cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis
Graphic processing units (GPUs) are rapidly gaining maturity as powerful general parallel computing devices. A key feature in the development of modern GPUs has been the advancement of the programming model and programming tools. Compute Unified Device Architecture (CUDA) is a software platform for...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Foundation for Open Access Statistics
2011-10-01
|
| Series: | Journal of Statistical Software |
| Subjects: | |
| Online Access: | http://www.jstatsoft.org/v44/i04/paper |
| _version_ | 1818355763714719744 |
|---|---|
| author | Adelino R. Ferreira da Silva |
| author_facet | Adelino R. Ferreira da Silva |
| author_sort | Adelino R. Ferreira da Silva |
| collection | DOAJ |
| description | Graphic processing units (GPUs) are rapidly gaining maturity as powerful general parallel computing devices. A key feature in the development of modern GPUs has been the advancement of the programming model and programming tools. Compute Unified Device Architecture (CUDA) is a software platform for massively parallel high-performance computing on Nvidia many-core GPUs. In functional magnetic resonance imaging (fMRI), the volume of the data to be processed, and the type of statistical analysis to perform call for high-performance computing strategies. In this work, we present the main features of the R-CUDA package cudaBayesreg which implements in CUDA the core of a Bayesian multilevel model for the analysis of brain fMRI data. The statistical model implements a Gibbs sampler for multilevel/hierarchical linear models with a normal prior. The main contribution for the increased performance comes from the use of separate threads for fitting the linear regression model at each voxel in parallel. The R-CUDA implementation of the Bayesian model proposed here has been able to reduce significantly the run-time processing of Markov chain Monte Carlo (MCMC) simulations used in Bayesian fMRI data analyses. Presently, cudaBayesreg is only configured for Linux systems with Nvidia CUDA support. |
| first_indexed | 2024-12-13T19:46:29Z |
| format | Article |
| id | doaj.art-c01fd90cefb142a6bee0bdc4a8213418 |
| institution | Directory Open Access Journal |
| issn | 1548-7660 |
| language | English |
| last_indexed | 2024-12-13T19:46:29Z |
| publishDate | 2011-10-01 |
| publisher | Foundation for Open Access Statistics |
| record_format | Article |
| series | Journal of Statistical Software |
| spelling | doaj.art-c01fd90cefb142a6bee0bdc4a82134182022-12-21T23:33:33ZengFoundation for Open Access StatisticsJournal of Statistical Software1548-76602011-10-014404cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data AnalysisAdelino R. Ferreira da SilvaGraphic processing units (GPUs) are rapidly gaining maturity as powerful general parallel computing devices. A key feature in the development of modern GPUs has been the advancement of the programming model and programming tools. Compute Unified Device Architecture (CUDA) is a software platform for massively parallel high-performance computing on Nvidia many-core GPUs. In functional magnetic resonance imaging (fMRI), the volume of the data to be processed, and the type of statistical analysis to perform call for high-performance computing strategies. In this work, we present the main features of the R-CUDA package cudaBayesreg which implements in CUDA the core of a Bayesian multilevel model for the analysis of brain fMRI data. The statistical model implements a Gibbs sampler for multilevel/hierarchical linear models with a normal prior. The main contribution for the increased performance comes from the use of separate threads for fitting the linear regression model at each voxel in parallel. The R-CUDA implementation of the Bayesian model proposed here has been able to reduce significantly the run-time processing of Markov chain Monte Carlo (MCMC) simulations used in Bayesian fMRI data analyses. Presently, cudaBayesreg is only configured for Linux systems with Nvidia CUDA support.http://www.jstatsoft.org/v44/i04/paperBayesian multilevel methodsfMRIRGPUCUDA |
| spellingShingle | Adelino R. Ferreira da Silva cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis Journal of Statistical Software Bayesian multilevel methods fMRI R GPU CUDA |
| title | cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis |
| title_full | cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis |
| title_fullStr | cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis |
| title_full_unstemmed | cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis |
| title_short | cudaBayesreg: Parallel Implementation of a Bayesian Multilevel Model for fMRI Data Analysis |
| title_sort | cudabayesreg parallel implementation of a bayesian multilevel model for fmri data analysis |
| topic | Bayesian multilevel methods fMRI R GPU CUDA |
| url | http://www.jstatsoft.org/v44/i04/paper |
| work_keys_str_mv | AT adelinorferreiradasilva cudabayesregparallelimplementationofabayesianmultilevelmodelforfmridataanalysis |