Advertisement

Abstract

With the growing acceptance of nonrigid registration as a useful tool to perform clinical research, and in particular group studies, the storage space needed to hold the resulting transforms is deemed to become a concern for vector field based approaches, on top of the traditional computation time issue. In a recent study we lead, which involved the registration of more than 22,000 pairs of T1 MR volumes, this constrain appeared critical indeed. In this paper, we propose to decompose the vector field on a wavelet basis, and let the registration algorithm minimize the number of non-zero coefficients by introducing an L 1 penalty. This enables a sparse representation of the vector field which, unlike parametric representations, does not confine the estimated transform into a small parametric space with a fixed uniform smoothness : nonzero wavelet coefficients are optimally distributed depending on the data. Furthermore, we show that the iconic feature registration framework allows to embed the non-differentiable L 1 penalty into a C 1 energy that can be efficiently minimized by standard optimization techniques.

Keywords

Iterative Close Point Nonrigid Registration Iterative Close Point Spline Wavelet Gaussian Pyramid 
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.

References

  1. 1.
    Ashburner, J., Friston, K.F.: Voxel-based morphometry — The methods. NeuroImage 12, 1427–1442 (2000)Google Scholar
  2. 2.
    Collins, D.L., Evans, A.C.: ANIMAL validation and applications of nonlinear registration based segmentation. Int. J. of Pattern Recognition and Artificial Intelligence 11, 1271–1294 (1997)CrossRefGoogle Scholar
  3. 3.
    Thompson, P., Toga, A.W.: Detection, visualization and animation of abnormal anatomic structures with a deformable probabilistic atlas based on random vector field transformations. Medical Image Analysis 1, 271–294 (1997)CrossRefGoogle Scholar
  4. 4.
    Rueckert, D., Frangi, A.F., Schnabel, J.A.: Automatic construction of 3D statistical deformation models of the brain using nonrigid registration. IEEE Trans. Med. Im. 22 (2003)Google Scholar
  5. 5.
    Cathier, P., Fung, G., J.-F., M.: Population classification based on global brain features inferred from massive T1 MRI registration. In: Proc. of HBM (2006)Google Scholar
  6. 6.
    Szeliski, R., Coughlan, J.: Spline-based image registration. Int. J. of Comp. Vision 22, 199–218 (1997)CrossRefGoogle Scholar
  7. 7.
    Kybic, J., Unser, M.: Fast parametric elastic image registration. IEEE Trans. Im. Proc. 12, 1427–1442 (2003)CrossRefGoogle Scholar
  8. 8.
    Tibshirani, R.: Regression selection and shrinkage via the lasso. J. Royal Stat. Soc. 1, 267–288 (1995)Google Scholar
  9. 9.
    Ng, L., Solo, V.: Optical flow estimation using wavelet zeroing. In: Proc. of ICIP, vol. III, pp. 722–726 (1999)Google Scholar
  10. 10.
    Cachier, P., Bardinet, E., Dormont, D., Pennec, X., Ayache, N.: Iconic feature based nonrigid registration: The PASHA algorithm. Comp. Vision Im. Underst. 89, 272–298 (2003)zbMATHCrossRefGoogle Scholar
  11. 11.
    Thirion, J.P.: Image matching as a diffusion process: an analogy with Maxwell’s demons. Medical Image Analysis 2, 243–260 (1998)CrossRefGoogle Scholar
  12. 12.
    Hayton, P.M., Brady, M., Smith, S.M., Moore, N.: A non-rigid registration algorithm for dynamic breast MR images. Artificial Intelligence 114, 124–156 (1999)CrossRefGoogle Scholar
  13. 13.
    Singh, A., Allen, P.: Image-flow computation: An estimation-theoretic framework and a unified perspective. CVGIP: Im. Underst. 56, 152–177 (1992)zbMATHCrossRefGoogle Scholar
  14. 14.
    Daubechies, I., Sweldens, W.: Factoring wavelet transforms into lifting steps. J. Fourier Anal. Appl. 3, 247–269 (1998)CrossRefMathSciNetGoogle Scholar
  15. 15.
    Mallat, S.G. (ed.): A wavelet tour of signal processing. Academic Press, London (1998)zbMATHGoogle Scholar
  16. 16.
    Wu, Y.T., Kanade, T., Li, C.C., Cohn, J.: Image registration using wavelet-based motion model. Int. J. of Comp. Vision 38, 129–152 (2000)zbMATHCrossRefGoogle Scholar
  17. 17.
    Cachier, P., Pennec, X.: 3D non-rigid registration by gradient descent on a Gaussian-windowed similarity measure using convolutions. In: Proc. of MMBIA, pp. 182–189 (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Pascal Cathier
    • 1
  1. 1.CEA, DSV, DRM, SHFJOrsayFrance

Personalised recommendations