Ziegenhein Peter, Ph Kamerling Cornelis, Oelfke Uwe
Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, SM2 5NG, UK.
Phys Med Biol. 2016 Mar 21;61(6):2457-70. doi: 10.1088/0031-9155/61/6/2457. Epub 2016 Mar 7.
In this work we present a novel treatment planning technique called interactive dose shaping (IDS) to be employed for the optimization of intensity modulated radiation therapy (IMRT). IDS does not rely on a Newton-based optimization algorithm which is driven by an objective function formed of dose volume constraints on pre-segmented volumes of interest (VOIs). Our new planning technique allows for direct, interactive adaptation of localized planning features. This is realized by a dose modification and recovery (DMR) planning engine which implements a two-step approach: firstly, the desired localized plan adaptation is imposed on the current plan (modification) while secondly inevitable, undesired disturbances of the dose pattern elsewhere are compensated for automatically by the recovery module. Together with an ultra-fast dose update calculation method the DMR engine has been implemented in a newly designed 3D therapy planning system Dynaplan enabling true real-time interactive therapy planning. Here we present the underlying strategy and algorithms of the DMR based planning concept. The functionality of the IDS planning approach is demonstrated for a phantom geometry of clinical resolution and size.
在这项工作中，我们提出了一种名为交互式剂量塑形（IDS）的新型治疗计划技术，用于优化调强放射治疗（IMRT）。IDS不依赖于基于牛顿法的优化算法，该算法由对预分割感兴趣体积（VOIs）的剂量体积约束构成的目标函数驱动。我们的新计划技术允许对局部计划特征进行直接、交互式调整。这通过剂量修改与恢复（DMR）计划引擎实现，该引擎采用两步法：首先，将所需的局部计划调整施加到当前计划上（修改），而其次，剂量模式在其他地方不可避免的、不期望的干扰由恢复模块自动补偿。与超快速剂量更新计算方法一起，DMR引擎已在新设计的3D治疗计划系统Dynaplan中实现，实现了真正的实时交互式治疗计划。在此，我们介绍基于DMR的计划概念的基本策略和算法。针对具有临床分辨率和尺寸的体模几何结构展示了IDS计划方法的功能。
