Advertisement

Canonical Correlation Analysis on Riemannian Manifolds and Its Applications

  • Hyunwoo J. Kim
  • Nagesh Adluru
  • Barbara B. Bendlin
  • Sterling C. Johnson
  • Baba C. Vemuri
  • Vikas Singh
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8690)

Abstract

Canonical correlation analysis (CCA) is a widely used statistical technique to capture correlations between two sets of multi-variate random variables and has found a multitude of applications in computer vision, medical imaging and machine learning. The classical formulation assumes that the data live in a pair of vector spaces which makes its use in certain important scientific domains problematic. For instance, the set of symmetric positive definite matrices (SPD), rotations and probability distributions, all belong to certain curved Riemannian manifolds where vector-space operations are in general not applicable. Analyzing the space of such data via the classical versions of inference models is rather sub-optimal. But perhaps more importantly, since the algorithms do not respect the underlying geometry of the data space, it is hard to provide statistical guarantees (if any) on the results. Using the space of SPD matrices as a concrete example, this paper gives a principled generalization of the well known CCA to the Riemannian setting. Our CCA algorithm operates on the product Riemannian manifold representing SPD matrix-valued fields to identify meaningful statistical relationships on the product Riemannian manifold. As a proof of principle, we present results on an Alzheimer’s disease (AD) study where the analysis task involves identifying correlations across diffusion tensor images (DTI) and Cauchy deformation tensor fields derived from T1-weighted magnetic resonance (MR) images.

