4D cone beam CT reconstruction for radiotherapy via motion vector optimization

描述（由申请人提供）：放射疗法用于肺癌和上腹部癌症治疗。有明确的证据表明，局部控制率和生存率与高辐射剂量水平相关。保留正常组织对于预防并发症也是至关重要的，特别是在同时进行化疗和放疗时。因此，肿瘤靶向非常重要，这对于受呼吸机运动影响的那些部位尤其具有挑战性。多年来，广泛的努力一直致力于开发新的模拟，规划和交付技术，以适应运动。然而，由于计划和治疗会话之间的大的运动变化，这些新的努力受到高质量的4D图像引导模态的缺失的影响。特别是在越来越多地使用的超分割或立体定向体部放射外科中，高分割剂量使得它们对靶向误差的宽容度更低。此外，自适应放射治疗（ART）具有补偿剂量输送误差的潜力。由于呼吸运动和肿瘤收缩，这对于肺和上腹部放射治疗特别重要。为了准确评估治疗适应性的输送剂量，需要高质量的4D锥形束CT（4DCBCT）。最近，4DCBCT已被开发为提供治疗指导的相位分辨图像。由于以下原因，其临床应用受到限制。(1)电流 重建算法需要长扫描协议。然而，即使在4分钟的扫描下，图像质量仍然低于静态物体的标准1分钟3D CBCT，更不用说CT了。(2)T易受散射污染影响。截断问题进一步降低图像质量，并导致视野外解剖结构缺失。这些会降低剂量计算的准确性。放射治疗的一个独特特征是在治疗模拟时患者特异性先验信息的可用性。我们相信，上述问题可以通过将先验信息，以及采用新的重建方法来解决。该提案的总体目标是开发和验证4DCBCT重建系统，该系统在标准1分钟3D CBCT扫描协议下检索具有组合几何和强度精度的高质量4DCBCT。我们的方法通过运动矢量域重建4DCBCT，这是从根本上不同于传统的方法，通过图像强度域。我们的初步研究已证明这个项目的可行性。我们的目标将通过追求以下具体目标（SA）来实现。SA 1.我们将开发一个完整的4DCBCT重建系统。SA 2.我们将验证我们的系统并展示其临床优势。完成后，将开发并全面测试新型4DCBCT重建系统。其临床优势将得到证实。这种系统的临床引入将为IGRT和未来ART下的肺癌和上腹部癌症患者带来巨大益处。

DESCRIPTION (provided by applicant): Radiotherapy is employed for lung and upper abdominal cancer treatments. There are clear evidences that local control and survival rates are correlated with high radiation dose level. Normal tissue sparing is also critical to prevent complications, especially in concurrent chemo- and radio- therapies. Tumor targeting is hence of significant importance, which is particularly challenging for those sites affected by respirator motions. Over the years, extensive efforts have been devoted to developing novel simulation, planning, and delivery techniques to accommodate motion. Yet, due to large motion variations between planning and treatment sessions, those novel efforts are compromised by the missing of high-quality 4D image-guidance modality. Especially in the increasingly used hyper-fractionation or stereotactic body radio-surgery, the high fractional dose makes them less forgiving to targeting error. Moreover, adaptive radiation therapy (ART) holds the potential to compensate errors in dose delivery. This is particularly important for lung and upper abdominal radiotherapy because of respiratory motion and tumor shrinkage. To accurately evaluate delivered dose for treatment adaptation calls for high-quality 4D cone beam CT (4DCBCT). Recently, 4DCBCT has been developed to provide respiratory-phase resolved images for treatment guidance. Its clinical applications are limited due to the following reasons. (1) Current reconstruction algorithms require long scan protocol. Yet, even under a 4-min scans, the image quality is still inferior to that of a standard 1-min 3D CBCT of static objects, let alone CT. (2) t is susceptible to scatter contamination. Truncation problem further degrades image quality and causes missing anatomy outside the field of view. These deteriorate dose calculations accuracy. One unique feature of radiotherapy is the availability of patient-specific prior information at treatment simulation. We believe that the aforementioned problems can be solved by incorporating the prior information, as well as employing novel reconstruction methods. The overall goal of this proposal is to develop and validate a 4DCBCT reconstruction system that retrieves high-quality 4DCBCT with combined geometry and intensity accuracies under a standard 1-min 3D CBCT scan protocol. Our method reconstructs 4DCBCT via motion vector domain, which is fundamentally different from conventional approaches via image intensity domain. The feasibility of this project has been demonstrated by our preliminary studies. Our goal will be accomplished by pursuing the following specific aims (SAs). SA1. We will develop a complete 4DCBCT reconstruction system. SA2. We will validate our system and demonstrate its clinical advantages. Upon completion, a novel 4DCBCT reconstruction system will have been developed and comprehensively tested. Its clinical advantages will have been demonstrated. Clinical introduction of such a system will lead to immense benefits to lung and upper abdominal cancer patients under IGRT and future ART.