get_patch_from_volume: Get Patches from 3D Volume

In neuroconductor-devel-releases/flexconn: FLEXCONN Model

Description

Get Patches from 3D Volume

Usage

get_patch_from_volume(
  vol,
  mask = NULL,
  patchsize,
  verbose = TRUE,
  pad = TRUE,
  normalize = TRUE,
  contrast,
  seed = NULL,
  run_mask_patches = TRUE,
  peak_estimator = c("Python", "R"),
  ...
)

get_num_patches(mask)

norm_pad(
  vol,
  mask = NULL,
  patchsize,
  verbose = TRUE,
  pad = TRUE,
  normalize = TRUE,
  peak_estimator = c("Python", "R"),
  contrast
)

get_mask_patches(mask, patchsize, pad = TRUE, seed = NULL, verbose = TRUE)

Arguments

vol	3D array or nifti image
mask	binary 3D array or nifti image, for the brain usually
patchsize	Vector of length 2 (or more)
verbose	print diagnostic messages
pad	Run pad_image on the image before getting patches (pads then normalizes if normalize = TRUE)
normalize	Run normalize_image on the image before getting patches
contrast	What imaging sequence of MRI is this volume, passed to normalize_image
seed	Seed for random sampling of indices. If NULL, no sampling is done, passed to mask_indices
run_mask_patches	Should the patches for the mask be made as well?
peak_estimator	Which functions to use to estimate peak, either wrapper Python code or native R code.
...	not used

Value

A list of image and mask Patches

Examples

user = Sys.getenv("USER")
if (user == "johnmuschelli") {
reticulate::use_python(paste0(
"/Library/Frameworks/Python.framework/Versions/3.5/bin/python3"))
mpl = reticulate::import("matplotlib")
mpl$use('TkAgg')
}
library(neurobase)
fname = system.file("extdata", "MPRAGE.nii.gz", package = "flexconn")
vol = readnii(fname)
mask = vol > 0
patchsize = c(5, 5)
verbose = TRUE
patch = get_patch_from_volume(vol, mask, patchsize = patchsize,
contrast = "T1")

rm(patch)

patch = get_patch_from_volume(vol, mask, patchsize = c(1,1,1),
contrast = "T1")
rm(patch)

neuroconductor-devel-releases/flexconn documentation built on May 6, 2020, 4:21 p.m.