Research on 3D Airway Trees Segmentation and Centerline Extraction Algorithm

Abstract

Abstract: Centerline extraction plays an important role in virtual bronchology. Organ segmentation and centerline extraction with correct topology structure is very difficult, especially for multiple branching structure organs. Therefore, a 3D airway tree segmentation and centerline extraction algorithm was proposed. A new region growing method was proposed to extract the main branch and sub-branch. The initial center was extracted based on the relationship of adjacent section. The center was corrected by computing the value of the determinant of Jacobian matrix, and the path with all the points was generated by Dijkstra's shortest path algorithm. The experimental results show that this method can extract the complete lung airway tree with magnitude 5 bronchus, and illustrate the accuracy for the single and multiple branch organs.

Key words: virtual bronchoscopy, airway tree segmentation, region growing, evaluation function, centerline, Jacobian matrix

CLC Number:

TP391.9

Li Yanbo, Yu Xiang. Research on 3D Airway Trees Segmentation and Centerline Extraction Algorithm[J]. Journal of System Simulation, 2015, 27(10): 2525-2530.

References

[1] M W Graham, J D Gibbs, W E Higgins. A robust system for human airway tree segmentation [C]// SPIE Medical Imaging: Image Processing. USA: SPIE, 2008: 6 9141J-1-69141J-18.
[2] M W Graham, J D Gibbs, W E Higgins.Computer based route definition system for peripheral bronchoscopy[J]. Journal of Digital Imaging (S0897-1889), 2012, 25(2): 307-317.
[3] A P Kiraly, W E Higgins, E A Hoffman, et al. 3D human airway segmentation method for virtual bronchoscopy[J]. Academic Radiology (S1076-6332), 2002, 19(10): 1153-1168.
[4] Wenjun Tan, Jinzhu Yang, Dazhe Zhao, et al. A Novel Method for Automated Segmentation of Airway Tree[C]// 24th Chinese Control and Decision Conference. USA: Institute of Electrical and Electronics Engineers (IEEE), 2012: 976-979.
[5] A Fabijanska.Two pass region growing algorithm for segmenting airway trees for MDCT chent scans[J]. Computerized Medical Imaging and Graphics (S0895-6111), 2009, 11(23): 1353-1364.
[6] T Y Law, P Heng.Automated extraction of bronchus from 3D CT images of lung based on genetic algorithm and 3D region growing[C]// Proceeding of SPIE Medical Imaging. USA: SPIE, 2000, 3979: 906-916.
[7] B van Ginneken, W Baggerman, E M Van Rikxorrt. Robust segmentation and anatomical labeling of the airway tree from thoracic CT scans[C]// Proceeding of MICCAI. Germany: Springer Berlin Heidelberg, 2008: 219-226.
[8] Michael W Graham, Jason D Gibbs, et al. Robust 3d airway tree segmentation for image guided peripheral bronchoscopy[J]. IEEE Transactions on Medical Imaging (S0278-0062), 2010, 29(4): 982-997.
[9] 王怡, 周明全, 耿国华. 虚拟内窥镜中心路径抽取技术[J]. 西北大学学报自然科学版, 2005, 35(6): 695-698.
[10] Saito T, Toriwaki J I.New algorithm for Euclidean distance transformation of an n-dimensional digitized picture with applications[J]. Pattern Recognition (S0031-3203), 1994, 27(11): 1551-1565.
[11] 彭延军, 石教英. 一种新的抽取中心路径算法[J]. 浙江大学学报(工学版), 2003, 37(3): 259-263.
[12] 胡英, 候悦, 徐心和. 基于双距离场的三维中心路径提取算法[J]. 中国图像图形学报, 2003, 8(11): 1272-1276.
[13] 竺海, 姬红兵, 高新波. 基于边界距离场的管腔中心路径自动提取算法[J]. 计算机辅助设计与图形学学报, 2006, 18(6): 860-864.
[14] 王海录. 体绘制下虚拟内窥镜系统自动漫游技术研究[D]. 包头: 内蒙古科技大学, 2014.
[15] 胡英, 侯悦, 徐心和. 基于双距离场的三维中心路径提取算法[J]. 中国图象图形学报, 2003, 8(11): 1272-1276.
[16] He TS, Hong LC, Chen DQ, Liang ZR.Reliable path for virtual endoscopy: Ensuring complete examination of human organs[J]. IEEE Transaction on Visualization and Computer Graphics (S1077-2626), 2001, 7(4): 33-342.
[17] 李光明, 田捷, 赵明昌, 等. 基于Hessian矩阵的中心路径提取算法[J]. 软件学报, 2003, 14(12): 2074-2081.