R/z_images.R

In muschellij2/msseg: Segmentation of Lesions for Multiple Sclerosis

#' @title Standardized to Template Images SyN
#' @description Normalizes to the template
#' @param orig_t1_pre Skull-stripped T1 image to register to Eve
#' @param t1_pre Normalized T1 Pre
#' @param t1_post Normalized T1 Post
#' @param flair Normalized FLAIR
#' @param t2 Normalized T2
#' @param pd Normalized PD
#' @param outdir output directory
#' @param add_stub Stub to add on the filenames
#'
#' @return NULL
#' @export
#' @importFrom extrantsr registration ants_apply_transforms
z_images = function(orig_t1_pre,
                    t1_pre,
                    t1_post,
                    flair,
                    t2,
                    pd,
                    outdir = ".",
                    add_stub = "_strip_norm"){
  verbose = TRUE

  niis = list(FLAIR = flair,
              T1_Pre = t1_pre,
              T1_Post = t1_post,
              T2 = t2,
              PD = pd)

  # REMOVE NULL
  nulls = sapply(niis, is.null)
  niis = niis[!nulls]

  keep_cols = nii_names = names(niis)

  imgs = check_nifti(niis)
  stopifnot(all(c("T1_Pre", "FLAIR") %in% nii_names))
  if (verbose) {
    message("Reading in Images")
  }

  #############################################
  # Columns to keep
  #############################################
  ofnames = file.path(outdir, keep_cols)
  ofnames = outer(ofnames,
                  c("_Z_Mean", "_Z_Median"),
                  paste0)
  ofnames = paste0(ofnames,
                   ".nii.gz")

  # if (!all_exists(ofnames))) {
  #############################################
  # Get stubs for whitestripes
  #############################################
  # reg_name = "T1_Pre"
  # reg_file = irow[, reg_name]
  reg_file = orig_t1_pre
  template_brain = system.file("JHU_MNI_SS_T1_brain.nii.gz",
                               package = "msseg")

  outprefix = tempfile()
  reg = registration(
    filename = reg_file,
    template.file = template_brain,
    typeofTransform = "SyN",
    interpolator = "LanczosWindowedSinc",
    verbose = verbose,
    remove.warp = FALSE,
    copy_origin = TRUE,
    outprefix = outprefix
  )


  # temp = check_nifti(template_brain)
  # nrows = length(temp)

  native = check_nifti(reg_file)

  icol = keep_cols[1]

  template_dir = system.file("Normed_Templates",
                             package = "msseg")
  for (icol in keep_cols) {

    out_stub = file.path(template_dir,
                         paste0("Template_", icol))

    out_name = paste0(out_stub, "_Median")
    med_img = readnii(out_name)

    out_name = paste0(out_stub, "_SD")
    sd_img = readnii(out_name)

    out_name = paste0(out_stub, "_Mean")
    mean_img = readnii(out_name)


    ofname = file.path(outdir, paste0(icol, add_stub))
    nim = check_nifti(imgs[[icol]])

    z_mean = finite_img(
      (nim - mean_img) /
        sd_img )

    z_med = finite_img(
      (nim - med_img) /
        sd_img )


    z_mean_native = ants_apply_transforms(
      fixed = native,
      moving = z_mean,
      transformlist = reg$invtransforms,
      interpolator = reg$interpolator)
    z_mean_native = robust_window(
      z_mean_native,
      probs = c(0.001, 0.999))
    writenii(z_mean_native,
             filename = paste0(ofname, "_Z_Mean"))

    z_med_native = ants_apply_transforms(
      fixed = native,
      moving = z_med,
      transformlist = reg$invtransforms,
      interpolator = reg$interpolator)
    z_med_native = robust_window(
      z_med_native,
      probs = c(0.001, 0.999))
    writenii(z_med_native,
             filename = paste0(ofname,
                               "_Z_Median"))

    if (verbose) {
      message(icol)
    }
  }
  return(invisible(NULL))
}