Skip to main content

Model-Based Initialisation for Segmentation

  • Conference paper
  • First Online:

Part of the Lecture Notes in Computer Science book series (LNCS,volume 1843)

Abstract

The initialisation of segmentation methods aiming at the localisation of biological structures in medical imagery is frequently regarded as a given precondition. In practice, however, initialisation is usually performed manually or by some heuristic preprocessing steps. Moreover, the same framework is often employed to recover from imperfect results of the subsequent segmentation. Therefore, it is of crucial importance for everyday application to have a simple and effective initialisation method at one’s disposal. This paper proposes a new model-based framework to synthesise sound initialisations by calculating the most probable shape given a minimal set of statistical landmarks and the applied shape model. Shape information coded by particular points is first iteratively removed from a statistical shape description that is based on the principal component analysis of a collection of shape instances. By using the inverse of the resulting operation, it is subsequently possible to construct initial outlines with minimal effort. The whole framework is demonstrated by means of a shape database consisting of a set of corpus callosum instances. Furthermore, both manual and fully automatic initialisation with the proposed approach is evaluated. The obtained results validate its suitability as a preprocessing step for semi-automatic as well as fully automatic segmentation. And last but not least, the iterative construction of increasingly point-invariant shape statistics provides a deeper insight into the nature of the shape under investigation.

Keywords

  • Basis Vector
  • Independent Component Analysis
  • Goal Function
  • Object Instance
  • Active Appearance Model

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. Andrew Blake and Michael Isard. Active Contours. Springer, first edition, 1998. ISBN 3-54-076217-5.

    Google Scholar 

  2. T. F. Cootes, G. J. Edwards, and C. J. Taylor. Active appearance models. In Proceedings 5 thEuropean Conference on Computer Vision, volume 2, pages 484–498. Springer, 1998. ISBN 3-540-64613-2.

    Google Scholar 

  3. T. F. Cootes, A. Hill, C. J. Taylor, and J. Haslam. The use of active shape models for locating structures in medical images. In H. H. Barrett and A. F. Gmitro, editors, Information Processing in Medical Imaging, pages 33–47. Springer, June 1993.

    Google Scholar 

  4. T. F. Cootes and C. J. Taylor. Active shape models-“Smart Snakes”. In Proceedings British Machine Vision Conference, pages 266–275, Leeds, 1992. Springer. ISBN 3-540-19777-X.

    Google Scholar 

  5. T. F. Cootes and C. J. Taylor. Statistical models of appearance for computer vision. Technical report, University of Manchester, Wolfson Image Analysis Unit, Imaging Science and Biomedical Engineering, Manchester M13 9PT, United Kingdom, September 1999. http://www.wiau.man.ac.uk.

    Google Scholar 

  6. M. A. Fischler, J. M. Tenenbaum, and H. C. Wolf. Detection of roads and linear structures in low-resolution aerial imagery using a multisource knowledge integration technique. Computer Graphics and Image Processing, 15:201–233, 1981.

    CrossRef  Google Scholar 

  7. David R. Forsey and Richard H. Bartels. Hierarchical B-spline refinement. In Computer Graphics Proceedings SIGGRAPH 1988, pages 205–212. Addison-Wesley, 1988.

    Google Scholar 

  8. J. Hug, C. Brechbühler, and G. Székely. Tamed Snake: A Particle System for Robust Semi-automatic Segmentation. In Medical Image Computing and Computer-Assisted Intervention-MICCAI’99, number 1679 in Lecture Notes in Computer Science, pages 106–115. Springer, September 1999.

    CrossRef  Google Scholar 

  9. Aapo Hyvärinen. Independent component analysis by minimization of mutual information. Technical Report A46, Helsinki University of Technology, Department of Computer Science and Engineering, Laboratory of Computer and Information Science, Rakentajanaukio 2 C, FIN-02150 Espoo, Finland, August 1997. http://www.cis.hut.fi/~aapo.

