Journal of the Korean Physical Society

, Volume 61, Issue 7, pp 1156–1162

Value of a probabilistic atlas in medical image segmentation regarding non-rigid registration of abdominal CT scans

Research Papers


A probabilistic atlas provides important information to help segmentation and registration applications in medical image analysis. We construct a probabilistic atlas by picking a target geometry and mapping other training scans onto that target and then summing the results into one probabilistic atlas. By choosing an atlas space close to the desired target, we construct an atlas that represents the population well. Image registration used to map one image geometry onto another is a primary task in atlas building. One of the main parameters of registration is the choice of degrees of freedom (DOFs) of the geometric transform. Herein, we measure the effect of the registration’s DOFs on the segmentation performance of the resulting probabilistic atlas. Twenty-three normal abdominal CT scans were used, and four organs (liver, spinal cord, left and right kidneys) were segmented for each scan. A well-known manifold learning method, ISOMAP, was used to find the best target space to build an atlas. In summary, segmentation performance was high for high DOF registrations regardless of the chosen target space, while segmentation performance was lowered for low DOF registrations if a target space was far from the best target space. At the 0.05 level of statistical significance, there were no significant differences at high DOF registrations while there were significant differences at low DOF registrations when choosing different targets.


Probabilistic atlas Image analysis Degree of freedom ISOMAP Segmentation Target selection 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A. Guimond, G. Subsol and J.-P. Thirion, Int. J. Pattern Recognit Artif Intell. 11, 8 (1997).CrossRefGoogle Scholar
  2. [2]
    M. Prastawa, J. Gilmore, W. Lin and G. Gerig, Med. Image Anal. 9, 5 (2005).CrossRefGoogle Scholar
  3. [3]
    M. Thompson and A. W. Toga, Med. Image Anal. 1, 4 (1997).CrossRefGoogle Scholar
  4. [4]
    A. Guimond, J. Meunier and J.-P. Thirion, Comput. Vision Image Understanding 77, 2 (2000).CrossRefGoogle Scholar
  5. [5]
    C. Studholme and V. Cardenas, Pattern Recognit. Lett. 25, 10 (2004).CrossRefGoogle Scholar
  6. [6]
    S. Marsland, C. Twining and C. Taylor, Lect. Notes Comput. Sci. 2879, 1 (2003).CrossRefGoogle Scholar
  7. [7]
    H. Park, A. Hero, P. Bland, M. Kessler, J. Seo and C. Meyer, IEICE Trans. Inf. Syst. E93-D, 8 (2010).Google Scholar
  8. [8]
    S. T. Roweis and L. K. Saul, Science 290, 5500 (2000).CrossRefGoogle Scholar
  9. [9]
    H. Tobias and H.-P. Meinzer, Med. Image Anal. 13, 4 (2009).Google Scholar
  10. [10]
    A. C. Zhu and A. Yuille, IEEE Trans. Pattern Anal. Mach. Intell. 18, 9 (1996).Google Scholar
  11. [11]
    S. Robbins, A. Evans, D. Collins and S. Whitesides, Med. Image Anal. 8, 3 (2004).CrossRefGoogle Scholar
  12. [12]
    D. L. G. Hill, P. G. Batchelor, M. Holden and D. J. Hawkes, Phys. Med. Biol. 46, r1 (2001).ADSCrossRefGoogle Scholar
  13. [13]
    C. Meyer, J. L. Boes, B. Kim, P. H. Bland, K. R. Zasadny, P. V. Kison, K. Koral, K. A. Frey and R. L. Wahl, Med. Image Anal. 3, 195 (1997).CrossRefGoogle Scholar
  14. [14]
    J. Ashburner, C. Hutton, R. Frackowiak, I. Johnsrude, C, Price and K. Friston, Hum. Brain Mapp. 6, 348 (1998).CrossRefGoogle Scholar
  15. [15]
    H. Park, P. Bland and C. Meyer, IEEE Trans. Med. Imaging 22, 4 (2003).Google Scholar

Copyright information

© The Korean Physical Society 2012

Authors and Affiliations

  1. 1.School of Electronic Electrical EngineeringSungkyunkwan UniversitySuwonKorea
  2. 2.Department of RadiologyUniversity of MichiganAnn ArborUSA

Personalised recommendations