Hi developers:
I have observed an error using the “-kernel 3d” option for sct_deepseg_sc (v.4.0.0-beta.2). The use of the “-kernel 2d” option works just fine when applied to the same data. The outputs are copied below.
I was previously using ver 3.2.7 and observed that the “-kernel 3d” option did not work, which is why I upgraded to v.4.0.0-beta.2 (although I don’t recall if the error message was the same in ver 3.2.7 - I can check if that’s important). I’d be happy to help test this further…
Thanks,
-Rob
barryrl% sct_deepseg_sc -i t2.nii -c t2 -brain 1 -kernel 2d
Method:
Centerline algorithm: svm
Assumes brain section included in the image: True
Dimension of the segmentation kernel convolutions: 2d
Create temporary folder (/var/folders/mf/5wv63z61295dcfvzm_l2llh8003jn9/T/sct-20190416141552.661809-v1M9wH)…
Reorient the image to RPI, if necessary…
Finding the spinal cord centerline…
Remove temporary files…
rm -rf /var/folders/mf/5wv63z61295dcfvzm_l2llh8003jn9/T/sct-20190416141620.975540-FSPxzN
Resample the image to 0.5 mm isotropic resolution…
Cropping the image around the spinal cord…
Normalizing the intensity…
Segmenting the spinal cord using deep learning on 2D patches…
Reassembling the image…
WARNING: To avoid intensity overflow due to convertion to uint8, intensity will be rescaled to the maximum quantization scale.
Resampling the segmentation to the original image resolution…
Binarizing the segmentation to avoid interpolation effects…
Removing small objects above slice#590
Removing small objects above slice#590
Filling a hole in the segmentation around z_slice #:55
Filling a hole in the segmentation around z_slice #:62
Filling a hole in the segmentation around z_slice #:67
Remove temporary files…
rm -rf /var/folders/mf/5wv63z61295dcfvzm_l2llh8003jn9/T/sct-20190416141552.661809-v1M9wH
barryrl% sct_deepseg_sc -i t2.nii -c t2 -brain 1 -kernel 3d
Method:
Centerline algorithm: svm
Assumes brain section included in the image: True
Dimension of the segmentation kernel convolutions: 3d
Create temporary folder (/var/folders/mf/5wv63z61295dcfvzm_l2llh8003jn9/T/sct-20190416141856.379420-OT_qWu)…
Reorient the image to RPI, if necessary…
Finding the spinal cord centerline…
Remove temporary files…
rm -rf /var/folders/mf/5wv63z61295dcfvzm_l2llh8003jn9/T/sct-20190416141926.368940-MHyE7_
Resample the image to 0.5 mm isotropic resolution…
Cropping the image around the spinal cord…
Normalizing the intensity…
Traceback (most recent call last):
File “/Applications/SCT/ver_4_0_0_beta_2/scripts/sct_deepseg_sc.py”, line 185, in
main()
File “/Applications/SCT/ver_4_0_0_beta_2/scripts/sct_deepseg_sc.py”, line 159, in main
brain_bool=brain_bool, kernel_size=kernel_size, remove_temp_files=remove_temp_files, verbose=verbose)
File “/Applications/SCT/ver_4_0_0_beta_2/spinalcordtoolbox/deepseg_sc/core.py”, line 640, in deep_segmentation_spinalcord
‘mm’, ‘linear’, verbose=0)
File “/Applications/SCT/ver_4_0_0_beta_2/spinalcordtoolbox/resampling.py”, line 138, in resample_file
nii = nipy.load_image(fname_data)
File “/Applications/SCT/ver_4_0_0_beta_2/python/lib/python2.7/site-packages/nipy/io/files.py”, line 57, in load
img = nib.load(filename)
File “/Applications/SCT/ver_4_0_0_beta_2/python/lib/python2.7/site-packages/nibabel/loadsave.py”, line 42, in load
raise FileNotFoundError(“No such file or no access: ‘%s’” % filename)
nibabel.py3k.FileNotFoundError: No such file or no access: ‘image_in_RPI_norm.nii’