spm_bwlabel: SPM BWLabel Clusters of Certain Size

In muschellij2/spm12r: Wrapper Functions for 'SPM' (Statistical Parametric Mapping) Version 12 from the 'Wellcome' Trust Centre for 'Neuroimaging'

SPM BWLabel Clusters of Certain Size

Description

Get Cluster of certain size from spm_bwlabel

Usage

spm_bwlabel(
  infile,
  outfile = NULL,
  retimg = TRUE,
  k = NULL,
  topN = NULL,
  margin = NULL,
  binary = TRUE,
  spmdir = spm_dir(verbose = verbose, install_dir = install_dir),
  reorient = FALSE,
  verbose = TRUE,
  install_dir = NULL
)

Arguments

infile	input filename
outfile	output filename
retimg	Return the image instead of matlab output
k	Minimum cluster size needed
topN	Top number of clusters kept (used if k is NULL)
margin	Margin to loop over if wanted in 2D
binary	(logical) Should the result be binary or numbered with cluster.
spmdir	SPM directory (for MATLAB)
reorient	If retimg, then this argument is passed to readNIfTI
verbose	Print Diagnostics
install_dir	directory to download SPM12

Value

Output from run_matlab_script or nifti object, depending on retimg

Result from run_matlab_script

Note

Taken from http://en.wikibooks.org/wiki/SPM/How-to#How_to_remove_clusters_under_a_certain_size_in_a_binary_mask.3F

Examples

if (matlabr::have_matlab()) { 
d = rep(25, 3)
arr = array(rbinom(prod(d), size = 1, prob = 0.5), dim = d)
nim = oro.nifti::nifti(arr)
res = spm_bwlabel(nim, k = 1, binary = FALSE, install_dir = tempdir())
tab = table(c(res))
tab
out_arr = as(res, "array")
}

muschellij2/spm12r documentation built on Aug. 31, 2022, 9:49 p.m.