Advertisement

On the Choice of a Tensor Distance for DTI White Matter Segmentation

  • Rodrigo de Luis-García
  • Carlos Alberola-López
  • Carl-Fredrik Westin
Conference paper
Part of the Mathematics and Visualization book series (MATHVISUAL)

Abstract

The segmentation of anatomical structures within the white matter of the brain from DTI is an important task for white matter analysis, and has therefore received considerable attention in the literature during the last few years. Any segmentation method relies on the choice of a tensor dissimilarity measure, which should be small between tensors belonging to the same region and large between tensors belonging to different structures. Many different tensor distances have been proposed in the literature (Frobenius, Kullback-Leibler, Geodesic, Log-Euclidean, Hybrid) for segmentation or other purposes, and there exist reasons (either theoretical or empirical) to justify the choice of any of them. Thus, determining which is the most appropriate tensor distance for a specific segmentation problem has become an extremely difficult decision. In this chapter we present a study on different tensor dissimilarity measures and their performance for white matter segmentation. The study is based on the use of two different DTI atlases of human brain, which provide a ground truth upon which the distances can be fairly compared. In order for the comparison to be independent of the segmentation method employed, it has been performed in terms of the separability of the different clases. Results show the Hybrid distance to perform better than other traditional tensor dissimilarity measures in terms of separability between classes, while the Frobenius, Kullback-Leibler, Geodesic and Log-Euclidean distances perform similarly.

Keywords

White Matter Corpus Callosum Diffusion Tensor Geodesic Distance Dissimilarity Measure 
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.

Notes

Acknowledgements

The authors acknowledge Junta de Castilla y León for grant VA0339A10-2 and grant belonging to order SAN/103/2011,6 and Ministerio de Ciencia e Innovación for grant TEC2010-17982. The first author was funded, at the time part of this research was carried out, by the Spanish MEC/Fulbright Commission grant 2007–1238. This is also acknowledged.

