Robust and Accurate Appearance Models Based on Joint Dictionary Learning Data from the Osteoarthritis Initiative

Data from the Osteoarthritis Initiative
  • Anirban Mukhopadhyay
  • Oscar Salvador Morillo Victoria
  • Stefan Zachow
  • Hans Lamecker
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9993)

Abstract

Deformable model-based approaches to 3D image segmentation have been shown to be highly successful. Such methodology requires an appearance model that drives the deformation of a geometric model to the image data. Appearance models are usually either created heuristically or through supervised learning. Heuristic methods have been shown to work effectively in many applications but are hard to transfer from one application (imaging modality/anatomical structure) to another. On the contrary, supervised learning approaches can learn patterns from a collection of annotated training data. In this work, we show that the supervised joint dictionary learning technique is capable of overcoming the traditional drawbacks of the heuristic approaches. Our evaluation based on two different applications (liver/CT and knee/MR) reveals that our approach generates appearance models, which can be used effectively and efficiently in a deformable model-based segmentation framework.

Keywords

Dictionary learning Appearance model Liver CT Knee MR 3D segmentation 

References

  1. 1.
    Aharon, M., et al.: K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. TSP 54, 4311–4322 (2006)Google Scholar
  2. 2.
    Belhumeur, P.N., et al.: Localizing parts of faces using a consensus of exemplars. In: CVPR (2011)Google Scholar
  3. 3.
    Cootes, T.F., et al.: Active shape models - their training and application. CVIU 61, 38–59 (1995)Google Scholar
  4. 4.
    Cristinacce, D., et al.: Automatic feature localisation with constrained local models. J. Pattern Recogn. 41, 3054–3067 (2008)CrossRefMATHGoogle Scholar
  5. 5.
    Heimann, T., et al.: Statistical shape models for 3D medical image segmentation: a review. MIA 13, 543–563 (2009)Google Scholar
  6. 6.
    Kainmueller, D., et al.: Shape constrained automatic segmentation of the liver basedon a heuristic intensity model. In: MICCAI Workshop 3D Segmentation in the Clinic: A Grand Challenge (2007)Google Scholar
  7. 7.
    Kainmueller, D., et al.: An articulated statistical shape model for accurate hip joint segmentation. In: EMBC (2009)Google Scholar
  8. 8.
    Lindner, C., et al.: Robust and accurate shape model matching using random forest regression-voting. PAMI 37, 1862–1874 (2015)CrossRefGoogle Scholar
  9. 9.
    Mukhopadhyay, A., Oksuz, I., Bevilacqua, M., Dharmakumar, R., Tsaftaris, S.A.: Data-driven feature learning for myocardial segmentation of CP-BOLD MRI. In: van Assen, H., Bovendeerd, P., Delhaas, T. (eds.) FIMH 2015. LNCS, vol. 9126, pp. 189–197. Springer, Heidelberg (2015)CrossRefGoogle Scholar
  10. 10.
  11. 11.
    Saragih, J.M., et al.: Deformable model fitting by regularized landmark mean-shift. IJCV 91, 200–215 (2011)MathSciNetCrossRefMATHGoogle Scholar
  12. 12.
    Toews, M., et al.: Efficient and robust model-to-image alignment using 3D scale-invariant features. MIA 17, 271–282 (2013)Google Scholar
  13. 13.
    Tropp, J.A., et al.: Signal recovery from random measurements via orthogonal matching pursuit. Trans. Inf. Theor. 53, 4655–4666 (2007)MathSciNetCrossRefMATHGoogle Scholar
  14. 14.
    Zhang, S., et al.: Deformable segmentation via sparse representation and dictionary learning. MIA 16, 1385–1396 (2012)Google Scholar

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  • Anirban Mukhopadhyay
    • 1
  • Oscar Salvador Morillo Victoria
    • 2
  • Stefan Zachow
    • 1
    • 2
  • Hans Lamecker
    • 1
    • 2
  1. 1.Zuse Institute BerlinBerlinGermany
  2. 2.1000 Shapes Gmbh.BerlinGermany

Personalised recommendations