Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Spatio-temporal autoregressive models defined over brain manifolds

Spatio-temporal autoregressive models defined over brain manifolds Multivariate Autoregressive time series models (MAR) are an increasingly used tool for exploring functional connectivity in Neuroimaging. They provide the framework for analyzing the Granger Causality of a given brain region on others. In this article, we shall limit our attention to linear MAR models, in which a set of matrices of autoregressive coefficients Ak (k = 1, …, p) describe the dependence of present values of the image on lagged values of its past. Methods for estimating the Ak and determining which elements that are zero are well-known and are the basis for directed measures of influence. However, to date, MAR models are limited in the number of time series they can handle, forcing the a priori selection of a (small) number of voxels or regions of interest for analysis. This ignores the full spatio-temporal nature of functional brain data which are, in fact, collections of time series sampled over an underlying continuous spatial manifold—the brain. A fully spatio-temporal MAR models (ST-MAR) is developed within the framework of functional data analysis. For spatial data, each row of a matrix Ak is the influence field of a given voxel. A Bayesian ST-MAR model is specified in which the influence fields for all voxels are required to vary smoothly over space. This requirement is enforced by penalizing the spatial roughness of the influence fields. This roughness is calculated with a discrete version of the spatial Laplacian operator. A massive reduction in dimensionality of computations is achieved via the singular value decomposition, making an interactive exploration of the model feasible. Use of the model is illustrated with an fMRI time series that was gathered concurrently with EEG in order to analyze the origin of resting brain rhythms. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Neuroinformatics Springer Journals

Spatio-temporal autoregressive models defined over brain manifolds

Valdes-Sosa, Pedro A.
Neuroinformatics , Volume 2 (2) – Jun 1, 2004
Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/spatio-temporal-autoregressive-models-defined-over-brain-manifolds-bZJnyxLhPw
Publisher
Springer Journals
Copyright
Copyright © Humana Press Inc 2004
ISSN
1539-2791
eISSN
1559-0089
DOI
10.1385/ni:2:2:239
Publisher site
See Article on Publisher Site

Abstract

Multivariate Autoregressive time series models (MAR) are an increasingly used tool for exploring functional connectivity in Neuroimaging. They provide the framework for analyzing the Granger Causality of a given brain region on others. In this article, we shall limit our attention to linear MAR models, in which a set of matrices of autoregressive coefficients Ak (k = 1, …, p) describe the dependence of present values of the image on lagged values of its past. Methods for estimating the Ak and determining which elements that are zero are well-known and are the basis for directed measures of influence. However, to date, MAR models are limited in the number of time series they can handle, forcing the a priori selection of a (small) number of voxels or regions of interest for analysis. This ignores the full spatio-temporal nature of functional brain data which are, in fact, collections of time series sampled over an underlying continuous spatial manifold—the brain. A fully spatio-temporal MAR models (ST-MAR) is developed within the framework of functional data analysis. For spatial data, each row of a matrix Ak is the influence field of a given voxel. A Bayesian ST-MAR model is specified in which the influence fields for all voxels are required to vary smoothly over space. This requirement is enforced by penalizing the spatial roughness of the influence fields. This roughness is calculated with a discrete version of the spatial Laplacian operator. A massive reduction in dimensionality of computations is achieved via the singular value decomposition, making an interactive exploration of the model feasible. Use of the model is illustrated with an fMRI time series that was gathered concurrently with EEG in order to analyze the origin of resting brain rhythms.

Journal

Neuroinformatics – Springer Journals

Published: Jun 1, 2004

Keywords: Granger causality; neuroimages; Bayesian multivariate autoregressive model; spatio-temporal model