References

  1. 1.
    Arsigny, V., Fillard, P., Pennec, X., Ayache, N.: Fast and simple calculus on tensors in the Log-Euclidean framework. In: Proceedings of the MICCAI’05, Volume 3749 of Lecture Notes in Computer Science, Palm Springs, CA, USA, pp. 115–122, Oct 2005Google Scholar
  2. 2.
    Arsigny, V., Fillard, P., Pennec, X., Ayache, N.: Log-Euclidean metrics for fast and simple calculus on diffusion tensors. Magn. Reson. Med. 56(2), 411–421 (2006)Google Scholar
  3. 3.
    Atkinson, C., Mitchell, A.: Rao’s distance measure. Sankhya 43(A), 345–365 (1981)Google Scholar
  4. 4.
    Awate, S.P., Gee, J.C.: A fuzzy, nonparametric segmentation framework for DTI and MRI analysis. In: Proceedings of Information Processing in Medical Imaging (IPMI), Kerkrade, The Netherlands (2007)Google Scholar
  5. 5.
    Awate, S.P., Zhang, H., Gee, J.C.: A fuzzy, nonparametric segmentation framework for DTI and MRI analysis: with applications to DTI-tract extraction. IEEE Trans. Med. Imaging 16(1 \(\frac{1} {4}\)), 1525–1536 (2007)Google Scholar
  6. 6.
    Basser, P., Mattiello, J., Bihan, D.L.: MR diffusion tensor spectroscopy and imaging. Biophysica 66, 259–267 (1994)Google Scholar
  7. 7.
    Le Bihan, D., Breton, E., Lallemand, D., Grenier, P., Cabanis, E., Laval-Jeantet, M.: Mr imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 161(2), 401–407 (1986)Google Scholar
  8. 8.
    Brox, T., Rousson, M., Deriche, R., Weickert, J.: Unsupervised segmentation incorporating colour, texture, and motion. Technical Report 4760, INRIA, Mar. 2003Google Scholar
  9. 9.
    Caselles, V., Kimmel, R., Sapiro, G.: Geodesic active contours. Int. J. Comput. Vis. 22, 61–79 (1997)Google Scholar
  10. 10.
    Chefd’hotel, C., Tschumperlé, D., Deriche, R., Faugeras, O.: Regularizing flows for constrained matrix-valued images. J. Math. Imaging Vis. 10(1–2), 147–162 (2004)Google Scholar
  11. 11.
    Cristianini, N., Shawe-Taylor, J., Kandola, J.: On kernel target alignment. Adv. Neural Inf. Process. Syst. 1(14), 367–374 (2001)Google Scholar
  12. 12.
    de Luis-Garcia, R., Alberola-Lopez, C.: Mixtures of Gaussians on tensor fields for the segmentation of DT-MRI. In: Proceeding of the MICCAI’07, Volume 4791 of Lecture Notes in Computer Science, Brisbane, Australia, pp. 319–326 (2007)Google Scholar
  13. 13.
    de Luis-Garcia, R., Alberola-Lopez, C., Kindlmann, G., Westin, C.-F.: Automatic segmentation of white matter structures from DTI using tensor invariants and tensor orientation. In: Proceedings of the ISMRM Scientific Meeting, Honolulu, HI, USA, April 2009Google Scholar
  14. 14.
    Duan, Y., Li, X., Xi, Y.: Thalamus segmentation from diffusion tensor magnetic resonance imaging. Int. J. Biom. Imaging 2007, 1–5 (2007)Google Scholar
  15. 15.
    Ennis, D.B., Kindlmann, G.: Orthogonal tensor invariants and the analysis of diffusion tensor magnetic resonance images. Magn. Reson. Med. 55(1–2), 136–146 (2006)Google Scholar
  16. 16.
    Feddern, C., Weickert, J., Burgeth, B.: Level set methods for tensor-valued images. In: Proceedings of the 9th IEEE Workshop on Variational, Geometric and Level Set Methods in Computer Vision, Nice, France, pp. 65–72, Oct. 2003Google Scholar
  17. 17.
    Graf, A.B.A., Borer, S.: Normalization in support vector machines. In: Proceedigns of the 23rd DAGM Symposium on Pattern Recognition, London, UK, pp. 277–282 (2001)Google Scholar
  18. 18.
    Horsfield, M., Jones, D.: Applications of diffusion-weighted and diffusion tensor MRI to white matter diseases-a review. NMR Biom. 15(7–8), 570–577 (2002)Google Scholar
  19. 19.
    Grandlund, G.H., Bigun, J.: Optimal orientation detection of linear symmetry. In: Proceedings of the 1st IEEE International Conference on Computer Vision, London, June 1987Google Scholar
  20. 20.
    Jonasson, L., Hagmann, P., Bresson, X., Meuli, R., Cuisenaire, O., Thiran, J.-P.: White matter mapping in DT-MRI using geometric flows. In: Proceedings. Eurocast 2003, Volume 2809 of Lecture Notes in Computer Science, pp. 585–595, Las Palmas de Gran Canaria, Spain, Feb. 2003Google Scholar
  21. 21.
    Jonasson, L.: Segmentation of diffusion weighted MRI using the level set framework. PhD dissertation, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland (2005).Google Scholar
  22. 22.
    Jonasson, L., Bresson, X., Hagmann, P., Cuisenaire, O., Meuli, R., Thiran, J.-P.: White matter fiber tract segmentation in DT-MRI using geometric flows. Med. Image Anal. 9, 223–236 (2005)Google Scholar
  23. 23.
    Kichenassamy, S, Kumar, A., Olver, P., Tannenbaum, A., Yezzi, A.: Gradient flows and geometric active contour models. In: Proceedings of the 5th International Conference on Computer Vision, Orlando, FL, USA, pp. 810–815, June 1995Google Scholar
  24. 24.
    Knutsson, H.: A tensor representation of 3-d structures. In: Proceedings 5th IEEE-ASSP and EURASIP Workshop on Multidimensional Signal Processing, Noordwijkerhout, The Netherlands, Sept. 1987Google Scholar
  25. 25.
    Kubicki, M., McCarley, R., Westin, C., Park, H., Maier, S., Kikinis, R., Jolesz, F., Shenton, M. E.: A review of diffusion tensor imaging studies in schizophrenia. J. Psychiatr. Res. 41(1–2), 15–30 (2007)Google Scholar
  26. 26.
    Lenglet, C., Rousson, M., Deriche, R.: Segmentation of 3d probability density fields by surface evolution: application to diffusion mri. In: Proceedings of the Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Saint Malo, France, Sept. 2004Google Scholar
  27. 27.
    Lenglet, C., Rousson, M., Deriche, R., Faugeras, O.: Statistics on multivariate normal distributions: a geometric approach and its application to diffusion tensor MRI. Technical Report 5242, INRIA, June 2004Google Scholar
