Advertisement

Bayesian Analysis of fMRI Data with ICA Based Spatial Prior

  • Deepti R. Bathula
  • Hemant D. Tagare
  • Lawrence H. Staib
  • Xenophon Papademetris
  • Robert T. Schultz
  • James S. Duncan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5242)

Abstract

Spatial modeling is essential for fMRI analysis due to relatively high noise in the data. Earlier approaches have been primarily concerned with the spatial coherence of the BOLD response in local neighborhoods. In addition to a smoothness constraint, we propose to incorporate prior knowledge of brain activation patterns learned from training samples. This spatially informed prior can significantly enhance the estimation process by inducing sensitivity to task related regions of the brain. As fMRI data exhibits intersubject variability in functional anatomy, we design the prior using Independent Component Analysis (ICA). Due to the non-Gaussian assumption, ICA does not regress to the mean activation pattern and thus avoids suppressing intersubject differences. Results from a real fMRI experiment indicate that our approach provides statistically significant improvement in estimating activation compared to the standard general linear model (GLM) based methods.

Keywords

General Linear Model Independent Component Analysis fMRI Data Independent Component Analysis Coronal Slice 
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.
    Friston, K.J., Holmes, A.P.: Statistical parametric maps in functional imaging: A general linear approach. Human Brain Mapping 2(4), 189–210 (1994)CrossRefGoogle Scholar
  2. 2.
    Descombes, X., Kruggel, F., von Cramon, Y.: fMRI signal restoration using an edge preserving spatio-temporal Markov random field. NeuroImage 8, 340–349 (1998)CrossRefGoogle Scholar
  3. 3.
    Rajapakse, J.C., Piyaratna, J.: Bayesian approach to segmentation of statistical parametric maps. IEEE TMI 48(10), 1186–1194 (2001)Google Scholar
  4. 4.
    Woolrich, M.W., Jenkinson, M., Brady, J.M., Smith, S.M.: Fully Bayesian spatio-temporal modeling of fMRI data. IEEE TMI 23(2), 213–231 (2004)Google Scholar
  5. 5.
    Hartvig, N., Jensen, J.: Spatial mixture modeling of fMRI data. Human Brain Mapping 11(4), 233–248 (2000)CrossRefGoogle Scholar
  6. 6.
    Penny, W.D., Trujillo-Barretob, N.J., Friston, K.J.: Bayesian fMRI time series analysis with spatial priors. NeuroImage 24, 350–362 (2005)CrossRefGoogle Scholar
  7. 7.
    Flandin, G., Bayesian, W.D.P.: fMRI data analysis with sparse spatial basis function priors. NeuroImage 34, 1108–1125 (2006)CrossRefGoogle Scholar
  8. 8.
    Schultz, R.T.: Developmental deficits in social perception in autism: The role of the amygdala and fusiform face area. Intl. J. Dev. Neuroscience 23, 125–141 (2005)CrossRefGoogle Scholar
  9. 9.
    Yang, J., Papademetris, X., Staib, L.H., Duncan, J.S.: Functional brain image analysis using joint function-structure priors. In: Barillot, C., Haynor, D.R., Hellier, P. (eds.) MICCAI 2004. LNCS, vol. 3217, pp. 736–744. Springer, Heidelberg (2004)Google Scholar
  10. 10.
    Hyvrinen, A., Oja, E.: A fast fixed-point algorithm for independent component analysis. Neural Computation 9(7), 1483–1492 (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Deepti R. Bathula
    • 1
  • Hemant D. Tagare
    • 2
    • 3
  • Lawrence H. Staib
    • 2
    • 3
  • Xenophon Papademetris
    • 2
    • 3
  • Robert T. Schultz
    • 4
  • James S. Duncan
    • 1
    • 2
    • 3
  1. 1.Departments of Biomedical EngineeringUSA
  2. 2.Electrical EngineeringUSA
  3. 3.Diagnostic RadiologyUSA
  4. 4.Child Study CenterYale UniversityNew HavenUSA

Personalised recommendations