    Google Scholar 

  10. M. Kass, A. Witkin, and D. Terzopoulos. Snakes: Active contour models. In Proceedings 1 stInternational Conference on Computer Vision, pages 259–268, June 1987.

    Google Scholar 

  11. András Kelemen. Elastic Model-Based Segmentation of 2-D and 3-D Neurora-diological Data Sets. PhD thesis, Swiss Federal Institute of Technology Zürich, Switzerland, 1998. ETH Dissertation No. 12907.

    Google Scholar 

  12. András Kelemen, Gábor Székely, and Guido Gerig. Three-dimensional model-based segmentation. In IEEE International Workshop on Model-Based 3D Image Analysis, pages 87–96, January 1998.

    Google Scholar 

  13. András Kelemen, Gábor Székely, and Guido Gerig. Elastic Model-Based Segmentation of 3-D Neuroradiological Data Sets. IEEE Transactions on Medical Imaging, 18(10):828–839, October 1999.

    Google Scholar 

  14. F. P. Kuhl and C. R. Giardina. Elliptic Fourier features of a closed contour. Computer Vision, Graphics, and Image Processing, 18:236–258, March 1982.

    Google Scholar 

  15. Frieder Kuhnert. Pseudoinverse Matrizen und die Methode der Regularisierung. Teubner, st edition 1976.

    Google Scholar 

  16. E. H. Moore. On the reciprocal of the general algebraic matrix (abstract). Bulletin of the American Mathematical Society, 26:394–395, 1920.

    Google Scholar 

  17. E. N. Mortensen, B. Morse, and W. A. Barret. Adaptive boundary detection using live-wire two-dimensional dynamic programming. Computers in Cardiology, pages 635–638, October 1992.

    Google Scholar 

  18. R. Penrose. A generalized inverse for matrices. Proceedings Cambridge Philosophical Society, 51:406–413, 1955.

    MathSciNet  CrossRef  MATH  Google Scholar 

  19. R. Penrose. On best approximate solutions of linear matrix equations. Proceedings Cambridge Philosophical Society, 52:17–19, 1956.

    MathSciNet  CrossRef  MATH  Google Scholar 

  20. E. Persoon and K. S. Fu. Shape discrimination using Fourier descriptors. IEEE Transactions on Systems, Man and Cybernetics, 7(3):170–179, March 1977.

    MathSciNet  CrossRef  Google Scholar 

  21. Lawrence H. Staib and James S. Duncan. Boundary finding with parametrically deformable models. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 14(11):1061–1075, November 1992.

    Google Scholar 

  22. Gábor Székely, András Kelemen, Christian Brechbühler, and Guido Gerig. Segmentation of 2-D and 3-D objects from MRI volume data using constrained elastic deformations of flexible Fourier contour and surface models. Medical Image Analysis, 1(1):19–34, 1996.

    CrossRef  Google Scholar 

  23. A. Yuille, D. Cohen, and P. Hallinan. Feature extraction from faces using deformable templates. In Proceedings Conference on Computer Vision and Pattern Recognition, pages 104–109, 1989.

    Google Scholar 

  24. Denis Zorin, Peter Schröder, and Wim Sweldens. Interactive multiresolution mesh editing. In Computer Graphics Proceedings SIGGRAPH 1997, pages 259–268. Addison-Wesley, 1997.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and Permissions

Copyright information

© 2000 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Hug, J., Brechbühler, C., Székely, G. (2000). Model-Based Initialisation for Segmentation. In: Vernon, D. (eds) Computer Vision — ECCV 2000. ECCV 2000. Lecture Notes in Computer Science, vol 1843. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45053-X_19

Download citation

  • DOI: https://doi.org/10.1007/3-540-45053-X_19

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-67686-7

  • Online ISBN: 978-3-540-45053-5

  • eBook Packages: Springer Book Archive

We’re sorry, something doesn't seem to be working properly.

Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.