Abstract
Statistical shape models show considerable promise as a basis for segmenting and interpreting images. One of the drawbacks of the approach is, however, the need to establish a set of dense correspondences between examples of similar structures, across a training set of images. Often this is achieved by locating a set of ‘landmarks’ manually on each of the training images, which is time-consuming and subjective for 2D images, and almost impossible for 3D images. This has led to considerable interest in the problem of building a model automatically from a set of training shapes. We extend previous work that has posed this problem as one of optimising a measure of model ‘quality’ with respect to the set of correspondences. We define model ‘quality’ in terms of the information required to code the whole set of training shapes and aim to minimise this description length. We describe a scheme for representing the dense correspondence maps between the training examples and show that a minimum description length model can be obtained by stochastic optimisation. Results are given for several different training sets of 2D boundaries, showing that the automatic method constructs better models than the manual landmarking approach. We also show that the method can be extended straightforwardly to 3D.
Acknowledgements
The authors would like to thank Dr. Alan Brett for his contribution to the ideas for the work in this paper. Tim Cootes is funded under an EPSRC Advanced Fellowship Grant. Rhodri Davies would like to thank the BBSRC and AstraZeneca Pharmaceuticals for their financial support.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Angenent, S., S. Haker, A. Tannenbaum and R. Kikinis: On the laplace-beltrami operator and brain surface flattening. IEEE Trans. Medical Imaging, 1999. 18: p. 700–711.
Baumberg, A. and D. Hogg, Learning Flexible Models from Image Sequences, in European Conference on Computer Vision, Stockholm, Sweden. 1994. p. 299–308.
Benayoun, A., N. Ayache, and I. Cohen. Adaptive meshes and nonrigid motion computation. in International Conference on Pattern Recognition. 1994. Jerusalem, Israel.
Bookstein, F.L., Landmark methods for forms without landmarks: morphometrics of group differences in outline shape. Medical Image Analysis, 1997. 1(3): p. 225–243.
Cootes, T., A. Hill, C. Taylor, and J. Haslam, The use of Active shape models for locating structures in medical images. Image and Vision Computing, 1994. 12: p. 355–366.
Cootes, T., C. Taylor, D. Cooper and J. Graham, Active shape models-their training and application. Computer Vision and Image Understanding, 1995. 61: p. 38–59.
Geiger, B., Three-dimensional modelling of human organs and its application to diagnosis and surgical planning,. 1993, Technical Report, INRIA, France.
Goldberg, D.E., Genetic Algorithms in Search, Optimisation and Machine Learning. 1989: Addison Wesley.
Goodall, C., Procrustes Methods in the Statistical Analysis of Shape. Journal of the Royal Statistical Society, 1991. 53(2): p. 285–339.
Hill, A. and C. Taylor. Automatic landmark generation for point distribution models. in British Machine Vision Conference. 1994. Birmingham, England: BMVA Press.
Hill, A. and C.J. Taylor, A framework for automatic landmark identification using a new method of non-rigid correspondence. IEEE Transactions on Pattern Analysis and Machine Intelligence, April, 2000.
Hurdal, M. K, P. L. Bowers, K. Stephenson, D. W. L. Sumners, K. Rehm, K. Schaper, D. A. Rottenberg, Quasi-conformally flat mapping the human cerebellum, MICCAI’99, p. 279–286
Kambhamettu, C. and D.B. Goldgof, Point Correspondence Recovery in Non-rigid Motion, in IEEE Conference on Computer Vision and Pattern Recognition. 1992. p. 222–227.
Kelemen, A., G. Szekely, and G. Gerig, Elastic model-based segmentation of 3-D neuroradiological data sets. IEEE Transactions On Medical Imaging, 1999. 18(10): p. 828–839.
Kirkpatrick, S., C. Gelatt, and M. Vecchi, Optimization by Simulated Annealing. Science, 1983. 220: p. 671–680.
Kotcheff, A.C.W. and C.J. Taylor, Automatic Construction of Eigenshape Models by Direct Optimisation. Medical Image Analysis, 1998. 2: p. 303–314.
Rangarajan, A., H. Chui and F. L. Bookstein,The Softassign Procrustes Matching Algorithm, in 15th IPMI 1997. p. 29–42.
Therrien, C.W., Decision Estimation and Classification. 1989: John Whiley and Sons.
Wang, Y., B. S. Peterson, and L. H. Staib. Shape-based 3D surface correspondence using geodesics and local geometry. CVPR 2000, v. 2: p. 644-51.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Davies, R.H., Cootes, T.F., Taylor, C.J. (2001). A Minimum Description Length Approach to Statistical Shape Modelling. In: Insana, M.F., Leahy, R.M. (eds) Information Processing in Medical Imaging. IPMI 2001. Lecture Notes in Computer Science, vol 2082. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45729-1_5
Download citation
DOI: https://doi.org/10.1007/3-540-45729-1_5
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42245-7
Online ISBN: 978-3-540-45729-9
eBook Packages: Springer Book Archive