Spatial 3D Matérn Priors for Fast Whole-Brain fMRI Analysis

Per Sidén*, Finn Lindgren, David Bolin, Anders Eklund, Mattias Villani

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


Bayesian whole-brain functional magnetic resonance imaging (fMRI) analysis with three-dimensional spatial smoothing priors has been shown to produce state-of-the-art activity maps without pre-smoothing the data. The proposed inference algorithms are computationally demanding however, and the spatial priors used have several less appealing properties, such as being improper and having infinite spatial range.We propose a statistical inference framework for whole-brain fMRI analysis based on the class of Matérn covariance functions. The framework uses the Gaussian Markov random field (GMRF) representation of possibly anisotropic spatial Matérn fields via the stochastic partial differential equation (SPDE) approach of Lindgren et al. (2011). This allows for more flexible and interpretable spatial priors, while maintaining the sparsity required for fast inference in the high-dimensional whole-brain setting. We develop an accelerated stochastic gradient descent (SGD) optimization algorithm for empirical Bayes (EB) inference of the spatial hyperparameters. Conditionally on the inferred hyperparameters, we make a fully Bayesian treatment of the brain activity.

Original languageEnglish (US)
Pages (from-to)1251-1278
Number of pages28
Issue number4
StatePublished - 2021


  • Efficient computation
  • Fmri
  • Gaussian markov random fields
  • Spatial priors
  • Spatiotemporal modeling

ASJC Scopus subject areas

  • Statistics and Probability
  • Applied Mathematics


Dive into the research topics of 'Spatial 3D Matérn Priors for Fast Whole-Brain fMRI Analysis'. Together they form a unique fingerprint.

Cite this