Skip to contents

Resample TemplateFlow Mask to fMRIPrep Image Using Python

Source: R/postprocess_functions.R
resample_template_to_img.Rd

This function uses Python (via reticulate) to identify and resample a TemplateFlow mask to match the resolution and spatial dimensions of an fMRIPrep BOLD image.

Usage

resample_template_to_img(
  in_file,
  output = NULL,
  template_resolution = 1,
  suffix = "mask",
  desc = "brain",
  extension = ".nii.gz",
  interpolation = "nearest",
  install_dependencies = TRUE,
  overwrite = FALSE
)

Arguments

in_file

Path to the BIDS-compliant NIfTI file (e.g., an fMRIPrep preprocessed BOLD image).

output

Optional path to write the resampled image. If NULL, a BIDS-style filename is constructed.

template_resolution

Integer specifying the TemplateFlow resolution index (e.g., 1 = 1mm).

suffix

TemplateFlow suffix (e.g., "mask", "T1w").

desc

TemplateFlow descriptor (e.g., "brain").

extension

File extension for the template image (default is ".nii.gz").

interpolation

Interpolation method to use during resampling. Options are "nearest", "linear", or "continuous".

install_dependencies

Logical. If TRUE (default), attempts to automatically install required Python packages (nibabel, nilearn, templateflow) if they are missing from the active environment. If FALSE, the function will raise an error if dependencies are not found.

overwrite

Logical. If TRUE, overwrite the existing output file (if present).

Value

Invisibly returns TRUE on success. A new NIfTI file is written to output.

Details

The appropriate template is inferred from the space- entity of the BIDS-formatted input filename. For example, an input such as: sub-221256_task-trust_run-1_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz will lead to selection of the MNI152NLin2009cAsym template.

This function depends on a companion Python script (fetch_matched_template_image.py) that is bundled with the BrainGnomes package and sourced at runtime.