GPSSI: Gaussian Process for Sampling Segmentations of Images

  • Matthieu Lê
  • Jan Unkelbach
  • Nicholas Ayache
  • Hervé Delingette
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9351)


Medical image segmentation is often a prerequisite for clinical applications. As an ill-posed problem, it leads to uncertain estimations of the region of interest which may have a significant impact on downstream applications, such as therapy planning. To quantify the uncertainty related to image segmentations, a classical approach is to measure the effect of using various plausible segmentations. In this paper, a method for producing such image segmentation samples from a single expert segmentation is introduced. A probability distribution of image segmentation boundaries is defined as a Gaussian process, which leads to segmentations that are spatially coherent and consistent with the presence of salient borders in the image. The proposed approach outperforms previous generative segmentation approaches, and segmentation samples can be generated efficiently. The sample variability is governed by a parameter which is correlated with a simple DICE score. We show how this approach can have multiple useful applications in the field of uncertainty quantification, and an illustration is provided in radiotherapy planning.


Image Segmentation Gaussian Process Discrete Fourier Transform Clinical Target Volume Gross Tumor Volume 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Fan, A.C., Fisher III, J.W., Wells III, W.M., Levitt, J.J., Willsky, A.S.: MCMC curve sampling for image segmentation. In: Ayache, N., Ourselin, S., Maeder, A. (eds.) MICCAI 2007, Part II. LNCS, vol. 4792, pp. 477–485. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  2. 2.
    Warfield, S.K., Zou, K.H., Wells, W.M.: Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation. IEEE Transactions on Medical Imaging 23(7), 903–921 (2004)CrossRefGoogle Scholar
  3. 3.
    Pohl, K.M., Fisher, J., Bouix, S., Shenton, M., McCarley, R.W., Grimson, W.E.L., Kikinis, R., Wells, W.M.: Using the logarithm of odds to define a vector space on probabilistic atlases. Medical Image Analysis 11(5), 465–477 (2007)CrossRefGoogle Scholar
  4. 4.
    Sabuncu, M.R., Yeo, B.T., Van Leemput, K., Fischl, B., Golland, P.: A generative model for image segmentation based on label fusion. IEEE Transactions on Medical Imaging 29(10), 1714–1729 (2010)CrossRefGoogle Scholar
  5. 5.
    Williams, O., Fitzgibbon, A.: Gaussian process implicit surfaces. In: Gaussian Proc. in Practice (2007)Google Scholar
  6. 6.
    Gerardo-Castro, M.P., Peynot, T., Ramos, F.: Laser-radar data fusion with gaussian process implicit surfaces. In: Corke, P., Mejias, L., Roberts, J. (eds.) The 9th Int. Conf. on Field and Service Robotics, Brisbane, Australia (2013)Google Scholar
  7. 7.
    Criminisi, A., Sharp, T., Blake, A.: GeoS: Geodesic image segmentation. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part I. LNCS, vol. 5302, pp. 99–112. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  8. 8.
    Rasmussen, C.E.: Gaussian processes for machine learning (2006)Google Scholar
  9. 9.
    Kozintsev, B.: Computations with Gaussian random fields (1999)Google Scholar
  10. 10.
    Mason, W., Del Maestro, R., Eisenstat, D., Forsyth, P., Fulton, D., Laperrière, N., Macdonald, D., Perry, J., Thiessen, B., Committee, C.G.R., et al.: Canadian recommendations for the treatment of glioblastoma multiforme. Current Oncology 14(3), 110 (2007)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Matthieu Lê
    • 1
  • Jan Unkelbach
    • 2
  • Nicholas Ayache
    • 1
  • Hervé Delingette
    • 1
  1. 1.Asclepios ProjectINRIA Sophia AntipolisValbonneFrance
  2. 2.Department of Radiation OncologyMassachusetts General Hospital and Harvard Medical SchoolBostonUSA

Personalised recommendations