spine-generic / data-multi-subject

Multi-subject data for the Spine Generic project
Creative Commons Attribution 4.0 International
21 stars 15 forks source link

Update T1w scans with re-defaced images #148

Closed mguaypaq closed 7 months ago

mguaypaq commented 11 months ago

Images supplied by @renelabounek (thanks!)

Fixes #146.

valosekj commented 11 months ago

Thanks for the push @mguaypaq!

I visually checked a few subjects in FSLeyes, and I also checked the orientation. The re-defaced images correspond with the original images 👍🏻

However, I noticed that some of the re-defaced images have changed intensity range:

$ cd data-multi-subject

# master branch
$ fslstats sub-fslAchieva02/anat/sub-fslAchieva02_T1w.nii.gz -R
0.000000 4022.933105

# mgp/redefaced branch
$ git fetch
$ git checkout mgp/redefaced
$ git pull && git annex get .
$ fslstats sub-fslAchieva02/anat/sub-fslAchieva02_T1w.nii.gz -R
0.000000 3219834.250000

The data type is the same for both:

# master branch
$ sct_image -i sub-fslAchieva02/anat/sub-fslAchieva02_T1w.nii.gz -header | grep data_type                                              
data_type   FLOAT64                                                                                                                                                                                         

# mgp/redefaced branch
$ sct_image -i sub-fslAchieva02/anat/sub-fslAchieva02_T1w.nii.gz -header | grep data_type                                      
data_type   FLOAT64

And both images look identical in FSLeyes. So I wonder if this is a problem.

Kapture 2023-08-17 at 17 29 44

mguaypaq commented 11 months ago

Ultimately, I don't know whether the difference in intensity range is a problem. But, I did look at the old and new versions of sub-fslAchieva02/anat/sub-fslAchieva02_T1w.nii.gz, and this is what I noticed: