Down-sample mri T1 image in python with Nipy
Question:
I have a T1 image (NIFTI), already aligned, with dimension 121 x 145 x 121.
The image is loaded by nibabel. The voxel size is 1.5 x 1.5 x 1.5 mm.
I want to down-sample it to an image with 2.0 x 2.0 x 2.0 mm resolution and keep the images aligned.
I have little knowledge in MRI image manipulation.
I couldn’t find a clear tutorial.
How do I do that ?
If you know any other Python library that can do it, it would also work.
Answers:
I would suggest using Nibabel. It can downsample your nifti file in just a few lines.
Example to resample to an image to a voxel size of 2x2x2:
import nibabel
import nibabel.processing
input_path = r'/input/path/input_img.nii.gz'
output_path = r'/output/path/output_img.nii.gz'
voxel_size = [2, 2, 2]
input_img = nibabel.load(input_path)
resampled_img = nibabel.processing.resample_to_output(input_img, voxel_size)
nibabel.save(resampled_img, output_path)
Just update input_path and output_path to reflect your files. The second argument in the resample_to_output function (voxel_size) needs to either match the dimensions of your input or be a single value, nibabel will then assume you want the same voxel size for all dimensions.
Nibabel info:
Docs: http://nipy.org/nibabel/.
Install instructions: https://anaconda.org/conda-forge/nibabel
I have a T1 image (NIFTI), already aligned, with dimension 121 x 145 x 121.
The image is loaded by nibabel. The voxel size is 1.5 x 1.5 x 1.5 mm.
I want to down-sample it to an image with 2.0 x 2.0 x 2.0 mm resolution and keep the images aligned.
I have little knowledge in MRI image manipulation.
I couldn’t find a clear tutorial.
How do I do that ?
If you know any other Python library that can do it, it would also work.
I would suggest using Nibabel. It can downsample your nifti file in just a few lines.
Example to resample to an image to a voxel size of 2x2x2:
import nibabel
import nibabel.processing
input_path = r'/input/path/input_img.nii.gz'
output_path = r'/output/path/output_img.nii.gz'
voxel_size = [2, 2, 2]
input_img = nibabel.load(input_path)
resampled_img = nibabel.processing.resample_to_output(input_img, voxel_size)
nibabel.save(resampled_img, output_path)
Just update input_path and output_path to reflect your files. The second argument in the resample_to_output function (voxel_size) needs to either match the dimensions of your input or be a single value, nibabel will then assume you want the same voxel size for all dimensions.
Nibabel info:
Docs: http://nipy.org/nibabel/.
Install instructions: https://anaconda.org/conda-forge/nibabel