Modeling of radiation dose deposition in time-varying geometries

放射治疗的目的是提供高剂量的电离辐射来摧毁恶性肿瘤细胞，同时保留周围的健康组织，以避免与治疗相关的副作用。传统上，放射治疗计划假定患者的几何形状在治疗过程中保持不变。然而，众所周知，由于呼吸运动，内部解剖结构经历了显著的运动和变形，这可能导致传递剂量分布的模糊和扭曲。 放射治疗的最新进展集中在测量和纠正呼吸引起的解剖变化的方法上。量化呼吸运动的影响和开发新的治疗计划方法需要剂量计算方法来估计沉积在时变(4D)几何形状中的辐射剂量。这些4D剂量计算方法依赖于可变形图像配准，这是一种确定一系列图像之间的几何变换的方法，以获得解剖状态之间的运动模型。 可变形图像配准和4D剂量计算的使用给治疗计划过程带来了额外的复杂性，并且必须仔细评估潜在的不准确性。这项研究旨在研究图像配准中的不确定性如何影响4D剂量计算的精度。我们还将对光子和质子放射治疗的4D剂量计算方法进行综合评估，以量化剂量计算方法带来的不确定度。这项研究的目的是了解图像配准和剂量计算中的不确定性如何限制复杂的4D治疗计划技术的可实现精度，并开发新的健壮算法，以安全地向移动的肿瘤提供预期的放射治疗。

The aim of radiation therapy is to deliver a high dose of ionizing radiation to destroy malignant tumor cells while sparing the surrounding healthy tissue to avoid treatment-related side effects. Traditionally, radiation therapy treatment planning has assumed that the patient's geometry remains static during the treatment delivery. However, it is known that the internal anatomy undergoes significant motion and deformation due to respiratory motion which can result in blurring and distortion of the delivered dose distribution. Recent advances in radiation therapy have focused on methods to measure and correct for respiration-induced anatomical changes. Quantification of the impact of respiratory motion and the development of new treatment planning methods requires dose calculation methods to estimate the radiation dose deposited in the time-varying (4D) geometry. These 4D dose calculation methods rely on deformable image registration, a method for determining the geometric transformation between a series of images, to obtain a model of the motion between anatomical states. The use of deformable image registration and 4D dose calculation introduce additional complexities into the treatment planning process and potential innaccuracies must be carefully evaluated. This research aims to investigate how uncertainties in image registration affect the accuracy of 4D dose calculation. We will also perform a comprehensive evaluation of 4D dose calculation methods for photon and proton radiotherapy to quantify the uncertainty introduced by the dose calculation method. The aim of this research is to understand how the uncertainties in the image registration and dose calculation limit the achievable accuracy of sophisticated 4D treatment planning techniques and to develop new robust algorithms which can safely deliver the intended radiation treatment to moving tumours.