References

  • Partial direc-ted coherence:a new concept in neural structure determination
    Baccala, L. A.; Sameshima, K.
  • On the directionality of cortical interactions studied by structural analysis of electrophysiological recordings
    Bernasconi, C.; Konig, P.
  • Investigation of cooperative cortical dynamics by multivariate autoregressive modeling of event-related local field potentials
    Bressler, S. L.; Ding, M. Z.; Yang, W. M.
  • The choice of variables in multivariate regression: A non-conjugate Bayesian decision theory approach
    Brown, P.J.; Fearn, T.; Vannucci, M.
  • Bayes model averaging with selection of regressors
    Brown, P.J.; Vannucci, M.; Fearn, T.
  • Assessing interactions among neuronal systems using functional neuroimaging
    Buchel, C.; Friston, K.
  • Interactions among neuronal systems assessed with functional neuroimaging
    Buchel, C.; Friston, K.
  • Graphical interaction models for multivariate time series
    Dahlhaus, R.
  • Identification of synaptic connections in neural ensembles by graphical models
    Dahlhaus, R.; Eichler, M.; Sandkuhler, J.
  • Some Matrix-Variate Distribution-Theory—Notational considerations and a Bayesian application
    Dawid, A. P.
  • Jackknife, bootstrap and other resampling methods in regression-analysis—Discussion
    Efron, B.
  • Testing non-linearity and directedness of interactions between neural groups in the macaque inferotemporal cortex
    Freiwald, W. A.; Valdes, P.; Bosch, J.; Biscay, R.; Jimenez, J. C.; Rodriguez, L. M.
  • Revealing interactions among brain systems with nonlinear PCA
    Friston, K.; Phillips, J.; Chawla, D.; Buchel, C.
  • Functional and effective connectivity in neuroimaging: a synthesis
    Friston, K. J.
  • Parametric time-series models for multivariate EEG analysis
    Gersch, W.; Yonemoto, J.
  • Measurement of linear-dependence and feedback between multiple time-series
    Geweke, J.
  • Measures of conditional linear-dependence and feedback between time-series
    Geweke, J. F.
  • Simultaneous EEG and fMRI of the alpha rhythm
    Goldman, R. I.; Stern, J. M.; Engel, J.; Cohen, M. S.
  • Generalized cross-validation as a method for choosing a good ridge parameter
    Golub, G. H.; Heath, M.; Wahba, G.
  • Investigating causal relations by econometric models and cross- spectral methods
    Granger, C. W. J.
  • Regularization tools version 3.0 for Matlab 5.2
    Hansen, P.C.
  • Multivariate autoregressive modeling of fMRI time series
    Harrison, L.; Penny, W. D.; Friston, K.
  • The use of time-variant EEG Granger causality for inspecting directed interdependencies of neural assemblies
    Hesse, W.; Moller, E.; Arnold, M.; Schack, B.
  • The decomposition and measurement of the interdependency between 2nd-order stationary-processes
    Hosoya, Y.
  • Evaluating causal relations in neural systems:Granger causality, directed transfer function and statistical assessment of significance
    Kaminski, M.; Ding, M. Z.; Truccolo, W. A.; Bressler, S. L.
  • A smoothness priors time-varying Ar coefficient modeling of nonstationary covariance time-series
    Kitagawa, G.; Gersch, W.
  • Penalized discriminant analysis of [O-15]-water PET brain images with prediction error selection of smoothness and regularization hyperparameters
    Kustra, R.; Strother, S.
  • A report of the functional connectivity workshop, Dusseldorf 2002
    Lee, L.; Harrison, L. M.; Mechelli, A.
  • The Kriged Kalman filter - rejoinder
    Mardia, K. V.; Goodall, C.; Redfern, E.; Alonso, F. J.
  • Structural equation modeling and its applications to network analysis in functional brain imaging
    McIntosh, A. R.; Gonzalez-Lima, F.
  • Analysis of fMRI data by blind separation into independent spatial components
    McKeown, M. J.; Makeig, S.; Brown, G. G.; Jung, T. P.; Kindermann, S. S.; Bell, A. J.
  • Estimation of parameters and eigenmodes of multivariate autoregressive models
    Neumaier, A.; Schneider, T.
  • Graphs, causality, and structural equation models
    Pearl, J.
  • Bayesian multivariate autoregressive models with structured priors
    Penny, W. D.; Roberts, S. J.
  • Locally regularized spatiotemporal modeling and model comparison for functional MRI
    Purdon, P. L.; Solo, V.; Weisskoff, R. M.; Brown, E. N.
  • Some tools for functional data-analysis
    Ramsay, J. O.; Dalzell, C. J.
  • Analysis of average evoked potentials making use of least mean square techniques
    Ruchkin, D. S.; John, E. R.; Villegas, J.
  • Blind signal separation from optical imaging recordings with extended spatial decorrelation
    Schiessl, I.; Stetter, M.; Mayhew, J. E. W.; McLoughlin, N.; Lund, J. S.; Obermayer, K.
  • Algorithm 808: ARfit—A matlab package for the estimation of parameters and eigenmodes of multivariate autoregressive models
    Schneider, T.; Neumaier, A.
  • Dynamic models for spatiotemporal data
    Stroud, J. R.; Muller, P.; Sanso, B.
  • A dimension-reduced approach to space-time Kalman filtering
    Wikle, C. K.; Cressie, N.
  • A unified statistical approach for determining significant signals in images of cerebral activation
    Worsley, K. J.; Marrett, S.; Neelin, P.; Vandal, A. C.; Friston, K. J.; Evans, A. C.