Department of Radiation Oncology, M D Anderson Cancer Center Orlando, 1400S Orange Ave., Orlando, FL 32806, USA.
Phys Med Biol. 2009 Oct 21;54(20):6165-80. doi: 10.1088/0031-9155/54/20/009. Epub 2009 Oct 1.
Lung tumors move during breathing depending on the patient's patho-physiological condition and orientation, thereby compromising the accurate deposition of the radiation dose during radiotherapy. In this paper, we present and validate a computer-based simulation framework to calculate the delivered dose to a 3D moving tumor and its surrounding normal tissues. The computer-based simulation framework models a 3D volumetric lung tumor and its surrounding tissues, simulates the tumor motion during a simulated dose delivery both as a self-reproducible motion and a random motion using the dose extracted from a treatment plan, and predicts the amount and location of radiation doses deposited. A radiation treatment plan of a small lung tumor (1-3 cm diameter) was developed in a commercial planning system (iPlan software, BrainLab, Munich, Germany) to simulate the radiation dose delivered. The dose for each radiation field was extracted from the software. The tumor motion was simulated for varying values of its rate, amplitude and direction within a single breath as well as from one breath to another. Such variations represent the variations in tumor motion induced by breathing variations. During the simulation of dose delivery, the dose on the target was summed to generate the real-time dose to the tumor for each beam independently. The simulation results show that the dose accumulated on the tumor varies significantly with both the tumor size and the tumor's motion rate, amplitude and direction. For a given tumor motion rate, amplitude and direction, the smaller the tumor size the smaller is the percentage of the radiation dose accumulated. The simulation results are validated by comparing the center plane of the 3D tumor with 2D film dosimetry measurements using a programmable 4D motion phantom moving in a self-reproducible pattern. The results also show the real-time capability of the framework at 40 discrete tumor motion steps per breath, which is higher than the number of four-dimensional computed tomography (CT) steps (approximately 20) during a single breath. The real-time capability enables the framework to be coupled with real-time tumor monitoring systems such as implanted fiducials for computing the dose delivered in real time during the treatment.
肺部肿瘤会在呼吸过程中移动，具体取决于患者的病理生理状况和体位，从而影响放疗过程中辐射剂量的精确沉积。本文提出并验证了一种基于计算机的模拟框架，以计算三维移动肿瘤及其周围正常组织的剂量分布。该模拟框架模型化了三维体积肿瘤及其周围组织，使用从治疗计划中提取的剂量来模拟模拟剂量输送过程中的肿瘤运动（包括自再现运动和随机运动），并预测辐射剂量的沉积量和位置。使用商业计划系统（iPlan 软件，BrainLab，慕尼黑，德国）为一个小的肺部肿瘤（直径为 1-3cm）开发了一个放射治疗计划，以模拟输送的辐射剂量。从软件中提取了每个辐射野的剂量。模拟了肿瘤在单个呼吸内以及从一次呼吸到另一次呼吸的速率、幅度和方向的变化，这些变化代表了呼吸变化引起的肿瘤运动的变化。在剂量输送模拟过程中，对每个射束独立地对目标进行求和，以生成实时肿瘤剂量。模拟结果表明，肿瘤上积累的剂量与肿瘤的大小以及肿瘤的运动速率、幅度和方向有很大的关系。对于给定的肿瘤运动速率、幅度和方向，肿瘤越小，积累的辐射剂量的百分比就越小。通过使用可编程 4D 运动体模在自再现模式下移动，将 3D 肿瘤的中平面与二维胶片剂量学测量进行比较，验证了模拟结果。结果还表明，该框架在每个呼吸的 40 个离散肿瘤运动步骤中具有实时能力，高于单次呼吸期间的四维 CT 步骤数（约 20 个）。实时能力使该框架能够与实时肿瘤监测系统（例如植入式基准标记）耦合，以在治疗过程中实时计算输送的剂量。
