BSPBSS: Bayesian Spatial Blind Source Separation

Gibbs sampling for Bayesian spatial blind source separation (BSP-BSS). BSP-BSS is designed for spatially dependent signals in high dimensional and large-scale data, such as neuroimaging. The method assumes the expectation of the observed images as a linear mixture of multiple sparse and piece-wise smooth latent source signals, and constructs a Bayesian nonparametric prior by thresholding Gaussian processes. Details can be found in our working paper: Ben et al. (2022+) "Bayesian Spatial Blind Source Separation via the Thresholded Gaussian Process".

Version: 1.0.2
Depends: R (≥ 3.4.0), movMF
Imports: rstiefel, Rcpp, ica, glmnet, gplots, BayesGPfit, svd, RandomFieldsUtils, neurobase, oro.nifti, gridExtra, ggplot2, gtools
LinkingTo: Rcpp, RcppArmadillo
Suggests: knitr, rmarkdown
Published: 2022-09-02
Author: Ben Wu [aut, cre], Ying Guo [aut], Jian Kang [aut]
Maintainer: Ben Wu <wuben at ruc.edu.cn>
License: GPL (≥ 3)
NeedsCompilation: yes
SystemRequirements: GNU make
Materials: README
CRAN checks: BSPBSS results

Documentation:

Reference manual: BSPBSS.pdf
Vignettes: BSPBSS-vignette

Downloads:

Package source: BSPBSS_1.0.2.tar.gz
Windows binaries: r-devel: BSPBSS_1.0.2.zip, r-release: BSPBSS_1.0.2.zip, r-oldrel: BSPBSS_1.0.2.zip
macOS binaries: r-release (arm64): BSPBSS_1.0.2.tgz, r-oldrel (arm64): BSPBSS_1.0.2.tgz, r-release (x86_64): BSPBSS_1.0.2.tgz, r-oldrel (x86_64): BSPBSS_1.0.2.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=BSPBSS to link to this page.