3D IMAGE ANAL ALGORITHMS FOR AUTOMATIC COMP OF GROUND-GLASS OPACITY LUNG TUMOR

这个子项目是许多研究子项目中利用 资源由NIH/NCRR资助的中心拨款提供。子项目和 调查员(PI)可能从NIH的另一个来源获得了主要资金， 并因此可以在其他清晰的条目中表示。列出的机构是 该中心不一定是调查人员的机构。 用于肺肿瘤磨玻璃不透明度自动计算的三维图像分析算法 派：钱德拉·坎巴梅图教授 磨玻璃是指HRCT表现为模糊的不混浊，不会遮盖 相关的肺血管。这种表现是由实质异常引起的。 低于HRCT的空间分辨率。可以观察到结节的GGO区域 在HRCT图像中，由于图像像素的出现，其中的像素是模糊的 并且与其他“厚/不透明”结节像素相比具有较少的不透明度。 CT数据来自克里斯蒂娜关怀中心的海伦·F·格雷厄姆癌症中心。 数据采集将在没有所有患者标识的情况下使用，并根据 IRB批准的方案。 目的1：肺和结节的时间序列比对(S)。我们的目标是获得全球刚性， CT肺和肿瘤数据之间的全局非刚性、局部非刚性对齐参数 在不同的时间情况下获得。我们分割肺部和结节，使用扩展的超二次曲面进行建模，然后在不同的时间实例数据之间进行基于样条的非刚性运动估计。 目的2：肺结节的GGO比值估算及三维测量。使用为每个结节提议的训练系统来确定GGO像素。然后进行测量，包括直径、覆盖面积、体积和GGO覆盖，并进行可视化。 研究问题包括建议的图像分析和可视化工具的可行性，以辅助高分辨率计算机断层图像的放射学分析，以便自动检测、提取、时间对准肺结节，并将其与 毛玻璃不透明度(GGO)比率。结果指标包括对所开发的系统进行定性和定量评估，以帮助选择根治性有限切除的候选人。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. 3D Image Analysis algorithms for Automatic computation of ground-glass opacity (GGO) of lung tumors PI:Prof. Chandra Kambhamettu Ground-glass refers to the HRCT appearance of a hazy opacity that does not obscure the associated pulmonary vessels. This appearance results from parenchymal abnormalities that are below the spatial resolution of HRCT. GGO area of a nodule can be observed in the HRCT images by the appearance of image pixels, where-in pixels are blurred and have less opacity compared to other "thick/opaque" nodule pixels. CT data is obtained from the Helen F. Graham Cancer Center of Christiana Care. Data acquisition will be used absent all patient identifiers and according to IRB approved protocol. Aim 1: Time-series alignment of lung and nodule(s). The goal is to obtain global rigid, global nonrigid, local nonrigid alignment parameters between the CT lung and tumor data obtained at different time instances. We segment the lung and nodules, use Extended-Superquadrics to model, and then perform spline based nonrigid motion estimation between different time instance data. Aim 2: GGO ratio estimation and 3D measurement of lung nodule. GGO pixels are identified using proposed training system for each nodule. Measurements are then performed consisting of the diameter, surface area of coverage, volume and GGO coverage, and visualized. Study questions include the feasibility of proposed image analysis and visualization tools for assistance in the radiologic analysis of high-resolution computed tomographic images in order to automatically detect, extract, time-align lung nodules, and classify them with ground-glass opacity (GGO) ratios. Outcome measures include the qualitative, quantitative evaluation of the developed system in assisting the selection of candidates for curative limited resection.