Vessel Tree Segmentation in Presence of Interstitial Lung Disease in MDCT

被引:20
作者
Korfiatis, Panayiotis D. [1 ]
Kalogeropoulou, Cristina [2 ]
Karahaliou, Anna N. [1 ]
Kazantzi, Alexandra D. [2 ]
Costaridou, Lena I. [1 ]
机构
[1] Univ Patras, Sch Med, Dept Med Phys, Patras 26500, Greece
[2] Univ Hosp Patras, Dept Radiol, Patras 26500, Greece
来源
IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE | 2011年 / 15卷 / 02期
关键词
Computed tomography; image enhancement; image segmentation; THORACIC CT; CLASSIFICATION; ENHANCEMENT; QUANTIFICATION; DIAGNOSIS; IMAGES;
D O I
10.1109/TITB.2011.2112668
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The automated segmentation of vessel tree structures is a crucial preprocessing stage in computer aided diagnosis (CAD) schemes of interstitial lung disease (ILD) patterns in multidetector computed tomography (MDCT). The accuracy of such preprocessing stages is expected to influence the accuracy of lung CAD schemes. Although algorithms aimed at improving the accuracy of lung fields segmentation in presence of ILD have been reported, the corresponding vessel tree segmentation stage is under-researched. Furthermore, previously reported vessel tree segmentation methods have only dealt with normal lung parenchyma. In this paper, an automated vessel tree segmentation scheme is proposed, adapted to the presence of pathologies affecting lung parenchyma. The first stage of the method accounts for a recently proposed method utilizing a 3-D multiscale vessel enhancement filter based on eigenvalue analysis of the Hessian matrix and on unsupervised segmentation. The second stage of the method is a texture-based voxel classification refinement to correct possible over-segmentation. The performance of the proposed scheme, and of the previously reported technique, in vessel tree segmentation was evaluated by means of area overlap (previously reported: 0.715 +/- 0.082, proposed: 0.931 +/- 0.027), true positive fraction (previously reported: 0.968 +/- 0.019, proposed: 0.935 +/- 0.036) and false positive fraction (previously reported: 0.400 +/- 0.181, proposed: 0.074 +/- 0.031) on a dataset of 210 axial slices originating from seven ILD affected patient scans (used for performance evaluation out of 15). The proposed method demonstrated a statistically significantly (p < 0.05) higher performance as compared to the previously reported vessel tree segmentation technique. The impact of suboptimal vessel tree segmentation in a reticular pattern quantification system is also demonstrated.
引用
收藏
页码:214 / 220
页数:7
相关论文
共 31 条
[1]   Vessel tree reconstruction in thoracic CT scans with application to nodule detection [J].
Agam, G ;
Armato, SG ;
Wu, CH .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2005, 24 (04) :486-499
[2]  
[Anonymous], 1995, NATURE STAT LEARNING
[3]   Automated lung segmentation for thoracic CT: Impact on computer-aided diagnosis [J].
Armato, SG ;
Sensakovic, WF .
ACADEMIC RADIOLOGY, 2004, 11 (09) :1011-1021
[4]   HRCT diagnosis of diffuse parenchymal lung disease: interobserver variation [J].
Aziz, ZA ;
Wells, AU ;
Hansell, DM ;
Bain, GA ;
Copley, SJ ;
Desai, SR ;
Ellis, SM ;
Gleeson, FV ;
Grubnic, S ;
Nicholson, AG ;
Padley, SPG ;
Pointon, KS ;
Reynolds, JH ;
Robertson, RJH ;
Rubens, MB .
THORAX, 2004, 59 (06) :506-511
[5]   Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis [J].
Chen, Shifeng ;
Zhou, Sumin ;
Yin, Fang-Fang ;
Marks, Lawrence B. ;
Das, Shiva K. .
MEDICAL PHYSICS, 2007, 34 (10) :3808-3814
[6]  
Einslein K., 1977, Statistical Methods for Digital Computer, P76
[7]  
Frangi AF, 1998, LECT NOTES COMPUT SC, V1496, P130, DOI 10.1007/BFb0056195
[8]   TEXTURAL FEATURES FOR IMAGE CLASSIFICATION [J].
HARALICK, RM ;
SHANMUGAM, K ;
DINSTEIN, I .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1973, SMC3 (06) :610-621
[9]   Fuzzy pulmonary vessel segmentation in contrast enhanced CT data [J].
Kaftan, Jens N. ;
Kiraly, Atilla P. ;
Bakai, Annemarie ;
Das, Marco ;
Novak, Carol L. ;
Aach, Til .
MEDICAL IMAGING 2008: IMAGE PROCESSING, PTS 1-3, 2008, 6914
[10]   Efficient tuning of SVM hyperparameters using radius/margin bound and iterative algorithms [J].
Keerthi, SS .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (05) :1225-1229