Shang Xuying, Liu Yaoying, Le Xiaoyun, Wang Zishen, Sun Xiaoqing, Fang Chunfeng, Qu Baolin, Zou Yue, Zhao Wei, Zhang Gaolong, Xu Shouping
School of Physics, Beihang University, Beijing 102206, People's Republic of China; Department of Radiation Oncology, PLA General Hospital, Beijing 100853, People's Republic of China; Hebei Yizhou Tumor Hospital, Hebei, Zhuozhou 072750, People's Republic of China; National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People's Republic of China.
School of Physics, Beihang University, Beijing 102206, People's Republic of China; Department of Radiation Oncology, PLA General Hospital, Beijing 100853, People's Republic of China; National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, People's Republic of China.
Phys Med. 2025 Sep;137:105074. doi: 10.1016/j.ejmp.2025.105074. Epub 2025 Aug 14.
We endeavor to present a comprehensive methodology for establishing a versatile Monte Carlo (MC) dose calculation platform that operates independently of the treatment planning system (TPS, e.g., RayStation).
We assumed the source emission plane was at the nozzle exit. To align with the measured data, we optimized the phase space parameters (spot size, angular spread, correlation, energy spread, and nominal energy) and the beam source model's absolute dose (the number of protons per MU) in the source emission plane. Additionally, we devised a method that automatically converts patient plans into executable MC scripts capable of running in the TOPAS MC software.
Our efforts successfully established a library encompassing the source model parameters. The disparities between the measured and simulated beam spot sizes were all below 0.3 mm. Moreover, the differences in the depth-dose curve's distal falloff (R) were less than 0.1 mm, and the mean point-to-point dose differences were less than 0.7 %. Remarkably, the 3D gamma passing rates (GPRs) for three spread-out Bragg peaks (SOBPs, 3 mm/3% criteria) were 100 %. Furthermore, when comparing TOPAS and TPS MCs for 23 pencil beam scanning (PBS) patient plans, the mean 3D GPRs for 2 mm/2% and 3 mm/3% criteria were 99.96 % and 100 %, respectively.
We have successfully developed a comprehensive MC framework for PBS, employing a well-defined beam source model. The method we presented for building the PBS MC framework holds potential to build a dose verification tool and for scientific research.
我们致力于提出一种全面的方法，以建立一个独立于治疗计划系统（TPS，例如RayStation）运行的通用蒙特卡罗（MC）剂量计算平台。
我们假设源发射平面位于喷嘴出口处。为了与测量数据对齐，我们在源发射平面中优化了相空间参数（光斑尺寸、角展度、相关性、能量展度和标称能量）以及束源模型的绝对剂量（每MU的质子数）。此外，我们设计了一种方法，可自动将患者计划转换为能够在TOPAS MC软件中运行的可执行MC脚本。
我们的努力成功建立了一个包含源模型参数的库。测量和模拟的束光斑尺寸之间的差异均低于0.3毫米。此外，深度剂量曲线远端下降（R）的差异小于0.1毫米，平均点对点剂量差异小于0.7%。值得注意的是，三个扩展布拉格峰（SOBP，3毫米/3%标准）的3D伽马通过率（GPR）为100%。此外，在比较TOPAS和TPS MC对23个笔形束扫描（PBS）患者计划时，2毫米/2%和3毫米/3%标准的平均3D GPR分别为99.96%和100%。
我们成功开发了一个用于PBS的综合MC框架，采用了定义明确的束源模型。我们提出的构建PBS MC框架的方法具有构建剂量验证工具和用于科学研究的潜力。
