Department of Physics and Astronomy, Louisiana State University and Agricultural and Mechanical College, 202 Tower Drive, Baton Rouge, LA 70803-4001, USA.
Department of Physics and Astronomy, Louisiana State University and Agricultural and Mechanical College, 202 Tower Drive, Baton Rouge, LA 70803-4001, USA; Department of Physics, Mary Bird Perkins Cancer Center, 4950 Essen Lane, Baton Rouge, LA 70809, USA.
Phys Med. 2021 Jul;87:136-143. doi: 10.1016/j.ejmp.2021.03.017. Epub 2021 Mar 26.
Previous literature has shown general trade-offs between plan complexity and resulting quality assurance (QA) outcomes. However, existing solutions for controlling this trade-off do not guarantee corresponding improvements in deliverability. Therefore, this work explored the feasibility of an optimization framework for directly maximizing predicted QA outcomes of plans without compromising the dosimetric quality of plans designed with an established knowledge-based planning (KBP) technique.
A support vector machine (SVM) was developed - using a database of 500 previous VMAT plans - to predict gamma passing rates (GPRs; 3%/3mm percent dose-difference/distance-to-agreement with local normalization) based on selected complexity features. A heuristic, QA-based optimization (QAO) framework was devised by utilizing the SVM model to iteratively modify mechanical treatment features most commonly associated with suboptimal GPRs. Specifically, leaf gaps (LGs) <50 mm were widened by random amounts, which impacts all aperture-based complexity features. 13 prostate KBP-guided VMAT plans were optimized via QAO using user-specified maximum LG displacements before corresponding changes in predicted GPRs and dose were assessed.
Predicted GPRs increased by an average of 1.14 ± 1.25% (p = 0.006) with QAO using a 3 mm maximum random LG displacement. There were small differences in dose, resulting in similarly small changes in tumor control probability (maximum increase = 0.05%) and normal tissue complication probabilities in the bladder, rectum, and femoral heads (maximum decrease = 0.2% in the rectum).
This study explored the feasibility of QAO and warrants future investigations of further incorporating QA endpoints into plan optimization.
先前的文献表明，计划复杂性和由此产生的质量保证（QA）结果之间存在普遍的权衡。然而，现有的控制这种权衡的解决方案并不能保证交付能力的相应提高。因此，这项工作探讨了开发一种优化框架的可行性，该框架可以直接最大化计划的预测 QA 结果，而不会牺牲使用既定基于知识的规划（KBP）技术设计的计划的剂量学质量。
开发了一个支持向量机（SVM）- 使用 500 个以前的 VMAT 计划的数据库- 根据选定的复杂性特征预测伽马通过率（GPR；3%/3mm 剂量差异/局部归一化的距离-一致性）。利用 SVM 模型设计了一种基于 QA 的启发式优化（QAO）框架，通过迭代修改与次优 GPR 最相关的机械处理特征来优化。具体来说，将叶片间隙（LG）<50mm 随机加宽，这会影响所有基于孔径的复杂性特征。使用用户指定的最大 LG 位移，通过 QAO 对 13 个前列腺 KBP 指导的 VMAT 计划进行了优化，然后评估了预测 GPR 变化和剂量变化。
使用 3mm 的最大随机 LG 位移进行 QAO 后，预测的 GPR 平均增加了 1.14±1.25%（p=0.006）。剂量差异很小，导致肿瘤控制概率（最大增加 0.05%）和膀胱、直肠和股骨头的正常组织并发症概率（直肠最大减少 0.2%）的变化也很小。
本研究探讨了 QAO 的可行性，并为进一步将 QA 终点纳入计划优化提供了未来的研究方向。
