Department of Radiation Oncology, Saint Francis Hospital and Medical Center, Hartford, CT 06105, USA.
Int J Radiat Oncol Biol Phys. 2010 May 1;77(1):277-84. doi: 10.1016/j.ijrobp.2009.08.001. Epub 2009 Dec 11.
To compare dose distributions calculated using the Monte Carlo algorithm (MC) and Ray-Trace algorithm (effective path length method, EPL) for CyberKnife treatments of lung tumors.
An acceptable treatment plan is created using Multiplan 2.1 and MC dose calculation. Dose is prescribed to the isodose line encompassing 95% of the planning target volume (PTV) and this is the plan clinically delivered. For comparison, the Ray-Trace algorithm with heterogeneity correction (EPL) is used to recalculate the dose distribution for this plan using the same beams, beam directions, and monitor units (MUs).
The maximum doses calculated by the EPL to target PTV are uniformly larger than the MC plans by up to a factor of 1.63. Up to a factor of four larger maximum dose differences are observed for the critical structures in the chest. More beams traversing larger distances through low density lung are associated with larger differences, consistent with the fact that the EPL overestimates doses in low-density structures and this effect is more pronounced as collimator size decreases.
We establish that changing the treatment plan calculation algorithm from EPL to MC can produce large differences in target and critical organs' dose coverage. The observed discrepancies are larger for plans using smaller collimator sizes and have strong dependency on the anatomical relationship of target-critical structures.
比较使用蒙特卡罗算法（MC）和射线追踪算法（有效路径长度法，EPL）计算 CyberKnife 治疗肺部肿瘤的剂量分布。
使用 Multiplan 2.1 和 MC 剂量计算创建可接受的治疗计划。将剂量规定为包含 95%计划靶体积（PTV）的等剂量线，这是临床交付的计划。为了进行比较，使用具有异质性校正的 Ray-Trace 算法（EPL），使用相同的射束、射束方向和监测单位（MU）重新计算该计划的剂量分布。
EPL 计算的靶 PTV 的最大剂量均匀地比 MC 计划大 1.63 倍。在胸部的关键结构中，最大剂量差异最大可达 4 倍。穿过低密度肺的更多射束穿过更大的距离与更大的剂量差异相关，这与 EPL 高估低密度结构中的剂量的事实一致，并且这种效应随着准直器尺寸的减小而更加明显。
我们确定，从 EPL 到 MC 改变治疗计划计算算法会导致靶区和关键器官的剂量覆盖产生很大差异。对于使用较小准直器尺寸的计划，观察到的差异更大，并且与靶-关键结构的解剖关系密切相关。
