voxel_selection: Voxel Selection Procedure

In neuroconductor-devel/oasis: Multiple Sclerosis Lesion Segmentation using Magnetic Resonance Imaging (MRI)

Voxel Selection Procedure

Description

This function creates a binary mask for the voxel selection procedure for OASIS.

Usage

voxel_selection(flair, brain_mask, cutoff)

Arguments

flair	FLAIR volume of class nifti
brain_mask	brain mask of class nifti
cutoff	the percentile cutoff for the thresholding, passed to quantile

Value

Returns the voxel selection mask as an object of class nifti.

Examples

library(neurobase)
library(fslr)
library(oasis)
niis = tempfile(fileext = ".nii.gz")
if (require(httr)) {
   url = paste0("https://s3.us-east-2.amazonaws.com/brainder/software/",
   "flair/templates/GG-853-FLAIR-2.0mm.nii.gz")
   req <- httr::GET(url,
   httr::write_disk(path = niis))
   httr::stop_for_status(req)

 flair <- readnii(niis)
 if (have.fsl()) {
   brain_mask = fslbet(niis) > 0
 } else {
   ind = list(c(10L, 81L), c(12L, 101L), c(3L, 78L))
   all.ind = lapply(ind, function(x) seq(x[1], x[2]))
   brain_mask = niftiarr(flair, 0)
   eg = expand.grid(all.ind)
   eg = as.matrix(eg)
   brain_mask[eg] = 1
 }
 voxel_selection_mask <- voxel_selection(flair,
   brain_mask, cutoff = .85)
}

neuroconductor-devel/oasis documentation built on July 3, 2025, 11:49 p.m.