ON LINE 3D VASCULATURE FROM BIPLANE ANGIOGRAMS

来自双平面血管造影的在线 3D 脉管系统

• 批准号: 2655258
• 负责人: KENNETH R HOFFMANN
• 金额: $ 23.76万
• 依托单位: UNIVERSITY OF CHICAGO
• 依托单位国家: 美国
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-02-01 至 2000-01-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2655258
• 关键词: angiography; biomedical automation; cardiovascular imaging /visualization; clinical research; computer assisted diagnosis; computer simulation; coronary vessels; human subject; image processing; model design /development; online computer

The goal of this proposed research is to develop a system for determination of imaging geometries for optimal views via computer display of a calculated three-dimensional (3D) coronary vascular tree and for retrospective determination of imaging geometries used in previous studies so that new acquisitions can be made with those geometries while the patient is still on the table. Specifically, we will develop methods (l) to estimate the imaging geometry from biplane angiograms, (2) to determine bifurcation points in biplane angiographic sequences, (3) to determine corresponding points in biplane image sequences, (4) to facilitate determination of optimal views, and (5) to determine retrospectively imaging geometries used in previous studies. Previously, methods have been proposed for determination of the 3D vasculature from biplane images; however, they involve the use of calibration objects or complex measurement protocols, and they are not easily automated. Because the 3D vasculature is not available, multiple acquisitions must be obtained, measurements in coronary images remain subjective and inaccurate, and the imaging geometry of current studies cannot be aligned with that of previous geometries, which reduces the accuracy and precision of comparisons performed in longitudinal studies. In the proposed research, the 3D coronary vasculature will be reconstructed automatically from biplane acquisitions without calibration objects for immediate evaluation. The accuracies in magnification and imaging geometry will be better than 3% and 2 degrees, respectively. With the 3D vasculature, optimal views can be identified without additional radiation dose or contrast load to the patient, and quantitative measurements become more reliable. In addition, we will develop methods for retrospective alignment of current imaging geometries with those of previous studies so that acquisitions with equivalent projections can be obtained to facilitate quantitative measurements of interval change. The significance of the proposed research is that the 3D vasculature will be determined accurately, automatically, and quickly. computerized visualization will allow identification of optimal view, thereby, reducing patient radiation exposure, contrast load, and risk. Immediate retrospective alignment will facilitate longitudinal studies of progression or regression of coronary disease. These techniques can be implemented on current digital biplane systems.

这项拟议研究的目的是为 通过计算机显示确定最佳视图的成像几何形状 计算出的三维（3D）冠状动脉树和用于 回顾性确定先前研究中使用的成像几何形状 因此，可以用这些几何来进行新的收购 病人仍在桌子上。 具体而言，我们将开发方法（L）来估计成像几何形状 从双翼血管造影，（2）确定双翼的分叉点 血管造影序列，（3）确定双翼飞机中的相应点 图像序列，（4）促进确定最佳视图，（5） 确定先前研究中使用的回顾性成像几何形状。 以前，已经提出了确定3D的方法 双翼图像的脉管系统；但是，它们涉及使用 校准对象或复杂的测量协议，它们不是 容易自动化。因为3D脉管系统不可用，所以 必须获得收购，冠状动脉图像中的测量仍在 主观和不准确，以及当前研究的成像几何形状 不能与以前的几何形状保持一致，这会减少 在纵向研究中进行的比较准确性和精度。 在拟议的研究中，3D冠状动脉血管将是 自动从双翼飞机采集自动重建而无需校准 立即评估的对象。放大倍数和 成像几何形状将分别大于3％和2度。和 3D脉管系统可以确定最佳视图，而无需其他 辐射剂量或对比度负荷给患者和定量 测量变得更加可靠。此外，我们将开发方法 为了回顾当前成像几何形状的对准 以前的研究，以便进行同等预测的收购可以是 获得以促进间隔变化的定量测量。 拟议的研究的意义是3D脉管系统将 准确，自动和快速确定。计算机化 可视化将允许识别最佳视图，从而减少 患者辐射暴露，对比负荷和风险。即时 回顾性一致性将促进 冠状动脉疾病的进展或消退。这些技术可以是 在当前的数字双翼系统上实施。