cudaBayesreg: CUDA Parallel Implementation of a Bayesian Multilevel Model for
fMRI Data Analysis
Compute Unified Device Architecture (CUDA) is a software
platform for massively parallel high-performance computing on
NVIDIA GPUs. This package provides a CUDA implementation of a
Bayesian multilevel model for the analysis of brain fMRI data.
A fMRI data set consists of time series of volume data in 4D
space. Typically, volumes are collected as slices of 64 x 64
voxels. Analysis of fMRI data often relies on fitting linear
regression models at each voxel of the brain. The volume of the
data to be processed, and the type of statistical analysis to
perform in fMRI analysis, call for high-performance computing
strategies. In this package, the CUDA programming model uses a
separate thread for fitting a linear regression model at each
voxel in parallel. The global statistical model implements a
Gibbs Sampler for hierarchical linear models with a normal
prior. This model has been proposed by Rossi, Allenby and
McCulloch in ‘Bayesian Statistics and Marketing’, Chapter 3,
and is referred to as ‘rhierLinearModel’ in the R-package
bayesm. A notebook equipped with a NVIDIA ‘GeForce 8400M GS’
card having Compute Capability 1.1 has been used in the tests.
The data sets used in the package's examples are available in
the separate package cudaBayesregData.