Methods Inf Med 2007; 46(03): 275-281
DOI: 10.1160/ME9043
paper
Schattauer GmbH

Automatic Generation of 3D Statistical Shape Models with Optimal Landmark Distributions

T. Heimann
1   Division of Medical and Biological Informatics, German Cancer Research Center, Heidelberg, Germany
,
I. Wolf
1   Division of Medical and Biological Informatics, German Cancer Research Center, Heidelberg, Germany
,
H.-P. Meinzer
1   Division of Medical and Biological Informatics, German Cancer Research Center, Heidelberg, Germany
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Publikationsdatum:
20. Januar 2018 (online)

Summary

Objectives: To point out the problem of non-uniform landmark placement in statistical shape modeling, to present an improved method for generating landmarks in the 3D case and to propose an unbiased evaluation metric to determine model quality.

Methods: Our approach minimizes a cost function based on the minimum description length (MDL) of the shape model to optimize landmark correspondences over the training set. In addition to the standard technique, we employ an extended remeshing method to change the landmark distribution without losing correspondences, thus ensuring a uniform distribution over all training samples. To breakthe dependency of the established evaluation measures generalization and specificity from the landmark distribution, we change the internal metric from landmark distance to volumetric overlap.

Results: Redistributing landmarks to an equally spaced distribution during the model construction phase improves the quality of the resulting models significantly if the shapes feature prominent bulges or other complex geometry.

Conclusions: The distribution of landmarks on the training shapes is – beyond the correspondence issue – a crucial point in model construction.

 
  • References

  • 1 Cootes TF, Taylor CJ, Cooper DH, Graham J. Active Shape Models – their training and application. Comput Vis Image Underst 1995; 61 (01) 38-59.
  • 2 Frangi AF, Rueckert D, Schnabel JA, Niessen WJ. Automatic construction of multiple-object three-dimensional statistical shape models: Application to cardiac modelling. IEEE Trans Med Imaging 2002; 21 (09) 1151-1166.
  • 3 Paulsen RR, Hilger KB. Shape modelling using Markov random field restoration of point correspondences. Taylor CJ, Noble JA. Information Processing in Medical Imaging, Proceedings of the 18th International Conference (IPMI). July 20-25, 2003 Ambleside, UK.: Springer; 2003: 1-12.
  • 4 Kaus MR, Pekar V, Lorenz C, Truyen R, Lobregt S, Weese J. Automated 3-D PDM construction from segmented images using deformable models. IEEE Trans Med Imaging 2003; 22 (08) 1005-1013.
  • 5 Brechbühler C, Gerig G, Kübler O. Parameterization of closed surfaces for 3-D shape description. Comput Vis Image Underst 1995; 61 (02) 154-170.
  • 6 Styner M, Rajamani KT, Nolte LP, Zsemlye G, Székely G, Taylor CJ, Davies RH. Evaluation of 3D correspondence method for model building. Taylor CJ, Noble JA. Information Processing in Medical Imaging, Proceedings of the 18th International Conference (IPMI). Jul 20-25, 2003 Ambleside, UK.: Springer; 2003: 63-75.
  • 7 Davies RH, Twining CJ, Cootes TF, Waterton JC, Taylor CJ. 3D statistical shape models using direct optimisation of description length. Heyden A, Sparr G, Nielsen M, Johansen P. Computer Vision-ECCV 2002, Proceedings of the 7th European Conference on Computer Vision, Part III. May 28-31, 2002 Copenhagen, Denmark.: Springer; 2002: 3-20.
  • 8 Heimann T, Wolf I, Meinzer H-P. Optimal landmark distributions for statistical shape model construction. Reinhardt JM, Pluim JPW. Proceedings of the SPIE, Volume 6144 – Medical Imaging 2006: Image Processing. Feb 13, 2006 San Diego, CA, USA.: SPIE Press; 2006: 518-528.
  • 9 Floater MS, Hormann K. Surface parameterization: a tutorial and survey. Dodgson NA, Floater MS, Sabin MA. Advances in Multiresolution for Geometric Modelling.. Springer; 2005: 157-186.
  • 10 Gu X, Wang Y, Chan TF, Thompson PM, Yau ST. Genus zero surface conformal mapping and its application to brain surface mapping. Taylor CJ, Noble JA. Information Processing in Medical Imaging, Proceedings of the 18th International Conference (IPMI). Jul 20-25, 2003 Ambleside, UK.: Springer; 2003: 172-184.
  • 11 Thodberg HH. Minimum description length shape and appearance models. Taylor CJ, Noble JA. Information Processing in Medical Imaging, Proceedings of the 18 th International Conference (IPMI). Jul 20-25, 2003 Ambleside, UK.: Springer; 2003: 51-62.
  • 12 Heimann T, Wolf I, Williams TG, Meinzer HP. 3D Active Shape Models using gradient descent optimization of description length. Christensen GE, Sonka M. Information Processing in Medical Imaging, Proceedings of the 19th International Conference (IPMI). Jul 10-15, 2005 Glenwood Springs, CO, USA.: Springer; 2005: 566-577.
  • 13 Alliez P, Meyer M, Desbrun M. Interactive geometry remeshing. ACM Trans Graphics, special issue for SIGGRAPH conference 2002; 21 (03) 347-354.
  • 14 Ostromoukhov V. A simple and efficient error-diffusion algorithm. Proceedings of the 28th annual conference on Computer graphics and interactive techniques (SIGGRAPH). Aug 12-17, 2001 Los Angeles, CA, USA.: ACM Press; 2001: 567-572.
  • 15 Tanimoto TT. An elementarymathematical theory of classification and prediction. Technical report. IBM Research; 1958
  • 16 Gu X, Yau S-T. Global conformal surfaceparameterization. Kobbelt L, Schröder P, Hoppe H. Proceedings of the Eurographics Symposium on Geometry processing. Jun 23-25, 2003 Aachen, Germany.: Eurographics Association; 2003: 127-137.