Graph-Based Medical Image Segmentation in 3D and 4D

基于图的 3D 和 4D 医学图像分割

• 批准号: 7918846
• 负责人: MILAN SONKA
• 金额: $ 37.2万
• 依托单位: UNIVERSITY OF IOWA
• 依托单位国家: 美国
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-04-01 至 2013-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7918846
• 关键词: 3-Dimensional; Address; Adoption; Attention; Biomedical Research; Clinical; Complex; Computational Science; Data; Development; Devices; Drug Formulations; Environment; Foundations; Four-dimensional; Graph; Image; Image Analysis; Knowledge; Medical Imaging; Medicine; Methodology; Methods; Performance; Phase; Physicians; Process; Publications; Research; Sampling; Seminal; Shapes; Slice; Solutions; Source; Surface; Technology; Testing; Time; Ultrasonography; Weight; Work; base; bioimaging; clinical practice; cost; design; flexibility; imaging Segmentation; improved; innovation; novel; peer; process optimization; processing speed; public health relevance; response; user-friendly

DESCRIPTION (provided by applicant): This is a competitive continuation of our Phase-I project. After successfully fulfilling all of its aims, a novel framework for optimal multi-surface and/or multi-object n-D biomedical image segmentation was developed, validated, and its practical utility demonstrated in clinical and translational image analysis tasks. This Phase-II proposal will develop several important extensions addressing identified limitations of the original framework while maintaining the ability of detecting optimal single and multiple interacting surfaces in n-D, including cylindrical shapes, closed-surface shapes, and shapes of complex topology. Novel methods will be developed for incorporation of shape-based a priori knowledge; substantial improvement of processing speed; and for interactive operator-guided segmentation. We hypothesize that by representing the segmentation problem in an arc-weighted graph (instead of the so-far utilized node-weighted graph), the 3-D and 4-D multi-surface multi-object optimal graph searching will offer significantly increased segmentation accuracy and robustness in volumetric image data from a variety of medical imaging sources, offering flexibility and higher processing speed, leading to real-time interactivity and practical applicability. We propose to: 1) Develop and validate a single- and multiple-surface n-D graph-based optimal segmentation method that uses arc-based graph representation, incorporates a priori shape knowledge using hard and soft constraints, and provides shape guidance while utilizing weighted combinations of edge-, region-, and shape-based costs. 2) Develop an approach for parallel (multi-core, multi-threaded) optimal graph search to significantly increase the processing speed and thus improving the method's applicability to higher-dimensional, multiply interacting, and overall larger problems. 3) Develop and evaluate an efficient real-time approach for interactive use of single- and multiplesurface segmentations incorporating expert-user guidance while maintaining highly automated character of 3-D or 4-D segmentation. The developed methods will be evaluated against the Phase-I methods to demonstrate statistically significant performance improvements in a variety of tasks with data samples of sufficient sizes. PUBLIC HEALTH RELEVANCE: Project Narrative Three- and four-dimensional (3D + time) analysis of medical image data from MR, CT, ultrasound, or OCT scanners is still performed visually and frequently either non- quantitatively, or only in 2-D slices. Clearly, the 3-D character of the image data provides additional information that may be overlooked by current approaches. The proposed research work is for development of globally optimal image segmentation methods that are practical in 3-D, 4-D and generally n-D medical image data. As such, the study has a promise for facilitating routine clinical analyses of volumetric data from medical image scanners.

描述（由申请人提供）：这是我们第一阶段项目的竞争性延续。在成功实现其所有目标后，开发、验证了一种用于最佳多表面和/或多对象 n 维生物医学图像分割的新颖框架，并在临床和转化图像分析任务中证明了其实用性。该第二阶段提案将开发几个重要的扩展，解决原始框架已确定的局限性，同时保持检测 n 维中最佳单个和多个相互作用表面的能力，包括圆柱形、闭合表面形状和复杂拓扑形状。将开发新方法来合并基于形状的先验知识；处理速度大幅提高；以及交互式操作员引导的分割。我们假设，通过在弧加权图（而不是迄今为止使用的节点加权图）中表示分割问题，3D和4D多表面多对象最优图搜索将显着提高来自各种医学成像源的体图像数据的分割精度和鲁棒性，提供灵活性和更高的处理速度，从而实现实时交互性和实际适用性。我们建议：1）开发和验证基于单表面和多表面n维图的最佳分割方法，该方法使用基于弧的图表示，利用硬约束和软约束结合先验形状知识，并在利用基于边缘、区域和基于形状的成本的加权组合的同时提供形状指导。 2）开发一种并行（多核、多线程）最优图搜索方法，显着提高处理速度，从而提高该方法对高维、多重交互和整体更大问题的适用性。 3) 开发和评估一种有效的实时方法，用于交互式使用单表面和多表面分割，结合专家用户指导，同时保持 3D 或 4D 分割的高度自动化特性。所开发的方法将根据第一阶段的方法进行评估，以证明在具有足够大小的数据样本的各种任务中具有统计显着的性能改进。 公共健康相关性：项目叙述 来自 MR、CT、超声波或 OCT 扫描仪的医学图像数据的三维和四维（3D + 时间）分析仍然以视觉方式进行，并且经常以非定量方式或仅在二维切片中进行。显然，图像数据的 3D 特征提供了当前方法可能忽略的附加信息。所提出的研究工作是为了开发适用于 3D、4D 和一般 n 维医学图像数据的全局最优图像分割方法。因此，该研究有望促进对医学图像扫描仪体积数据的常规临床分析。