The BHMSMAfMRI package performs BHMSMA (Sanyal & Ferreira, 2012) of fMRI data, or other multiscale data, using wavelet based prior that borrows strength across subjects and provides posterior smoothed images of the effect sizes and samples from the posterior distribution. The package currently considers analysis of 2D slices/grids only.

Details

Package:BHMSMAfMRI
Type:Package
Version:2.1
Date:2022-10-01
License:GPL (>= 2)

Import fMRI data using:
readfmridata

The main analysis function, which provides subject-specific posterior estimates, is:
BHMSMA

The main function sucessively calls the following functions:
glmcoef (get regression coefficients)
waveletcoef (get wavelet coefficients)
hyperparamest (estimate model hyperparameters)
postmixprob (estimate posterior mixture probabilities of wavelet coefficients)
postwaveletcoef (compute posterior estimates of wavelet coefficients)
postglmcoef (compute posterior estimates of regression coefficients)

For posterior group estimates of regression coefficients use:
postgroupglmcoef

For posterior uncertainty estimates use:
postsamples

Internal sample data:
fmridata

Miscellaneous:
substituteWaveletCoef

Author

Nilotpal Sanyal <nilotpal.sanyal@gmail.com>, Marco Ferreira <marf@vt.edu>

Maintainer: Nilotpal Sanyal <nilotpal.sanyal@gmail.com>

References

Sanyal, Nilotpal, and Ferreira, Marco A.R. (2012). Bayesian hierarchical multi-subject multiscale analysis of functional MRI data. Neuroimage, 63, 3, 1519-1531.