Advertisement

Comparative Evaluation of Regression Methods for 3D-2D Image Registration

  • Ana Isabel Rodrigues Gouveia
  • Coert Metz
  • Luís Freire
  • Stefan Klein
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7553)

Abstract

We perform a comparative evaluation of different regression techniques for 3D-2D registration-by-regression. In registration-by-regression, image registration is treated as a nonlinear regression problem that relates image features of 2D projection images to the transformation parameters of the 3D image. In this work, we evaluate seven regression methods: Multiple Linear and Polynomial Regression (LR and PR), k-Nearest Neighbour (k-NN), Multiple Layer Perceptron with conjugate gradient optimization (MLP-CG) and with Levenberg-Marquardt optimization (MLP-LM), Radial Basis Function network (RBF) and Support Vector Regression (SVR). The experiments are performed using simulated X-ray images (DRRs) of nine coronary vessel trees, allowing us to compute the mean target registration error (mTRE) to the ground truth. All methods were robust to large initial misalignment and the highest accuracy was achieved using MLP-LM and RBF.

Keywords

3D-2D image registration regression Multiple Layer Perceptron Radial Basis Function network Support Vector Regression coronary arteries 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Markelj, P., et al.: A review of 3D/2D registration methods for image-guided interventions. Medical Image Analysis (2010)Google Scholar
  2. 2.
    van de Kraats, E.B., et al.: Standardized evaluation methodology for 2D – 3D registration. IEEE Transactions on Medical Imaging 24, 1177–1189 (2005)CrossRefGoogle Scholar
  3. 3.
    Gouveia, A.R., et al.: 3D-2D Image registration by nonlinear regression. In: IEEE-ISBI (in press, 2012)Google Scholar
  4. 4.
    Freire, L.C., Gouveia, A.R., Godinho, F.M.: A Neural Network-Based Method for Affine 3D Registration of FMRI Time Series Using Fourier Space Subsets. In: Diamantaras, K., Duch, W., Iliadis, L.S. (eds.) ICANN 2010, Part I. LNCS, vol. 6352, pp. 22–31. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  5. 5.
    Zhang, J., et al.: Rapid surface registration of 3D volumes using a neural network approach. Image and Vision Computing 26, 201–210 (2008)CrossRefGoogle Scholar
  6. 6.
    Wachowiak, M.P., et al.: A supervised learning approach to landmark-based elastic biomedical image registration and interpolation. In: IEEE – IJCNN 2002, pp. 1625–1630 (2002)Google Scholar
  7. 7.
    Qi, W., et al.: Effective 2D-3D medical image registration using Support Vector Machine. In: IEEE - EMBS 2008, pp. 5386–5389 (2008)Google Scholar
  8. 8.
    Banks, S., et al.: Accurate measurement of three-dimensional knee replacement kinematics using single-plane fluoroscopy. IEEE TBE 43, 638–649 (1996)Google Scholar
  9. 9.
    Hastie, T., et al.: The elements of statistical learning: data mining, inference and prediction. Springer (2009)Google Scholar
  10. 10.
    Haykin, S.: Neural networks: a comprehensive foundation. Prentice Hall, Delhi (1999)zbMATHGoogle Scholar
  11. 11.
    Hagan, M.: Training feedforward networks with the Marquardt algorithm. IEEE Transactions on Neural Networks 5, 2–6 (1994)CrossRefGoogle Scholar
  12. 12.
    Sarle, W.: Neural network frequently asked questions (2005), ftp://ftp.sas.com/pub/neural/FAQ.html
  13. 13.
    Chen, S., et al.: Orthogonal least squares learning algorithm for radial basis function networks. IEEE Transactions on Neural Networks 2, 302–309 (1991)CrossRefGoogle Scholar
  14. 14.
    Basak, D., et al.: Support Vector Regression. Neural Information Processing – Letters and Reviews 11, 207–224 (2007)Google Scholar
  15. 15.
    Vapnik, V.N.: The nature of statistical learning theory (1995)Google Scholar
  16. 16.
    Smola, A.J., Schölkopf, B.: A tutorial on support vector regression. Statistics and Computing 14, 199–222 (2004)CrossRefMathSciNetGoogle Scholar
  17. 17.
    Metz, C.T., et al.: Alignment of 4D Coronary CTA with Monoplane X-Ray Angiography. In: MICCAI 2011 (2011), http://www.bigr.nl/publication/694
  18. 18.
    Metz, et al.: GPU accelerated alignment of 3-D CTA with 2-D X-ray data for improved guidance in coronary interventions. In: IEEE-ISBI (2009)Google Scholar
  19. 19.
    Fitzpatrick, J.M., West, J.B.: The distribution of target registration error in rigid-body point-based registration. IEEE Transactions on Medical Imaging 20, 917–927 (2001)CrossRefGoogle Scholar
  20. 20.
    Cherkassky, V., Ma, Y.: Practical selection of SVM parameters and noise estimation for SVM regression. Neural Networks: the Official Journal of the INNS 17, 113–126 (2004)CrossRefzbMATHGoogle Scholar
  21. 21.
    Hsu, C., et al.: A Practical Guide to Support Vector Classification. Bioinformatics 1, 1–16 (2010)Google Scholar
  22. 22.
    Chang, C., et al.: LIBSVM: A Library for Support Vector Machines. Science, 1–39 (2011)Google Scholar
  23. 23.
    Bi, J., et al.: Regression error characteristic curves. In: ICML (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Ana Isabel Rodrigues Gouveia
    • 1
    • 2
  • Coert Metz
    • 3
  • Luís Freire
    • 4
  • Stefan Klein
    • 3
  1. 1.CICS-UBI – Health Sciences Research CentreUniversity of Beira InteriorCovilhãPortugal
  2. 2.Institute of Biophysics and Biomedical EngineeringUniversity of LisbonLisbonPortugal
  3. 3.Depts. of Medical Informatics & Radiology, Erasmus MCBiomedical Imaging Group RotterdamRotterdamThe Netherlands
  4. 4.Escola Superior de Tecnologia da Saúde de LisboaInstituto Politécnico de LisboaLisbonPortugal

Personalised recommendations