Dynamic multi-organ anatomical models for hypofractionated RT design and delivery

描述（由申请人提供）：低分割放疗技术的进步在治疗传统上与高发病率和局部控制差相关的癌症（如肺癌和肝癌）方面显示出希望。少量的高剂量治疗组分要求在治疗时靶标划定、适形治疗计划和靶标定位方面具有较高的精度和准确性。成像技术在目标识别、治疗时的体积成像能力和时间成像技术方面的进步，提高了在模拟、计划和输送过程中识别肿瘤的能力。该信息的空间配准对于关联来自每张图像的独特信息至关重要，但由于缺乏将所有可用信息整合到一个患者综合模型中的能力而受到限制。早期使用动态多器官解剖模型进行可变形登记的经验使人们认为，变形技术将提高治疗质量，并在临床上显著改善肿瘤控制和降低毒性。虽然验证这一假设需要一个多机构临床试验的综合计划，但这些方法需要在临床研究部署之前建立和评估。该提案提出了三个具体目标，以确保这些技术准备好转化为临床环境，特别是在肺，肝脏和胰腺。在具体目标1中，将开发并验证肺、肝和胰腺的动态多器官解剖模型。这些模型的准确性和呼吸状态之间的线性插值将被量化。异质材料模型将针对肺和肝进行优化。多器官可变形配准对低分割放射治疗的设计和靶向的影响将在具体目标2中进行研究。将评估可变形配准的多模态治疗计划准确性的提高。由于呼吸引起的运动和变形所带来的剂量学准确度的提高将被量化。将研究将准确度的提高转化为临床剂量效应模型。具体目的3评估可变形配准对低分割放疗中剂量记录和计算的影响。将评估使用可变形配准的图像制导精度的改进。将研究累积剂量超过治疗的记录准确性的提高，以及这种改进在剂量效应模型中的转化。本研究的目的是提高放射治疗的准确性和减少不确定性。通过使用动态多器官解剖模型，从先进的成像技术中获得的丰富信息将被组合成一个清晰的患者模型。这种增强的患者模型将提高设计和实施治疗的准确性。

DESCRIPTION (provided by applicant): Advances in hypofractionated radiotherapy techniques have shown promise in the treatment of cancers that are conventionally associated with high morbidity and poor local control (e.g. lung and liver cancer). The small number of high dose treatment fractions requires superior precision and accuracy in target delineation, conformal treatment planning, and target localization at the time of treatment. Advances in imaging for target identification, volumetric imaging capabilities at treatment, and temporal imaging technologies, increase the capability of identifying the tumor during simulation, planning, and delivery. The spatial registration of this information, which is critical to correlate the unique information from each image, is limited by the lacking ability to integrate all available information into one comprehensive model of the patient. Early experience with dynamic multi-organ anatomical models for deformable registration has lead to the hypothesis that deformation technologies will improve the quality of treatment and lead to clinically significant improvements in tumor control and reduced toxicity. While testing this hypothesis will require a comprehensive program of multi-institution clinical trials, these methods need to be established and evaluated prior to deployment in clinical studies. This proposal sets out three specific aims to assure that the technologies are ready for translation into the clinical context, specifically in the lung, liver, and pancreas. In specific aim 1, dynamic multi-organ anatomical models will be developed and validated for the lung, liver, and pancreas. The accuracy of these models and linear interpolation between breathing states will be quantified. Heterogeneous material models will be optimized for the lung and liver. The influence of multi-organ deformable registration on the design and targeting of hypofractionated radiotherapy will be investigated in specific aim 2. The increase in accuracy of multi-modality treatment planning with deformable registration will be evaluated. The improvements in dosimetric accuracy with the inclusion of motion and deformation due to breathing will be quantified. The translation of this increase in accuracy into clinical dose effect models will be investigated. Specific aim 3 evaluates the impact of deformable registration on documentation and accounting of dose in hypofractionated radiotherapy. The improvements in accuracy of image guidance using deformable registration will be assessed. The increase in accuracy of the documentation of accumulated dose over treatment will be investigated, as well as the translation of this improvement in dose effect models. The goal of this research is to improve the accuracy and reduce the uncertainty in radiation therapy. Through the use of dynamic multi-organ anatomical models the wealth of information obtained from advanced imaging techniques will be combined into one, clear, model of the patient. This enhanced patient model will allow improved accuracy in the design and implementation of treatment.