Prospective 3-D Treatment Planning for MR-Guided Laser Induced Thermal Therapy Pr

描述（由申请人提供）：磁共振引导激光诱导热疗法（MRgLITT）用于治疗脑部癌变病变，是传统手术的一种微创替代方法，目前正在全球范围内进行多项正在进行的临床试验。这些疗法结合了实时磁共振温度成像（MRTI）来提供反馈，使这些过程安全可行。然而，当这些疗法转化为临床研究时，有两个不足值得解决。首先，在存在组织界面和对流热传递（例如，心室、血管和组织灌注）的情况下，激光加热可能会使治疗计划变得困难，因为假设对称椭球状病变产生困难，特别是当使用任意放置的多个涂抹器时。其次，MRTI信息在某些区域可能会损坏（例如，从温度依赖的信号丢失，血液或界面的敏感性影响），温度信息可能丢失或变得非常不确定，使得对治疗结果的决策变得困难。因此，为了提高治疗交付的安全性、有效性和一致性，迫切需要MRgLITT程序的前瞻性3D治疗计划以及更强大的实时监测来补充MRTI。该研究的主要目标是开发和验证算法，为大脑中的MRgLITT程序提供前瞻性3D治疗计划和实时模型驱动的治疗监测。这样的系统将结合目前用于神经外科和立体定向治疗计划的3D MRI采集的使用，允许用户交互式地导航组织，模拟各种涂抹器轨迹和激光照射的效果，一旦激光定位，更新这些计划，并根据获得的多层MR温度成像输入提供实时3D损伤估计。该项目的完成将不仅提供上述工具，而且为我们的长期目标提供框架，将这些模型应用于计算机优化的反向治疗计划以及自适应治疗控制的实时建模。这个项目既创新又及时，因为目前还没有技术可以对这种新兴的治疗方案进行前瞻性的3D规划，更不用说将这种算法与数据驱动的模型预测相结合，以帮助监测和治疗评估。这项研究是高风险的，因为在接近实时的情况下要解决的计算问题的复杂性存在物理和理论上的障碍，但我们之前在该领域的经验表明它是可行的。

DESCRIPTION (provided by applicant): MR-guided laser induced thermal therapy (MRgLITT) for treatment of cancerous lesions in the brain presents a minimally invasive alternative to conventional surgery, gaining use worldwide with multiple on-going clinical trials currently. These therapies incorporate real-time MR temperature imaging (MRTI) to provide feedback which makes these procedures safe and feasible. However, as these therapies translate into clinical studies, there are two deficiencies worth addressing. Firstly, laser heating in the presence of tissue interfaces and convective heat transfer (e.g., ventricles, vessels and tissue perfusion) may render planning of treatments by assuming symmetric ellipsoidal lesion generation difficult, particularly when multiple applicators with arbitrary placement are used. Second, MRTI information may become corrupt in some regions (e.g., from temperature dependent signal losses, blood, or susceptibility effects at interfaces) and the temperature information may be lost or become extremely uncertain, making decisions about treatment outcome difficult. Therefore, in order to enhance the safety, efficacy and conformality of treatment delivery, there exists a critical need for prospective 3D treatment planning of MRgLITT procedures as well as more robust real-time monitoring to complement MRTI. The primary goal of the proposed research is to develop and validate algorithms to provide both prospective 3D treatment planning and real-time model driven treatment monitoring for MRgLITT procedures in the brain. Such a system will incorporate the use of 3D MRI acquisitions currently used for neurosurgical and stereotactic treatment planning and allow the user to interactively navigate the tissue and simulate the effects of various applicator trajectories and laser exposures, update these plans once the laser is positioned, as well as provide real-time 3D estimation of damage based on the acquired multi-slice MR temperature imaging input. Completion of this project will provide not only the tools outlined above, but a framework for our long term goal of applying these models for computer optimized inverse treatment planning as well as real-time modeling for adaptive control of therapy. This project is both innovative and timely in that no technology presently exists for prospective 3D planning of this emerging therapeutic option, much less, integration of such an algorithm with data driven model prediction assistance for monitoring and treatment assessment. This research is high risk in that the complexity of the computational problems to be addressed in near real-time present both physical and theoretical hurdles to overcome, but our previous experience in this area indicate it is feasible. PUBLIC HEALTH RELEVANCE: MR-guided laser induced thermal therapy (MRgLITT) for the treatment of cancerous lesions in the brain is a minimally invasive alternative to conventional surgery and is now available as a commercialized product with many clinical trials in operation or beginning. The primary goal of the proposed research is to develop, investigate and validate 3D prospective treatment planning and real-time MRTI driven model prediction assisted monitoring for MRgLITT procedures in the brain in order to enhance the efficacy and safety of these procedures. This project is both innovative and timely in that no technology presently exists to support prospective 3D treatment planning of this emerging therapeutic option nor data driven real-time modeling for prediction of temperature and tissue damage.