新型基于分层多假设跟踪的冠脉骨架提取算法

朱文博; 李彬; 田联房; 陈侃; 鲍盈含

doi:10.3724/SP.J.1004.2014.01783

新型基于分层多假设跟踪的冠脉骨架提取算法

doi: 10.3724/SP.J.1004.2014.01783

1.
华南理工大学自动化科学与工程学院广州 510641

基金项目:

国家自然科学基金（61305038，61273249），广东省自然科学基金（S2012010009886，S2011010005811），中央高校基本科研业务费重点项目（2013ZZ045），自主系统与网络控制教育部重点实验室资助

详细信息

作者简介:
朱文博华南理工大学自动化科学与工程学院博士研究生.主要研究方向为医学图像处理与模式识别.E-mail：moonkeeper86@126.com

通讯作者:
李彬博士，华南理工大学自动化科学与工程学院副教授.主要研究方向为医学图像处理与模式识别.E-mail：binlee@scut.edu.cn

计量
- 文章访问数: 1975
- HTML全文浏览量: 50
- PDF下载量: 1037
- 被引次数: 0
出版历程
- 收稿日期: 2013-06-27
- 修回日期: 2014-02-20
- 刊出日期: 2014-08-20

A New Coronary Artery Skeleton Extraction Method Based on Layered Multiple Hypothesis Tracking

1.
School of Automation Science and Engineering, South China University of Technology, Guangzhou 510641

Funds:

Supported by National Natural Science Foundation of China (61305038, 61273249), Natural Science Foundation of Guangdong Province (S2012010009886, S2011010005811), the Fundamental Research Fund for the Central Universities (2013ZZ045), and The Key Laboratory of Autonomous Systems and Network Control of Ministry of Education

摘要

摘要: 为解决大多数脉管骨架提取算法中存在的运算复杂、准确率低以及无法同步获取脉管半径问题，提出了一种新型基于分层多假设跟踪的冠脉骨架提取算法. 首先，提出改进局部形状分析方法用于冠脉预分割，通过引入单连通约束和体积约束和降低非血管型结构及细小类血管型结构误分割率；其次，定义新的中心检测能量函数，增强骨架定位能力，并提出分层多假设策略，避免跟踪过程产生局部最优解和实现脉管半径同步获取；此外，通过生成水平集图，使算法可根据脉管树分支情况自动初始化多条跟踪路径，具有较好的拓扑适应性. 实验表明，与其他骨架提取算法相比，该算法可以同步获取冠脉骨架及半径等信息，且结果精度较高.
- 骨架提取 /
- 三维冠脉分割 /
- 局部形状分析 /
- 多假设跟踪 /
- 心脏图像
Abstract: To address the most common problems such as high computational complexity, low extraction accuracy and having difficulty capturing skeleton points and radius simultaneously, a new coronary skeleton extraction method based on layered multiple hypothesis tracking is proposed. First, the improved local shape analysis (ILSA) method is proposed for coronary artery pre-segmentation. Through introducing simply connected and volume constraints, segmentation error rate caused by non-vascular structures and tiny vascular-like structures can be reduced. Secondly, a new medialness measuring energy function (MMEF) is defined to enhance the skeleton point positioning performance. A layered multiple hypothesis tracking (LMHT) strategy is proposed to avoid locally optimal results and obtain vessel radius simultaneously. Additionally, by the level-set graph, multiple tracking paths can be adaptively initialized with better topological flexibility. Experimental analysis shows that compared with other methods, the proposed algorithm can obtain better skeleton extraction performance.
- Skeleton extraction /
- 3D coronary artery segmentation /
- local shape analysis /
- multiple hypothesis tracking /
- cardiac image

HTML全文

参考文献(36)

