Toward precision radiotherapy: Physiological modeling of respiratory motion based on ultra-quality 4D-MRI

迈向精准放疗：基于超高质量 4D-MRI 的呼吸运动生理模型

• 批准号: 10653082
• 负责人: Jing Cai
• 金额: $ 41.67万
• 依托单位: UNIVERSITY OF VIRGINIA
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-07-18 至 2024-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10653082
• 关键词: 3-Dimensional; 4D Imaging; 4D MRI; Abdomen; Algorithms; Biological; Biomechanics; Breathing; Cancer Patient; Chest; Communities; Data; Development; Dose; Four-dimensional; Future; Goals; Human; Hybrids; Image; Imaging Techniques; Imaging technology; Liver; Lung; Magnetic Resonance Imaging; Malignant Neoplasms; Measurement; Measures; Medical Imaging; Medicine; Methods; Modeling; Modernization; Morphologic artifacts; Morphology; Motion; Normal tissue morphology; Paint; Patients; Phase; Physiological; Predisposition; Process; Prospective Studies; Radiation Dose Unit; Radiation therapy; Research; Resolution; Sampling; Scheme; Techniques; Technology; Therapeutic; Time; Trees; Walking; cancer radiation therapy; clinical implementation; digital; image guided; image registration; improved; improved outcome; multimodality; novel; personalized medicine; physiologic model; public health relevance; radiation mitigation; radiation-induced injury; radiomics; respiratory; respiratory imaging; spatiotemporal; therapy development; tool; tumor

ABSTRACT Despite numerous advances in radiotherapy in the past decade, which have effectively enhanced local or locoregional tumor control for many patients, there remains substantial room for improvement. A compelling need in today's era of precision radiotherapy is to further widen the therapeutic window and improve radiation dose conformity to the defined target volume, through technological improvements such as advanced image guidance and motion management. Four-dimensional (4D) imaging and deformable image registration (DIR) are two of the most important tools behind many recent radiotherapy advances, but both are facing significant challenges as the requirement for precision increases. Major limitations of current 4D imaging technology include low temporal/spatial resolution, long image acquisition time, suboptimal tumor contrast, and susceptibility to artifacts caused by irregular breathing. Meanwhile, current DIR techniques focus on morphological similarity but not on the physiological plausibility of the deformation, leading to unrealistic results in various applications. These limitations have significantly hampered the advancement of precision radiotherapy. Our long-term goal is to enhance precision radiotherapy through the development and clinical implementation of advanced image guidance and motion management techniques. The overall objective of this application is to develop, cross-- fertilize, and evaluate two techniques: (a) ultra-quality 4D-MRI and (b) physiologically-based motion modeling, for precision radiotherapy applications. Aim 1 is to develop and optimize a 4D-MRI technique for imaging respiratory motion in the thorax and abdomen at ultra-high spatiotemporal resolution. Aim 2 is to develop a physiologically-based motion modeling method for respiratory motion estimation. Aim 3 is to evaluate ultra- quality 4D-MRI and physiological motion modeling in a patient study. Aim 4 is to construct physiologically realistic 4D digital phantoms for future development of precision radiotherapy applications. Successful completion of these aims will yield powerful image guidance and motion management tools for precision radiotherapy. Such improvements will take precision radiotherapy to a whole new level, by significantly improving radiation dose conformity and opening doors for biological-based treatment adaptation for more effective personalized treatment. The proposed research will have a high impact to the fields of both radiotherapy and medical imaging. It will trigger a wave of extensive studies on a number of new and existing applications such as 4D radiotherapy, radiomics, human digital phantom, function-based dose painting, adaptive planning, etc. Most importantly, it will ultimately improve outcomes for cancer patients by improving our ability to precisely deliver radiation treatment to the target and mitigate radiation-induced injury to normal tissues.

