ANTsPyNet: Incorrect sform/qform in `croppedMni152`

Hi ANTs devs,

I was using preprocessing pipeline proposed by your paper for cortical thickness and brain segmentation. I want to obtain the MNI coordinate from the transforms (inside antspynet.preprocess_brain_image) Then I realize the sample image croppedMni152 has incorrect sform/qform (Image IJK voxel space to RAS, in this case MNI152).

>>> import nibabel
>>> import antspynet
>>> img = nibabel.load(antspynet.get_antsxnet_data("croppedMni152"))
>>> img.header.get_sform()
array([[0., 0., 0., 0.],
       [0., 0., 0., 0.],
       [0., 0., 0., 0.],
       [0., 0., 0., 1.]])
>>> img.header.get_qform()
array([[ 1.,  0.,  0., -0.],
       [ 0.,  1.,  0., -0.],
       [ 0.,  0.,  1.,  0.],
       [ 0.,  0.,  0.,  1.]])
>>> 

I wonder if I have an IJK point in croppedMni152, How can I obtain the MNI-152 template coordinate?

About this issue

  • Original URL
  • State: closed
  • Created a year ago
  • Comments: 17 (17 by maintainers)

Most upvoted comments

Great. Just wanted to make sure you weren’t doing any extra work.

There’s no single MNI coordinate system, there are multiple series of MNI152 templates that are approximately the same but not exactly.

I would recommend taking the brain image output of the thickness pipeline and registering it to a specific public MNI template for analysis purposes.