[1]	Jaeyoun Y, Jong B R. A locally adaptive region growing algorithm for vascular segmentation. Journal of Imaging Systems and Technology, 2003, 13(4): 208-214
[2]	[2] Jia Y Y, Zhuang T G. Applying improved fast marching method to endocardial boundary detection in echocardiographic images. Pattern Recognition Letters, 2003, 24(15): 2777-2784
[3]	[3] Mille J, Bon R, Cohen L D. Region-based 2D deformable generalized cylinder for narrow structures segmentation. In: The European Conference on Computer Vision-ECCV 2008. Marseille, France: Springer, 2008. 392-404
[4]	[4] Mille J, Cohen L D. Deformable tree models for 2D and 3D branching structures extraction. In: Proceedings of the Computer Vision and Pattern Recognition Workshops-CVPR2009. Miami, USA: IEEE, 2009. 149-156
[5]	[5] Descoteaux M, Collins L, Siddiqi K. A Multi-scale Geometric Flow for Segmenting Vasculature in MRI. Berlin: Springer-Verlag, 2004. 169-180
[6]	[6] Kimmel R, Bruckstein A M. Regularized Laplacian zero crossings as optimal edge integrators. International Journal of Computer Vision, 2003, 53(3): 225-243
[7]	[7] Mohammadi A, Asif A. Distributed particle filter implementation with intermittent/irregular consensus convergence. IEEE Transactions on Signal Processing, 2013, 61(10): 2572-2587
[8]	[8] Florin C, Paragios N, Williams J. Globally optimal active contours, sequential Monte Carlo and on-line learning for vessel segmentation. In: Proceedings of the European Conference on Computer Vision-ECCV 2006. Graz, Austria: Springer, 2006. 476-489
[9]	[9] Allen K, Yau C, Noble J. A recursive, stochastic vessel segmentation framework that robustly handles bifurcations. In: Proceedings of Medical Image Understanding and Analytics. Dundee, England: 2008
[10]	Liu Song-Tao, Yin Fu-Liang. The basic principle and its new advances of image segmentation methods based on graph cuts. Acta Automatica Sinica, 2012, 32(1): 911-922(刘松涛, 殷福亮. 基于图割的图像分割方法及其新进展. 自动化学报, 2012, 32(1): 911-922)
[11]	Han Shou-Dong, Zhao Yong, Tao Wen-Bing, Sang Nong. Gaussian super-pixel based fast image segmentation using graph cuts. Acta Automatica Sinica, 2011, 32(1): 11-20(韩守东, 赵勇, 陶文兵, 桑农. 基于高斯超像素的快速Graph Cuts图像分割方法. 自动化学报, 2011, 32(1): 11-20)
[12]	Schaap M, van Walsum T, Neefjes L, Metz C, Capuano E, de Bruijne M, Niessen W. Robust shape regression for supervised vessel segmentation and its application to coronary segmentation in CTA. IEEE Transactions on Medical Imaging, 2011, 30(11): 1974-1986
[13]	Frangi A F, Niessen W J, Vincken K L, Viergever M A. Multi-scale vessel enhancement filtering. In: Proceedings of the 1st International Conference on Medical Image Computing and Computer-assisted Intervention-MICCAI1998. Cambridge, England: Springer, 1998. 130-137
[14]	Lin Q F. Enhancement, Extraction, and Visualization of 3D Volume Data [Ph.D. dissertation], Linkping University, Sweden: 2003
[15]	Bauer C, Pock T, Sorantin E, Bischof H, Beichel R. Segmentation of interwoven 3D tubular tree structures utilizing shape priors and graph cuts. Medical Image Analysis, 2010, 14(2): 172-184
[16]	Diciotti S, Lombardo S, Falchini M, Picozzi G, Mascalchi M. Automated segmentation refinement of small lung nodules in CT scans by local shape analysis. IEEE Transactions on Biomedical Engineering, 2011, 58(12): 3418-3428
[17]	Maurer C R, Qi R, Raghavan V. A linear time algorithm for computing exact Euclidean distance transforms of binary images in arbitrary dimensions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(2): 265-270
[18]	Liu Jun-Tao, Liu Wen-Yu, Wu Cai-Hua, Yuan Liang. A new method of extracting object curve-skeleton. Acta Automatica Sinica, 2008, 32(1): 617-622(刘俊涛, 刘文予, 吴彩华, 原亮. 一种提取物体线形骨架的新方法. 自动化学报, 2008, 32(1): 617-622)
[19]	Pal
[20]	gyi K. A 3D fully parallel surface-thinning algorithm. Theoretical Computer Science, 2008, 406(1-2): 119-135
[21]	Hesselink W H, Roerdink J B. Euclidean skeletons of digital image and volume data in linear time by the integer medial axis transform. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(12): 2204-2217
[22]	Tang H, van Walsum T, van Onkelen R S, Hameeteman R, Klein S, Schaap M, Tori FL, van den Bouwhuijsen Q J, Witteman J C, van der Lugt A, van Vliet L J, Niessen W J. Semiautomatic carotid lumen segmentation for quantification of lumen geometry in multi-spectral MRI.
[23]	Medical Image Analysis, 2012, 16(6): 1202-1215
[24]	Hassouna M S, Faragand A A. Variational curve skeletons using gradient vector flow. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12): 2257-2274
[25]	Xu Y, Zhang H, Li H, Hu G S. An improved algorithm for vessel centerline tracking in coronary angiograms. Computer Methods and Programs in Biomedicine, 2007, 88(2): 131-143
[26]	Bouix S, Siddiqi K, Tannenbaum A. Flux driven automatic centerline extraction. Medical Image Analysis, 2005, 9(3): 209-221
[27]	Liu Wen-Yu, Bai Xiang, Zhu Guang-Xi. A skeleton-growing algorithm based on boundary curve evolution. Acta Automatica Sinica, 2008, 32(1): 255-262(刘文予, 白翔, 朱光喜. 基于边界曲线演化模型的生长骨架算法. 自动化学报, 2006, 32(1): 255-262)
[28]	Wrz S, Rohr K. A new 3D parametric intensity model for accurate segmentation and quantification of human vessels. In: Proceedings of the 7th International Conference on Medical Image Computing and Computer-Assisted Intervention-MICCAI2004. Saint-Malo, France: Springer, 2004. 491-499
[29]	Wrz S, Rohr K. Segmentation and quantification of human vessels using a 3D cylindrical intensity model. IEEE Transactions on Image Processing, 2007, 16(8): 1994-2004
[30]	Friman O, Hindennach M, K
[31]	hnel, Peitgen H O. Multiple hypothesis template tracking of small 3D vessel structures. Medical Image Analysis, 2010, 14(2): 160-171
[32]	Li H, Yezzi A. Vessels as 4D curves: Global minimal 4D paths to extract 3D tubular surfaces and centerlines. IEEE Transactions on Medical Imaging, 2007, 26(9): 1213-1223
[33]	Kang D G, Suh D C, Ra J B. Three-dimensional blood vessel quantification via centerline deformation. IEEE Transactions on Medical Imaging, 2009, 28(3): 405-414
[34]	Battagliero S, Puglia G, Vicario S, Rubino F, Scioscia G, Leo P. An efficient algorithm for approximating geodesic distances in tree space. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2011, 8(5): 1196-1207
[35]	Bharath R S, Ramanathan M. The shortest path in a simply-connected domain having a curved boundary. Computer-Aided Design, 2011, 43(8): 923-933
[36]	Friman O, Hindennach M, Peitgen H O. Template-based multiple hypotheses tracking of small vessels. In: Proceedings of the 5th IEEE International Symposium on Biomedical Imaging. Paris, France: IEEE, 2008. 1047-1050