全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

应用球形算子层次聚类的3维冠脉跟踪提取

DOI: 10.11834/jig.20140814

Keywords: 计算机断层血管造影术,分割,球形算子,方向聚类

Full-Text   Cite this paper   Add to My Lib

Abstract:

目的计算机断层血管造影术(CTA)是冠脉疾病诊疗过程中最为常用的成像方法之一。然而造影图像中冠脉与其周边组织的灰度分布较为接近,冠脉整体结构的识别较为困难。为此提出一种基于方向聚类的冠脉结构跟踪提取方法。方法该方法首先使用闭运算及灰度拉伸对影像进行预处理,然后在选定的种子点构造球形算子,该球形算子由多对方向相反且从球心指向球壳的射线组成,通过对多组数据进行分析得到血管在球形算子内部的灰度差异率特征,对灰度差异率特征曲线构造2维凸包可得到射线与血管壁的交点,以此获得血管段的分割结果以及指向血管走势的方向向量的集合,应用最小距离层次聚类法对向量集合进行分类得到跟踪方向,遍历整条血管完成对冠状动脉的分割。结果实验结果表明,本文方法能够获得较为精确的冠脉提取结果,与近年流行算法相比在临床有效血管段的跟踪上具有一定优势。能够达到0.39mm的跟踪精度。结论本文方法的优势在于所提出的球形算子对不同形态的3维血管具有较好的适应性和鲁棒性。用于向量分类的层次聚类方法无需训练,算法自动化程度高,且分类较为准确。血管跟踪提取方法精度较高。

References

[1]  Yang G Y, Kitslaar P, Frenay M, et al. Automatic centerline extraction of coronary arteries in coronary computed tomographic angiography[J]. The International Journal of Cardiovascular Imaging, 2012, 28(4): 921-933.
[2]  Frangi A, Niessen W, Vincken K, et al. Multiscale vessel enhancement filtering[J]. Medical Image Computing and Computer-Assisted Interventation, 1998: 130-137.
[3]  Egger J, Kapur J, Dukatz T, et al. Sqware-Cut: a segmentation atgorithm on the basis of a rectangte shape[J]. Peos one, 2012, 7(2): 1-13.
[4]  Egger J, Grosskopf S, Nimsky C, et al. Modeling and visualization techniques for virtual stenting of aneurysms and stenoses[J]. Computerized Medical Imaging and Graphics, 2012, 36(3):183-203.
[5]  Egger J, Bauer M, Kuhnt D, et al. Nugget-cut: a segmentation scheme for spherically-and elliptically-shaped 3D objects[J]. Pattern Recognition, 2010, 6376: 373-382.
[6]  Boykov Y, Kolmogorov V. An experimental comparison of min-cut/max-flow algorithms for energy minimization in Vision[J]. IEEE Transactions on PAMI, 2004, 26: 1124-1137.
[7]  Whited B, Rossignac J, Slabaugh G, et al. Pearling: stroke segmentation with crusted pearl strings[J]. Pattern Recognition and Image Analysis, 2009, 19:277-283.
[8]  Tyrrell J A, di Tomaso E, Fuja D, et al. Robust 3-D modeling of vasculature imagery using superellipsoids[J]. IEEE Transactions on Medical Imaging, 2007, 26:223-237.
[9]  Cloutier G, Soulez G, Qanadli S D, et al. A multimodality vascular imaging phantom with fiducial markers visible in DSA, CTA, MRA, and ultrasound[J]. Medical Physics, 2004, 31:14-24.
[10]  Schaap M, Metz C T, Walsum T, et al. Standardized evaluation methodology and reference database for evaluating coronary artery centerline extraction algorithms[J]. Medical Image Analysis, 2009, 13:701-714.
[11]  Zambal S, Hladuvka J, Kanitsar A, et al, Shape and appearance models for automatic coronary artery tracking[J/OL]. The Midas Journal, 2008. http://hdl.handle.net/10380/1460.
[12]  Wang C, Smedby O. An automatic seeding method for coronary artery segmentation and skeletonization in CTA[J/OL]. The Midas Journal, 2008. http://hdl.handle.net/10380/1434.
[13]  Kitslaar P, Frenay M, Oost E, et al. Connected component and morpholgy based extraction of arterial centerlines of the heart (CocomoBeach)[J/OL]. The Midas Journal, 2008. http://hdl.handle.net/10380/1460.
[14]  Gülsün M A, Tek H. Robust vessel tree modeling[C]//Medical Image Computing and Computer-Assisted Intervention. New York: Springer, 2008, 602-611.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133