheng Hospital（佛山佛山水磨石医院）；School of Medicine, Washington University in St. Louis（华盛顿大学圣路易斯分校医学院）
Beam data modeling of linear accelerators (linacs) through machine learning and its potential applications in fast and robust linac commissioning and quality assurance
背景和目的：提出一种基于机器学习的新型直线加速器束流数据建模方法，用于直线加速器调试和质量保证过程。材料与方法：我们假设束流数据是固有的直线加速器功能和百分比深度剂量（PDDs）的函数，不同的字段大小的配置文件是相互关联的。使用机器学习框架将相关性公式化为多变量回归问题。使用从多家机构获得的Varian TrueBeam射束数据集（n=43）评价框架。数据集包括不同能量和场大小的PDD和分布。使用10 × 10 cm $^2$射野作为输入，训练多元回归模型以预测射束特定PDD和不同射野尺寸的分布。结果如下：在所研究的不同射束能量中，PDD的预测值的平均绝对百分比相对误差（%RE）为0.19-0.35%。最大平均绝对%RE为0.93%。对于曲线预测，平均绝对%RE为0.66-0.93%，最大绝对%RE为3.76%。最大的不确定性在PDD和配置文件的预测被发现在集结区和现场半影，分别。随着训练集数量的增加，预测精度增加到大约20个训练集。结论：通过这种新的基于机器学习的方法，我们已经证明了用于常规放射治疗的直线加速器调试的射束数据的准确和可再现的生成。该方法具有简化直线加速器调试程序、节省时间和人力的潜力，同时提高了调试过程的准确性。
PyDoseRT Proton: A GPU Pencil-Beam Engine with a Convolutional Residual-Correction Network for Fast Proton Dose Calculation
Medical University of Vienna（维也纳医科大学）；Christian Doppler Laboratory for Image and Knowledge Driven Precision Radiation Oncology（克里斯蒂安·多普勒图像与知识驱动精准放射肿瘤学实验室）；Umeå University（于默奥大学）
