New dataset_hc 3T T2* from Marseille

[Nilser3]

Description

Dataset acquired in 25 HCs at 3T T2Star sequence from C1 to L2, contains GM/SC/CSF masks. (available in OSF)

This dataset was used in the papers 1 and 2

Details:

The images were uploaded in splitted 2D slices (one by level):

Their masks have the multilabel form (a single file contains different classes according to pixel intensity)

[jcohenadad]

Great!

• data should be volumes (not individual 2d slices)
• make sure it is BIDS compliant
• can you pls create a new label for this issue, i suggest the dataset name: marseille-t2s-template

[Nilser3]

There are some subjects with some missing levels, (example)

So, I propose merging the existing slices, and creating a levels file in derivatives (according to PAM50_levels convention).

[jcohenadad]

There are some subjects with some missing levels, (example)

what do you mean by "missing levels"?

• missing level labels?
• missing cord/WM/GM segmentation for some levels?
• missing image for some levels?

[Nilser3]

missing image and mask for some levels For example, sub-02 does not include: C1, T1-T7, L1-L2

[jcohenadad]

missing image and mask for some levels?

I assume the "?" is a typo.

In this case, yes, you can add a vertebral level file under the derivatives so we know what level each slice correspond to. Thanks!

[Nilser3]

The data has been converted to volumes with BIDS compliant (see warning).

