Variational Bayesian Learning of Sparse Representations and Its Application in Functional Neuroimaging

  • Evangelos Roussos
  • Steven Roberts
  • Ingrid Daubechies
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7263)


Recent theoretical and experimental work in imaging neuroscience reveals that activations inferred from functional MRI data have sparse structure. We view sparse representation as a problem in Bayesian inference, following a machine learning approach, and construct a structured generative latent-variable model employing adaptive sparsity-inducing priors. The construction allows for automatic complexity control and regularization as well as denoising. Experimental results with benchmark datasets show that the proposed algorithm outperforms standard tools for model-free decompositions such as independent component analysis.


Sparse representations variational Bayesian learning hierarchical generative models complexity control wavelets fMRI 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Worsley, K., Friston, K.: Analysis of fMRI time series revisited—again. NeuroImage 2, 173–181 (1995)CrossRefGoogle Scholar
  2. 2.
    Beckmann, C., Smith, S.: Probabilistic independent component analysis for functional magnetic resonance imaging. IEEE Trans. Med. Imag. 23, 137–152 (2004)CrossRefGoogle Scholar
  3. 3.
    McKeown, M., Sejnowski, T.: Independent component analysis of fMRI data: examining the assumptions. Hum. Brain Mapp. 6, 5–6, 368–372 (1998)CrossRefGoogle Scholar
  4. 4.
    Daubechies, I., Roussos, E., Takerkart, S., Benharrosh, M., Golden, C., D’Ardenne, K., Richter, W., Cohen, J., Haxby, J.: Independent component analysis for brain fMRI does not select for independence. PNAS 106(26), 10415–10422 (2009)CrossRefGoogle Scholar
  5. 5.
    Carroll, M., Cecchi, G., Rish, I., Garg, R., Rao, A.: Prediction and interpretation of distributed neural activity with sparse models. NeuroImage 44(1), 112–122 (2009)CrossRefGoogle Scholar
  6. 6.
    Li, Y., Cichocki, A., Amari, S.-I., Shishkin, S., Cao, J., Gu, F.: Sparse representation and its applications in blind source separation. In: Proceedings of the Annual Conference on Neural Information Processing Systems 17 (2003)Google Scholar
  7. 7.
    Turkheimer, F., Brett, M., Aston, J., Leff, A., Sargent, P., Wise, R., Grasby, P., Cunningham, V.: Statistical modelling of PET images in wavelet space. Journal of Cerebral Blood Flow and Metabolism 20, 1610–1618 (2001)Google Scholar
  8. 8.
    Bullmore, E., Fadili, J., Breakspear, M., Salvador, R., Suckling, J., Brammer, M.: Wavelets and statistical analysis of functional magnetic resonance images of the human brain. Statistical Methods in Medical Research 12, 375–399 (2003)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    MacKay, D.: Probable Networks and Plausible Predictions - A Review of Practical Bayesian Methods for Supervised Neural Networks. Network: Computation in Neural Systems 6, 469–505 (1995)CrossRefzbMATHGoogle Scholar
  10. 10.
    Attias, H.: A Variational Bayesian Framework for Graphical Models. In: Proceedings of Advances in Neural Information Processing Systems, vol. 12 (2000)Google Scholar
  11. 11.
    Penny, W., Roberts, S.: Variational Bayes for 1-dimensional Mixture Models. Techn. Rep. PARG–00–2, Dept. of Engineering Science, University of Oxford (2000)Google Scholar
  12. 12.
    Smith, S., Jenkinson, M., Woolrich, M., Beckmann, C., Behrens, T., Johansen-Berg, H., Bannister, P., De Luca, M., Drobnjak, I., Flitney, D., Niazy, R., Saunders, J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J., Matthews, P.: Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage 23(S1), 208–219 (2004)CrossRefGoogle Scholar
  13. 13.
    Choudrey, R., Penny, W., Roberts, S.: An Ensemble Learning Approach to Independent Component Analysis. In: Proceedings of Neural Networks for Signal Processing (2000)Google Scholar
  14. 14.
    Roussos, E., Roberts, S., Daubechies, I.: Variational Bayesian Learning for Wavelet Independent Component Analysis. In: 25th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, vol. 25 (2005)Google Scholar
  15. 15.
    Groves, A.: Bayesian Learning Methods for Modelling Functional MRI. D.Phil. Thesis, Department of Clinical Neurology, University of Oxford (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Evangelos Roussos
    • 1
  • Steven Roberts
    • 1
  • Ingrid Daubechies
    • 2
  1. 1.Dept. of Engineering ScienceUniversity of OxfordOxfordUK
  2. 2.Dept. of MathematicsDuke UniversityDurhamUSA

Personalised recommendations