Hybrid virtual-MRI/CBCT: A new paradigm for image guidance in liver SBRT

混合虚拟 MRI/CBCT：肝脏 SBRT 图像引导的新范例

• 批准号: 10463464
• 负责人: Lei Ren
• 金额: $ 51.34万
• 依托单位: UNIVERSITY OF MARYLAND BALTIMORE
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-30 至 2023-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10463464
• 关键词: 3-Dimensional; Address; Algorithms; Anatomy; Atlases; Award; Breathing; Cessation of life; Clinic; Clinical; Clinical Trials; Data; Dose; Elements; Ensure; Failure; Financial compensation; Fractionated radiotherapy; Goals; Hepatotoxicity; Hybrids; Image; Imaging Techniques; Lead; Liver; Liver neoplasms; Magnetic Resonance Imaging; Malignant neoplasm of liver; Malignant neoplasm of prostate; Maps; Methods; Motion; Multimodal Imaging; Normal tissue morphology; Oncology; Patients; Psychological Transfer; Publications; Publishing; Radiation Oncology; Radiation therapy; Residual state; Site; Surface; System; Techniques; Tissues; Toxic effect; Treatment outcome; Tumor Volume; Variant; Visualization; base; clinical practice; clinical research site; clinically significant; cone-beam computed tomography; convolutional neural network; cost; falls; image guided; image guided radiation therapy; imaging capabilities; imaging system; improved; improved outcome; industry partner; machine learning algorithm; malignant breast neoplasm; multimodality; next generation; novel; prospective; random forest; recursive neural network; respiratory; simulation; soft tissue; tumor; virtual

ABSTRACT This proposal aims to establish a hybrid virtual-MRI/CBCT system for precision image guidance in radiation therapy of liver cancer. Stereotactic body radiation therapy (SBRT) improves the treatment outcome over conventional fractionated radiotherapy by delivering high fractional dose in few fractions. However, SBRT of liver cancer still has high local failure rate up to 20-30% at 2-3 years and high overall grade>=3 toxicity rate up to 30% with patient deaths from the treatment. Previous studies demonstrated a strong correlation between the treatment outcome and the tumor localization accuracy of the image guidance technique used. Currently, cone beam CT (CBCT) is the most widely used imaging technique for liver SBRT. However, CBCT has extremely limited soft tissue contrast with no visualization of most liver tumors. As a result, its localization error is frequently over 6mm, and the dose coverage of tumor volume can fall below 90%, which is far from enough for tumor control. Drastically reducing localization errors is imperative to improve the outcome of liver SBRT. Combined MRI-Radiotherapy machines could provide MRI contrast for liver tumor localization. However, such machines are rare and expensive for most clinics. As such, an alternative solution is urgently required to meet the ubiquitous need of all radiation oncology departments: a solution in the form of a novel low-cost imaging system with MRI equivalent contrast. Since CBCT is readily available on most radiotherapy machines, creating CBCT-based imaging systems would be the most effective strategy. Our recent publication demonstrated the feasibility of such strategy. (Harris, Med Phys 2018) Our long-term goal is to establish the next generation of CBCT systems with virtual multi-modality imaging capabilities for high precision tumor localization and adaptive therapy in radiation therapy. In this application, our overall objective is to establish a novel system to generate hybrid virtual-MRI/CBCT from CBCT, with virtual-MRI inside the liver for target localization and augmented CBCT outside the liver for healthy tissue verification. The specific aims are: (1). Establish a 3D hybrid virtual- MRI/CBCT system for image guidance in breath-hold liver SBRT without respiratory motion. (2). Establish a 4D hybrid virtual-MRI/CBCT system for image guidance in free-breathing liver SBRT with respiratory motion. (3). Optimize and clinically evaluate the systems through simulation studies and a prospective clinical trial. Our expected deliverables through the academic-industrial partnership are novel hybrid virtual-MRI/CBCT systems ready to be integrated to all radiotherapy machines with only CBCT systems. This new imaging capability will have significant clinical impact because it will create a paradigm shift in image-guided radiation therapy by revolutionizing the widely used CBCT systems to enable a new era of low-cost high-precision virtual multi- modality guided radiotherapy. This precision guidance can lead to breakthrough in escalating the tumor control and minimizing the toxicities of liver SBRT as well as treatments of other sites with tumors in the soft tissue.

摘要 本提案旨在建立一个混合的虚拟MRI/CBCT系统，用于辐射中的精确图像引导 肝癌的治疗。立体定向放射治疗（SBRT）改善了治疗效果， 通过在少数分次中递送高分次剂量，常规分次放射治疗是有效的。然而，SBRT 肝癌在2-3年时仍具有高达20-30%的高局部失败率， 30%的患者死于治疗。先前的研究表明， 治疗结果和所用图像引导技术的肿瘤定位准确性。目前，CONE 束CT（CBCT）是用于肝脏SBRT的最广泛使用的成像技术。然而，CBCT非常 软组织对比度有限，大多数肝脏肿瘤无法显示。因此，其定位误差为 通常超过6 mm，肿瘤体积的剂量覆盖率可以下降到90%以下，这远远不够， 肿瘤控制减少定位误差是改善肝脏SBRT预后的关键。 MRI-放疗联合机可为肝脏肿瘤定位提供MRI对比。但这种 对于大多数诊所来说，机器是罕见且昂贵的。因此，迫切需要一种替代解决方案， 所有放射肿瘤科的普遍需求：一种新型低成本成像形式的解决方案 MRI等效对比度系统。由于CBCT在大多数放射治疗机器上都很容易获得， 基于CBCT的成像系统将是最有效的战略。我们最近的出版物表明， 这种战略的可行性。（Harris，Med Phys 2018）我们的长期目标是建立下一代 具有虚拟多模态成像功能的CBCT系统，用于高精度肿瘤定位和自适应 放射治疗中的治疗。在这个应用程序中，我们的总体目标是建立一个新的系统， CBCT的混合虚拟MRI/CBCT，肝脏内的虚拟MRI用于靶定位和增强 肝外CBCT，用于健康组织验证。具体目标是：（1）.建立三维混合虚拟- MRI/CBCT系统用于无呼吸运动的屏气肝脏SBRT中的图像引导。（二）、建立4D 混合虚拟MRI/CBCT系统在具有呼吸运动自由呼吸肝脏SBRT中用于图像引导。（三）、 通过模拟研究和前瞻性临床试验对系统进行优化和临床评估。我们 通过学术界和工业界的合作，预期可交付的成果是新型混合虚拟MRI/CBCT系统 可集成到所有仅配备CBCT系统的放射治疗机中。这种新的成像能力将 具有显著的临床影响，因为它将通过以下方式在图像引导放射治疗中创造范式转变： 彻底改变了广泛使用的CBCT系统，开创了低成本高精度虚拟多通道CT的新时代。 模态引导放射治疗。这种精确的指导可以导致肿瘤控制升级的突破 以及最小化肝SBRT的毒性以及软组织中具有肿瘤的其他部位的治疗。