...
首页> 外文期刊>Sensing and imaging >Three-Dimensions Segmentation of Pulmonary Vascular Trees for Low Dose CT Scans
【24h】

Three-Dimensions Segmentation of Pulmonary Vascular Trees for Low Dose CT Scans

机译:低剂量CT扫描的肺血管树的三维分割

获取原文
获取原文并翻译 | 示例
           

摘要

Due to the low contrast and the partial volume effects, providing an accuraternand in vivo analysis for pulmonary vascular trees from low dose CT scans is a challengingrntask. This paper proposes an automatic integration segmentation approach forrnthe vascular trees in low dose CT scans. It consists of the following steps: firstly, lungrnvolumes are acquired by the knowledge based method from the CT scans, and then therndata are smoothed by the 3D Gaussian filter; secondly, two or three seeds are gotten byrnthe adaptive 2D segmentation and the maximum area selecting from different positionrnscans; thirdly, each seed as the start voxel is inputted for a quick multi-seeds 3D regionrngrowing to get vascular trees; finally, the trees are refined by the smooth filter. Throughrnskeleton analyzing for the vascular trees, the results show that the proposed methodrncan provide much better and lower level vascular branches.
机译:由于低对比度和部分体积效应,通过低剂量CT扫描对肺血管树进行准确的体内分析是一项艰巨的任务。本文提出了一种在低剂量CT扫描中对血管树进行自动积分分割的方法。它包括以下步骤:首先,通过基于知识的方法从CT扫描中获取肺体积,然后通过3D高斯滤波器对数据进行平滑处理。其次,通过自适应二维分割和从不同位置扫描中选择最大面积来获得两到三个种子。第三,输入每个种子作为起始体素,以快速生长多种子的3D区域以生长出维管树。最后,通过平滑过滤器对树木进行细化。通过对血管树的骨架分析,结果表明所提出的方法可以提供更好,更低水平的血管分支。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号