bids-validator

```console (base) nilaia@rosenberg:~/data_nvme_nilaia$ bids-validator marseille-t2s-template/ --verbose bids-validator@1.9.7 bids-specification@disable 1: [WARN] Not all subjects/sessions/runs have the same scanning parameters. (code: 39 - INCONSISTENT_PARAMETERS) ./sub-01/anat/sub-01_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,19 (voxels). ./sub-02/anat/sub-02_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,10 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 30.00mm. ./sub-04/anat/sub-04_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,16 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-05/anat/sub-05_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,6 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-06/anat/sub-06_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,10 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-07/anat/sub-07_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,16 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-08/anat/sub-08_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,12 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-09/anat/sub-09_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,9 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-10/anat/sub-10_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,11 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-11/anat/sub-11_T2star.nii.gz The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-12/anat/sub-12_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,16 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-13/anat/sub-13_T2star.nii.gz The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-14/anat/sub-14_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,17 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-15/anat/sub-15_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,12 (voxels). The most common resolution is: 0.47mm x 0.47mm x 17.50mm, This file has the resolution: 0.47mm x 0.47mm x 24.50mm. ./sub-16/anat/sub-16_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,17 (voxels). ./sub-17/anat/sub-17_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,8 (voxels). ./sub-19/anat/sub-19_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,7 (voxels). ./sub-20/anat/sub-20_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,15 (voxels). ./sub-22/anat/sub-22_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,10 (voxels). ./sub-23/anat/sub-23_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,14 (voxels). ./sub-24/anat/sub-24_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,14 (voxels). ./sub-25/anat/sub-25_T2star.nii.gz The most common set of dimensions is: 384,288,18 (voxels), This file has the dimensions: 384,288,17 (voxels). Please visit https://neurostars.org/search?q=INCONSISTENT_PARAMETERS for existing conversations about this issue. Summary: Available Tasks: Available Modalities: 29 Files, 36.97MB MRI 25 - Subjects 1 - Session If you have any questions, please post on https://neurostars.org/tags/bids. ```

@mguaypaq could you please create a repository for marseille-t2s-template

Thank you

[vcallot]

Description

Dataset acquired in 25 HCs at 3T T2Star sequence from C1 to L2, contains GM/SC/CSF masks. (available in OSF)

@Nilser3 @jcohenadad

I am happy to see that the 3T T2* data could be of interest.

When using (part of) this dataset, please DO NOT FORGET to refer to its DOI (10.17605/OSF.IO/YMRGK), and to the 2 associated publications (Fradet et al, 2014 , doi: 10.1097/BRS.0000000000000125 and Taso et al, 2014, doi: 10.1007/s10334-013-0403-6 ) (and to keep me updated ideally 😜). Thanks!

[jcohenadad]

yes good point-- @Nilser3 can you pls add this info to the README

[Nilser3]

Thanks you @vcallot , @jcohenadad Of course, I have this first draft for the README:

# Dataset marseille-t2s-template

Content: Sub-millimetric cross-sectional T2*-weighted MR images of the whole spinal cord (C1 to L2), acquired in vivo, and allowing to discriminate white and gray matter
Masks from manual segmentation of White Matter (WM), Gray Matter (GM) and Cerebro Spinal Fuild (CSF)
Supplementary data (such as age, body mass index, sex, etc..) can be provided upon request (please contact virginie.callot@univ-amu.fr))

See OSF project: https://osf.io/ymrgk/
See GH repository https://github.com/neuropoly/data-management/issues/279

## Citation

CALLOT, V., LAINES MEDINA, N., Taso, M., & Fradet, L. (2022, December 30). In Vivo Human Spinal Cord MRI data – From cervical to thoraco-lumbar levels. https://doi.org/10.17605/OSF.IO/YMRGK
DOI https://doi.org/10.17605/OSF.IO/YMRGK

## Related works:

Morphometrics of the entire human spinal cord and spinal canal measured from in vivo high-resolution anatomical magnetic resonance imaging. Fradet L, Arnoux PJ, Ranjeva JP, Petit Y, Callot V. Spine (Phila Pa 1976). 2014 Feb 15;39(4):E262-9 doi: 10.1097/BRS.0000000000000125 PMID: 24253776
Construction of an in vivo human spinal cord atlas based on high-resolution MR images at cervical and thoracic levels: preliminary results. Taso M, Le Troter A, Sdika M, Ranjeva JP, Guye M, Bernard M, Callot V. MAGMA. 2014 Jun;27(3):257-67 doi: 10.1007/s10334-013-0403-6 PMID: 24052240

## Dataset structure

Dataset of 25 healthy controls from Marseille, acquired at 3T cross-sectional T2Star-weighted
0.5x0.5 mm2 in-plane resolution

## Naming convention

- sub-XX_T2Star

## derivatives
- sub-XX_T2star_label-CSF_seg
- sub-XX_T2star_label-SC_seg
- sub-xx_T2star_label-GM_seg
- sub-XX_T2star_label-levels

## Authors:

- Virginie Callot,
- Nilser Laines Medina

thank you

[mguaypaq]

Ok, I've created the repository marseille-t2s-template, with the sample files from the wiki, and given write access to @Nilser3. Please add your data in a new branch, push your changes to the server, and let me know once it's ready for review.

[Nilser3]

Done @mguaypaq , I have pushed to my branch nlm/initial_data

[mguaypaq]

Looks good! I configured bids-validator to ignore INCONSISTENT_PARAMETERS and merged into master.

[jcohenadad]

my bad-- sorry

[jcohenadad]

actually-- would be good to also add Manuel Taso and Louis Fradet

[Nilser3]

The authors were added in the branch nlm/update_authors I'm ready for PR

[jcohenadad]

Checking: 5f61ef57b3b9c47366eb76567ca60aa5de3716a8

Binary segmentations are unnecessarily encoded as FLOAT32, eg:

julien-macbook:~/data.neuro/marseille-t2s-template $ fslhd derivatives/labels/sub-01/anat/sub-01_T2star_label-SC_seg.nii.gz 
filename    derivatives/labels/sub-01/anat/sub-01_T2star_label-SC_seg.nii.gz

sizeof_hdr  348
data_type   FLOAT32

I recommend changing them to UINT8.

[Nilser3]

I agree, the mask encoding was changed thanks to sct_image -i IMAGE -type uint8

Last update: 4583ed295d697b0341fee164cdbfda687a840bd7 in the branch nlm/update_authors

[jcohenadad]

4583ed295d697b0341fee164cdbfda687a840bd7 looks good 👍

[mguaypaq]

Great, I merged into master and deleted the branch nlm/update_authors.