ARI {ARIbrain}

Valid Circular Inference (ARI) for Brain Imaging

Description

Valid Circular Inference (ARI) for Brain Imaging

Usage

ARI(Pmap, clusters, mask = NULL, alpha = 0.05, Statmap = function(ix)
  -qnorm(Pmap[ix]), summary_stat = c("max", "center-of-mass"),
  silent = FALSE)

Arguments

Pmap

3D array of p-values or a (character) nifti file name.

clusters

3D array of cluster ids (0 when voxel does not belong to any cluster) or a (character) nifti file name.

mask

3D array of locicals (i.e. TRUE/FALSE in/out of the brain). Alternatively it may be a (character) nifti file name. If mask=NULL, it is assumed that non of the voxels have to be excluded.

alpha

Significance level. alpha=.05 by default.

Statmap

Statistics (usually t-values) on which the summaries are based. Can be either a 3D array, a (character) nifti file name or a function with argument ix used in the function to select the voxels belonging to a given cluster. By default Statmap = function(ix) -qnorm(Pmap[ix]) which convert the p-values in one-sided z-score.

summary_stat

Choose among =c("max", "center-of-mass").

silent

FALSE by default.

Value

A matrix reporting Size, FalseNull, TrueNull, ActiveProp and other statistics for each cluster.

Examples

pvalue_name <- system.file("extdata", "pvalue.nii.gz", package="ARIbrain")
cluster_name <- system.file("extdata", "cluster_th_3.2.nii.gz", package="ARIbrain")
zstat_name <- system.file("extdata", "zstat.nii.gz", package="ARIbrain")
mask_name <- system.file("extdata", "mask.nii.gz", package="ARIbrain")

print(mask_name)
print(pvalue_name)
print(cluster_name)
print(zstat_name)

ARI(Pmap = pvalue_name, clusters= cluster_name, 
    mask=mask_name, Statmap = zstat_name)