Coding issue with nibabel--- + Torchio

Hi. Thx for this wonderful repository.

I have a question. If I get the image + labels array for all my data, and corresponding ground truths with img=nib.load(os.path…), with gt… Then, I use img=img_getdata()…, then return them. So then, my question=> in order to use Torchio, which I guess whill be provided by the (subjects, channel, h,w,d)----nibabel, right? If that’s correct, do I need to use then as—

For example a class,

if transform is None:

            if mode == 'train':

               transform = ([
               tio.RandomAffine(): 0.75,
               tio.RandomElasticDeformation(): 0.25,
               tio.RandomAffine(scales=(0.9, 1.2), degrees=(10)),
               tio.RandomFlip(axes=('LR'))..... Likewise
               ])

[fepegar]

Typically, array = img.get_data() will return a NumPy array of shape (w, h, d). You can add a channels dimension doing for example array[np.newaxis]and passing that to the tensor kwarg of tio.ScalarImage. But this is probably not necessary. Just pass the image path to tio.ScalarImage and it will take care of reading it.

I suggest you take a look at the Getting started section in the docs, and go through all the notebooks.

[fepegar]

I’m not sure what the question is. Can you maybe try to be more specific?