抽象的 尽管过去十年放射治疗取得了许多进展，有效地增强了局部或 对于许多患者来说，局部肿瘤控制仍有很大的改进空间。一个引人注目的 当今精准放疗时代需要的是进一步拓宽治疗窗口和提高放射治疗水平 通过先进图像等技术改进，剂量符合定义的目标体积 指导和运动管理。四维 (4D) 成像和变形图像配准 (DIR) 是最近许多放射治疗进展背后最重要的工具，但两者都面临着重大挑战 随着精度要求的提高，挑战不断增加。当前 4D 成像技术的主要局限性包括 时间/空间分辨率低、图像采集时间长、肿瘤对比度欠佳以及对 由不规则呼吸引起的伪影。同时，当前的 DIR 技术侧重于形态相似性，但 不考虑变形的生理合理性，导致在各种应用中产生不切实际的结果。 这些限制极大地阻碍了精准放射治疗的进步。我们的长期目标 是通过先进影像的开发和临床实施来增强精准放射治疗 指导和运动管理技术。该应用程序的总体目标是开发、跨-- 施肥并评估两种技术：(a) 超高质量 4D-MRI 和 (b) 基于生理学的运动建模， 用于精准放射治疗应用。目标 1 是开发和优化 4D-MRI 成像技术 超高时空分辨率下胸部和腹部的呼吸运动。目标 2 是开发一个 用于呼吸运动估计的基于生理学的运动建模方法。目标 3 是评估超 患者研究中的高质量 4D-MRI 和生理运动建模。目标 4 是构建生理上真实的 用于未来精准放射治疗应用开发的 4D 数字体模。顺利完成 这些目标将为精准放射治疗带来强大的图像引导和运动管理工具。这样的 通过显着提高辐射剂量，改进将把精确放射治疗提升到一个全新的水平 顺应性并为基于生物的治疗适应打开大门，以实现更有效的个性化治疗 治疗。拟议的研究将对放射治疗和医学成像领域产生重大影响。 它将引发一波针对许多新的和现有的应用的广泛研究浪潮，例如 4D 放射治疗、 放射组学、人体数字模型、基于功能的剂量绘制、自适应规划等。最重要的是，它将 通过提高我们精确提供放射治疗的能力，最终改善癌症患者的治疗结果 到达目标并减轻辐射对正常组织造成的损伤。

项目成果

Improving liver tumor image contrast and synthesizing novel tissue contrasts by adaptive multiparametric MRI fusion.

通过自适应多参数 MRI 融合提高肝脏肿瘤图像对比度并合成新的组织对比度。

• 发表时间: 2022-09
• 作者: Zhang, Lei;Yin, Fang;Lu, Ke;Moore, Brittany;Han, Silu;Cai, Jing
• 通讯作者: Cai, Jing

MRI super-resolution via realistic downsampling with adversarial learning.

通过真实的下采样和对抗性学习实现 MRI 超分辨率。

• 发表时间: 2021
• 影响因子: 3.5
• 作者: Huang, Bangyan;Xiao, Haonan;Liu, Weiwei;Zhang, Yibao;Wu, Hao;Wang, Weihu;Yang, Yunhuan;Yang, Yidong;Miller, G Wilson;Li, Tian;Cai, Jing
• 通讯作者: Cai, Jing

A dual-supervised deformation estimation model (DDEM) for constructing ultra-quality 4D-MRI based on a commercial low-quality 4D-MRI for liver cancer radiation therapy.

一种双监督变形估计模型 (DDEM)，用于基于用于肝癌放射治疗的商业低质量 4D-MRI 构建超质量 4D-MRI。

• 发表时间: 2022-05
• 影响因子: 3.8
• 作者: Xiao, Haonan;Ni, Ruiyan;Zhi, Shaohua;Li, Wen;Liu, Chenyang;Ren, Ge;Teng, Xinzhi;Liu, Weiwei;Wang, Weihu;Zhang, Yibao;Wu, Hao;Lee, Ho;Cheung, Lai;Chang, Hing;Li, Tian;Cai, Jing
• 通讯作者: Cai, Jing

Motion robust 4D-MRI sorting based on anatomic feature matching: A digital phantom simulation study.

基于解剖特征匹配的运动鲁棒 4D-MRI 分类：数字体模模拟研究。

• 发表时间: 2020-03
• 作者: Yang, Zi;Ren, Lei;Yin, Fang;Liang, Xiao;Cai, Jing
• 通讯作者: Cai, Jing