ravetools (version 0.1.3)

register_volume: Imaging registration using 'NiftyReg'

Description

Registers 'CT' to 'MRI', or 'MRI' to another 'MRI'

Usage

register_volume(
  source,
  target,
  method = c("rigid", "affine", "nonlinear"),
  interpolation = c("cubic", "trilinear", "nearest"),
  threads = detect_threads(),
  symmetric = TRUE,
  verbose = TRUE,
  ...
)

Value

See niftyreg

Arguments

source

source imaging data, or a 'nifti' file path; for example, 'CT'

target

target imaging data to align to; for example, 'MRI'

method

method of transformation, choices are 'rigid', 'affine', or 'nonlinear'

interpolation

how volumes should be interpolated, choices are 'cubic', 'trilinear', or 'nearest'

threads, symmetric, verbose, ...

see niftyreg

Examples

if(interactive()) {

source <- system.file("extdata", "epi_t2.nii.gz", package="RNiftyReg")
target <- system.file("extdata", "flash_t1.nii.gz", package="RNiftyReg")
aligned <- register_volume(source, target, verbose = FALSE)

source_img <- aligned$source[[1]]
target_img <- aligned$target
aligned_img <- aligned$image

par(mfrow = c(2, 2), mar = c(0.1, 0.1, 3.1, 0.1))

pal <- grDevices::grey.colors(256, alpha = 1)
image(source_img[,,30], asp = 1, axes = FALSE,
      col = pal, main = "Source image")
image(target_img[,,64], asp = 1, axes = FALSE,
      col = pal, main = "Target image")
image(aligned_img[,,64], asp = 1, axes = FALSE,
      col = pal, main = "Aligned image")

# bucket fill and calculate differences
aligned_img[is.nan(aligned_img) | aligned_img <= 1] <- 1
target_img[is.nan(target_img) | aligned_img <= 1] <- 1
diff <- abs(aligned_img / target_img - 1)
image(diff[,,64], asp = 1, axes = FALSE,
      col = pal, main = "Percentage Difference")

}