Keywords

Riemannian Manifold Canonical Correlation Analysis Parallel Transport Geodesic Curve Dictionary Learning 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Akaho, S.: A kernel method for canonical correlation analysis. In: International Meeting on Psychometric Society (2001)Google Scholar
  2. 2.
    Andrew, G., Arora, R., Bilmes, J., Livescu, K.: Deep canonical correlation analysis. In: ICML (2013)Google Scholar
  3. 3.
    Avants, B.B., Cook, P.A., Ungar, L., Gee, J.C., Grossman, M.: Dementia induces correlated reductions in white matter integrity and cortical thickness: a multivariate neuroimaging study with SCCA. Neuroimage 50(3), 1004–1016 (2010)CrossRefGoogle Scholar
  4. 4.
    Bach, F.R., Jordan, M.I.: Kernel independent component analysis. The Journal of Machine Learning Research 3, 1–48 (2003)zbMATHMathSciNetGoogle Scholar
  5. 5.
    Callaghan, P.T.: Principles of nuclear magnetic resonance microscopy. Oxford University Press (1991)Google Scholar
  6. 6.
    Chung, M., Worsley, K., Paus, T., Cherif, C., Collins, D., Giedd, J., Rapoport, J., Evans, A.: A unified statistical approach to deformation-based morphometry. NeuroImage 14(3), 595–606 (2001)CrossRefGoogle Scholar
  7. 7.
    Do Carmo, M.P.: Riemannian geometry (1992)Google Scholar
  8. 8.
    Fletcher, P.T.: Geodesic regression and the theory of least squares on riemannian manifolds. International Journal of Computer Vision 105(2), 171–185 (2013)CrossRefMathSciNetGoogle Scholar
  9. 9.
    Fletcher, P.T., Lu, C., Pizer, S.M., Joshi, S.: Principal geodesic analysis for the study of nonlinear statistics of shape. Medical Imaging 23(8), 995–1005 (2004)CrossRefGoogle Scholar
  10. 10.
    Goh, A., Lenglet, C., Thompson, P.M., Vidal, R.: A nonparametric Riemannian framework for processing high angular resolution diffusion images (HARDI), pp. 2496–2503 (2009)Google Scholar
  11. 11.
    Hardoon, D.R., Szedmak, S., Shawe-Taylor, J.: CCA: An overview with application to learning methods. Neural Computation 16(12), 2639–2664 (2004)CrossRefzbMATHGoogle Scholar
  12. 12.
    Hardoon, D.R., et al.: Unsupervised analysis of fMRI data using kernel canonical correlation. NeuroImage 37(4), 1250–1259 (2007)CrossRefGoogle Scholar
  13. 13.
    Hinkle, J., Fletcher, P.T., Joshi, S.: Intrinsic polynomials for regression on Riemannian manifolds. Journal of Mathematical Imaging and Vision, 1–21 (2014)Google Scholar
  14. 14.
    Ho, J., Xie, Y., Vemuri, B.: On a nonlinear generalization of sparse coding and dictionary learning. In: ICML, pp. 1480–1488 (2013)Google Scholar
  15. 15.
    Hotelling, H.: Relations between two sets of variates. Biometrika 28(3/4), 321–377 (1936)CrossRefzbMATHGoogle Scholar
  16. 16.
    Hsieh, W.W.: Nonlinear canonical correlation analysis by neural networks. Neural Networks 13(10), 1095–1105 (2000)CrossRefGoogle Scholar
  17. 17.
    Hua, X., Gutman, B., et al.: Accurate measurement of brain changes in longitudinal MRI scans using tensor-based morphometry. Neuroimage 57(1), 5–14 (2011)CrossRefGoogle Scholar
  18. 18.
    Hua, X., Leow, A.D., Parikshak, N., Lee, S., Chiang, M.C., Toga, A.W., Jack Jr., C.R., Weiner, M.W., Thompson, P.M.: Tensor-based morphometry as a neuroimaging biomarker for Alzheimer’s disease: an MRI study of 676 AD, MCI, and normal subjects. Neuroimage 43(3), 458–469 (2008)CrossRefGoogle Scholar
  19. 19.
    Huang, H., He, H., Fan, X., Zhang, J.: Super-resolution of human face image using canonical correlation analysis. Pattern Recognition 43(7), 2532–2543 (2010)CrossRefzbMATHGoogle Scholar
  20. 20.
    Huckemann, S., Hotz, T., Munk, A.: Intrinsic shape analysis: Geodesic PCA for Riemannian manifolds modulo isometric Lie group actions. Statistica Sinica 20, 1–100 (2010)zbMATHMathSciNetGoogle Scholar
  21. 21.
    Jayasumana, S., Hartley, R., Salzmann, M., Li, H., Harandi, M.: Kernel methods on the Riemannian manifold of symmetric positive definite matrices. In: CVPR, pp. 73–80 (2013)Google Scholar
  22. 22.
    Karcher, H.: Riemannian center of mass and mollifier smoothing. Communications on Pure and Applied Mathematics 30(5), 509–541 (1977)CrossRefzbMATHMathSciNetGoogle Scholar
  23. 23.
    Kim, H.J., Adluru, N., Collins, M.D., Chung, M.K., Bendlin, B.B., Johnson, S.C., Davidson, R.J., Singh, V.: Multivariate general linear models (MGLM) on Riemannian manifolds with applications to statistical analysis of diffusion weighted images. In: CVPR (2014)Google Scholar
  24. 24.
    Kim, T.K., Cipolla, R.: CCA of video volume tensors for action categorization and detection. PAMI 31(8), 1415–1428 (2009)CrossRefGoogle Scholar
  25. 25.
    Lai, P.L., Fyfe, C.: A neural implementation of canonical correlation analysis. Neural Networks 12(10), 1391–1397 (1999)CrossRefGoogle Scholar
  26. 26.
    Lebanon, G., et al.: Riemannian geometry and statistical machine learning. Ph.D. thesis, Carnegie Mellon University, Language Technologies Institute, School of Computer Science (2005)Google Scholar
  27. 27.
    Li, P., Wang, Q., Zuo, W., Zhang, L.: Log-Euclidean kernels for sparse representation and dictionary learning. In: ICCV, pp. 1601–1608 (2013)Google Scholar
  28. 28.
    Moakher, M.: A differential geometric approach to the geometric mean of symmetric positive-definite matrices. SIAM Journal on Matrix Analysis and Applications 26(3), 735–747 (2005)CrossRefzbMATHMathSciNetGoogle Scholar
  29. 29.
    Niethammer, M., Huang, Y., Vialard, F.X.: Geodesic regression for image time-series. In: MICCAI, pp. 655–662 (2011)CrossRefGoogle Scholar
  30. 30.
    Said, S., Courty, N., Le Bihan, N., Sangwine, S.J., et al.: Exact principal geodesic analysis for data on SO(3). In: Proceedings of the 15th European Signal Processing Conference, pp. 1700–1705 (2007)Google Scholar
  31. 31.
    Shi, X., Styner, M., Lieberman, J., Ibrahim, J.G., Lin, W., Zhu, H.: Intrinsic regression models for manifold-valued data. In: Yang, G.-Z., Hawkes, D., Rueckert, D., Noble, A., Taylor, C. (eds.) MICCAI 2009, Part II. LNCS, vol. 5762, pp. 192–199. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  32. 32.
    Sommer, S.: Horizontal dimensionality reduction and iterated frame bundle development. In: Nielsen, F., Barbaresco, F. (eds.) GSI 2013. LNCS, vol. 8085, pp. 76–83. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  33. 33.
    Sommer, S., Lauze, F., Nielsen, M.: Optimization over geodesics for exact principal geodesic analysis. Advances in Computational Mathematics, 1–31Google Scholar
  34. 34.
    Steinke, F., Hein, M., Schölkopf, B.: Nonparametric regression between general Riemannian manifolds. SIAM Journal on Imaging Sciences 3(3), 527–563 (2010)CrossRefzbMATHMathSciNetGoogle Scholar
  35. 35.
    Witten, D.M., Tibshirani, R., Hastie, T.: A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. Biostatistics 10(3), 515–534 (2009)CrossRefGoogle Scholar
  36. 36.
    Xie, Y., Vemuri, B.C., Ho, J.: Statistical analysis of tensor fields. In: Jiang, T., Navab, N., Pluim, J.P.W., Viergever, M.A. (eds.) MICCAI 2010, Part I. LNCS, vol. 6361, pp. 682–689. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  37. 37.
    Yger, F., Berar, M., Gasso, G., Rakotomamonjy, A.: Adaptive canonical correlation analysis based on matrix manifolds. In: ICML (2012)Google Scholar
  38. 38.
    Yu, S., Tan, T., Huang, K., Jia, K., Wu, X.: A study on gait-based gender classification. IEEE Transactions on Image Processing 18(8), 1905–1910 (2009)CrossRefMathSciNetGoogle Scholar
  39. 39.
    Zhu, H., Chen, Y., Ibrahim, J.G., Li, Y., Hall, C., Lin, W.: Intrinsic regression models for positive-definite matrices with applications to diffusion tensor imaging. Journal of the American Statistical Association 104(487) (2009)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Hyunwoo J. Kim
    • 1
  • Nagesh Adluru
    • 1
  • Barbara B. Bendlin
    • 1
  • Sterling C. Johnson
    • 1
  • Baba C. Vemuri
    • 2
  • Vikas Singh
    • 1
  1. 1.University of Wisconsin-MadisonUSA
  2. 2.University of FloridaUSA

Personalised recommendations