Methods for 2-D and 3-D Endobronchial Ultrasound Image Segmentation

被引:21
|
作者
Zang, Xiaonan [1 ]
Bascom, Rebecca [2 ]
Gilbert, Christopher [2 ]
Toth, Jennifer [2 ]
Higgins, William [1 ,3 ]
机构
[1] Penn State Univ, Sch Elect Engn & Comp Sci, University Pk, PA 16802 USA
[2] Penn State Milton S Hershey Med Ctr, Dept Med, Hershey, PA USA
[3] Penn State Univ, Dept Biomed Engn, University Pk, PA 16802 USA
关键词
Bronchoscopy; endobronchial ultrasound (EBUS); image-guided intervention system; image segmentation; lung cancer; FAST-MARCHING METHOD; LIVE WIRE; ULTRASONOGRAPHY; DIAGNOSIS; DIFFUSION; MODEL;
D O I
10.1109/TBME.2015.2494838
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
Endobronchial ultrasound (EBUS) is now commonly used for cancer-staging bronchoscopy. Unfortunately, EBUS is challenging to use and interpreting EBUS video sequences is difficult. Other ultrasound imaging domains, hampered by related difficulties, have benefited from computer-based image-segmentation methods. Yet, so far, no such methods have been proposed for EBUS. We propose image-segmentation methods for 2-D EBUS frames and 3-D EBUS sequences. Our 2-D method adapts the fast-marching level-set process, anisotropic diffusion, and region growing to the problem of segmenting 2-D EBUS frames. Our 3-D method builds upon the 2-D method while also incorporating the geodesic level-set process for segmenting EBUS sequences. Tests with lung-cancer patient data showed that the methods ran fully automatically for nearly 80% of test cases. For the remaining cases, the only user-interaction required was the selection of a seed point. When compared to ground-truth segmentations, the 2-D method achieved an overall Dice index = 90.0% +/- 4.9%, while the 3-D method achieved an overall Dice index = 83.9 +/- 6.0%. In addition, the computation time (2-D, 0.070 s/frame; 3-D, 0.088 s/frame) was two orders of magnitude faster than interactive contour definition. Finally, we demonstrate the potential of the methods for EBUS localization in a multimodal image-guided bronchoscopy system.
引用
收藏
页码:1426 / 1439
页数:14
相关论文
共 50 条
  • [1] 2-D and 3-D endoluminal ultrasound
    Liu, JB
    Goldberg, BB
    [J]. ULTRASOUND IN MEDICINE AND BIOLOGY, 2000, 26 : S137 - S139
  • [2] Investigation of the relative contributions of 3-D and 2-D image cues in texture segmentation
    Guyader, N
    Jingling, L
    Lewis, AS
    Zhaoping, L
    [J]. PERCEPTION, 2005, 34 : 55 - 55
  • [3] MOVING BY 2-D METHODS IN A 3-D WORLD
    STEFANID.EJ
    [J]. DESIGN NEWS, 1970, 25 (14) : 80 - +
  • [4] 3-D/2-D registration by integrating 2-D information in 3-D
    Tomazevic, D
    Likar, B
    Pernus, F
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2006, 25 (01) : 17 - 27
  • [5] Image appraisal for 2-D and 3-D electromagnetic inversion
    Alumbaugh, DL
    Newman, GA
    [J]. GEOPHYSICS, 2000, 65 (05) : 1455 - 1467
  • [6] Learning Structured Models for Segmentation of 2-D and 3-D Imagery
    Lucchi, Aurelien
    Marquez-Neila, Pablo
    Becker, Carlos
    Li, Yunpeng
    Smith, Kevin
    Knott, Graham
    Fua, Pascal
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2015, 34 (05) : 1096 - 1110
  • [7] A unified algebraic approach to 2-D and 3-D motion segmentation
    Vidal, R
    Ma, Y
    [J]. COMPUTER VISION - ECCV 2004, PT 1, 2004, 3021 : 1 - 15
  • [8] 2-D and 3-D endoluminal ultrasound: Vascular and nonvascular applications
    Liu, JB
    Goldberg, BB
    [J]. ULTRASOUND IN MEDICINE AND BIOLOGY, 1999, 25 (02): : 159 - 173
  • [9] Detecting 3-D Mirror Symmetry in a 2-D Camera Image for 3-D Shape Recovery
    Sawada, Tadamasa
    Li, Yunfeng
    Pizlo, Zygmunt
    [J]. PROCEEDINGS OF THE IEEE, 2014, 102 (10) : 1588 - 1606
  • [10] 3D Segmentation and Reconstruction of Endobronchial Ultrasound
    Zang, Xiaonan
    Breslav, Mikhail
    Higgins, William E.
    [J]. MEDICAL IMAGING 2013: ULTRASONIC IMAGING, TOMOGRAPHY, AND THERAPY, 2013, 8675