  28. 28.
    Lenglet, C., Rousson, M., Deriche, R., Faugeras, O.: Toward segmentation of 3D probability density fields by surface evolution: application to diffusion MRI. Technical Report 5243, INRIA, June 2004Google Scholar
  29. 29.
    Lenglet, C., Rousson, M., Deriche, R.: DTI segmentation by statistical surface evolution. IEEE Trans. Med. Imaging 25(6), 685–700 (2006)Google Scholar
  30. 30.
    Lenglet, C., Rousson, M., Deriche, R.: DTI segmentation by statistical surface evolution. Technical Report 5843, INRIA, Feb. 2006Google Scholar
  31. 31.
    Lenglet, C., Rousson, M., Deriche, R., Faugeras, O.: Statistics on the manifold of multivariate normal distributions: theory and application to diffusion tensor MRI processing. J. Math. Imaging Vis. 25, 423–444 (2006)Google Scholar
  32. 32.
    Mori, S., Wakana, S., van Zijl, P.C.M., Nagae-Poetscher, L.M.: MRI Atlas of Human White Matter, 1st edn. Elsevier, Amsterdam (2005)Google Scholar
  33. 33.
    Paragios, N., Deriche, R.: Geodesic active regions: a new framework to deal with frame partition problems in computer vision. J. Vis. Commun. Image Represent. 13, 249–268 (2002)Google Scholar
  34. 34.
    Pasternak, O., Sochen, N., Basser, P.J.: The effect of metric selection on the analysis of diffusion tensor MRI data. Neuroimage 49(3), 2190–2204 (2010)Google Scholar
  35. 35.
    Peeters, T.H.J.M., Rodrigues, P.R., Vilanova, A., ter Haar Romeny, B.M.: Analysis of Distance/Similarity Measures for Diffusion Tensor Imaging. Springer, Berlin (2008)Google Scholar
  36. 36.
    Pekalaska, E., Paclik, P., Duin, R.: A generalized kernel approach to dissimilarity based classification. J. Mach. Learn. Res. 2, 175–211 (2001)Google Scholar
  37. 37.
    Peled, S., Gudbjartsson, H., Westin, C.-F., Kikinis, R., Jolesz, F.A.: Magnetic resonance imaging shows orientation and asymmetry of white matter fiber tracts. Brain Res. 780, 27–33 (1998)Google Scholar
  38. 38.
    Rao, C.R.: Information and accuracy attainable in the estimation of statistical parameters. Bull. Calcutta Math. Soc. 37, 81–91 (1945)Google Scholar
  39. 39.
    Rousson, M., Brox, T., Deriche, R.: Active unsupervised texture segmentation on a diffusion based feature space. Technical Report 4695, INRIA, Jan. 2003Google Scholar
  40. 40.
    Rousson, M., Brox, T., Deriche, R.: Active unsupervised texture segmentation on a diffusion based feature space. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Madison, Wisconsin, USA, Jun. 2003Google Scholar
  41. 41.
    Rousson, M., Lenglet, C., Deriche, R.: Level set and region based surface propagation for diffusion tensor mri segmentation. In: Proceedings of the Computer Vision Approaches to Medical Image Analysis Workshop, Prague, Czech Republic, May 2004Google Scholar
  42. 42.
    Salmenpera, T.M., Simister, R.J., Bartlett, P., Symms, M.R., Boulby, P.A., Free, S.L., Barker, G.J., Duncan, J.S.: High-resolution diffusion tensor imaging of the hippocampus in temporal lobe epilepsy. Epilepsy Res. 71, 102–106 (2006)Google Scholar
  43. 43.
    Scholkopf, B.: Learning with Kernels. MIT, Cambridge MA (2002)Google Scholar
  44. 44.
    Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press, New York (2004)Google Scholar
  45. 45.
    Sotak, C.H.: The role of diffusion tensor imaging in the evaluation of ischemic brain injury – a review. NMR Biom. 15(7–8), 561–569 (2002)Google Scholar
  46. 46.
    Sundgren, P.C., Dong, Q., Gómez-Hassan, D., Mukherji, S.K., Maly, P., Welsh, R.: Diffusion tensor imaging of the brain: review of clinical applications. Neuroradiology 46(5), 339–350, Aug. 2004Google Scholar
  47. 47.
    Wakana, S., Jiang, H., Nagae-Poetscher, M., van Zijl, P.C.M., Mori, S.: A fiber-tract based atlas of human white matter anatomy. Radiology 230, 77–87 (2004)Google Scholar
  48. 48.
    Wang, Z., Vemuri, B.: An affine invariant tensor dissimilarity measure and its applications to tensor-valued image segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Washington DC, USA, pp. 228–233 (2004)Google Scholar
  49. 49.
    Wang, Z., Vemuri, B.: Tensor field segmentation using region based active contour model. In: Proceedings of the European Conference on Computer Vision, Prague, Czech Republic, pp. 304–315 (2004)Google Scholar
  50. 50.
    Weldeselassie, Y.T., Hamarneh, G.: DT-MRI segmentation using graph cuts. In: Pluim, J.P.W., Reinhardt, J.M. (eds.) Proceedings of Medical Imaging, SPIE, San Diego, CA, USA, Feb. 2007Google Scholar
  51. 51.
    Wiegell, M.R., Tuch, D.S., Larsson, H.B.W., Wedeen, V.J.: Automatic segmentation of thalamic nuclei from diffusion tensor magnetic resonance imaging. NeuroImage 19, 391–401 (2003)Google Scholar
  52. 52.
    Zhukov, L., Museth, K., Breen, D., Whitaker, R., Barr, A.H.: Level set segmentation and modeling of dt-mri human brain data. J. Electron. Imaging 12(1), 125–133 (2003)Google Scholar
  53. 53.
    Ziyan, U., Tuch, D., Westin, C.-F.: Segmentation of thalamic nuclei from DTI using spectral clustering. In: Proceedings of the MICCAI’06, Volume 4191 of Lecture Notes in Computer Science, Copenhagen, Denmark, pp. 807–814 (2006)Google Scholar

Copyright information

© Springer Berlin Heidelberg 2012

Authors and Affiliations

  • Rodrigo de Luis-García
    • 1
  • Carlos Alberola-López
    • 2
  • Carl-Fredrik Westin
    • 1
  1. 1.Laboratory of Mathematics in ImagingHarvard Medical SchoolBostonUSA
  2. 2.Laboratorio de Procesado de ImagenUniversidad de ValladolidValladolidSpain

Personalised recommendations