New release r20240118 discussion

[rohanbanerjee]

After the previous release, I have updated the segmentation file, disc file (added a disc level) and the vertebrae level file (added a vertebrae level). I am attaching the folder in this issue as a zip folder (a gdrive link). If this looks good, I will publish the release with the release name r20240118 with the same r20240118 in the release tag.

https://drive.google.com/file/d/1UT2tqePQulXSopJwxAkpjdlytkewHT5e/view?usp=sharing

Discussion regarding registration using this is in the issue: #14

[jcohenadad]

should not be here

[jcohenadad]

to remove:

[jcohenadad]

[jcohenadad]

The cord segmentation is incorrect:

[jcohenadad]

cord seg should be symmetric-- see code in sct PR (if you don,t find it, ask @jcohenadad )

[jcohenadad]

Mismatch between t1 and t2. Consider re-doing the template:

[jcohenadad]

The cord seg is better done with starting from propseg on the t2 and then manually correct:

[jcohenadad]

plan moving forward

• we ditch the t1w for now
• we re-do the seg on the t2w template

[rohanbanerjee]

Redone t2w template segmentation. It is part of the below link. If the cord segmentation looks good, it will be the next release.
The modified file is called: templatedog_cord.nii.gz. In order to review, it, overlay it on the t2w template file called: templatedog_t2w.nii.gz.

Here is a link to download the template: https://drive.google.com/file/d/1bko995GOvosglwy6FlBBiQEeUzYD8MFN/view?usp=sharing

[jcohenadad]

Here is the result of registration of single dog image on the updated version of the template: qc_templatedog_t2.zip

The first row is a trial and should be ignored. The second row is the one to look at.

As per #14 (comment), here is the command used to produce this registration:

# only keep 2 disc labels
sct_label_utils -i ~/data.neuro/template_dog_virginiatech/derivatives/labels/sub-HarshmanDobby/anat/sub-HarshmanDobby_T2_labels-disc.nii.gz -keep 4,10 -o labels_4-10.nii.gz
# run registration (params not optimized)
sct_register_to_template -i ~/data.neuro/template_dog_virginiatech/sub-HarshmanDobby/anat/sub-HarshmanDobby_T2.nii.gz -s ~/data.neuro/template_dog_virginiatech/derivatives/labels/sub-HarshmanDobby/anat/sub-HarshmanDobby_T2_label-centerline.nii.gz -ldisc labels_4-10.nii.gz -t . -c t2 -s-template-id 2 -ofolder test_discs4-10 -param step=1,type=seg,algo=centermass:step=2,type=im,algo=syn,metric=CC,iter=3 -qc qc~~~

SCT version used: Spinal Cord Toolbox (git-jca/4298-flow-d396ad3a9090900cc62d4b6eb4ccbc6d63d4a6eb)

[jcohenadad]

Information needed about how the new segmentation in #16 (comment) was produced. Write EXACT syntax and protocol that was used to produce the segmentation.

[rohanbanerjee]

This release r20240118 has already been done and we are set to release the next version. Closing the issue therefore.