AMU7T: 3D qT1 and T2*w in vivo template + atlas

AMU7T : high-resolution multimodal quantitative T1 and T2*-weighted templates with white and gray matter spinal cord and substructures atlas, aligned with the PAM50 space

Data

This data repository contains:

• template/AMU26_qt1.nii.gz: 7T qT1 template from 26 subjects.
• template/AMU72_t2s.nii.gz: 7T T2*w template from 72 subjects.
• template/AMU7T_gm.nii.gz: Probabilistic gray matter mask.
• template/AMU7T_wm.nii.gz: Probabilistic white matter mask.
• template/AMU7T_sc.nii.gz: Probabilistic spinal cord mask.
• atlas/AMU7T_50_labels.nii.gz: Atlas with 50 labels (WM parcels from Lévy et al., NeuroImage (2015)., and GM parcels derivated from Hausman, L. Atlases of the Spinal Cord and Brainstem and the Forebrain. (Thomas, 1962).
• atlas/AMU7T_atlas_ID_bin.nii.gz: Binary masks by ID.

Labels: Multilabel atlas in in atlas/AMU7T_50_labels.nii.gz containing:

ID - Region

• 0 - WM Left_fasciculus-gracilis
• 1 - WM Right_fasciculus-gracilis
• 2 - WM Left_fasciculus-cuneatus
• 3 - WM Right_fasciculus-cuneatus
• 4 - WM Left_lateral-corticospinal
• 5 - WM Right_lateral-corticospinal
• 6 - WM Left_ventral-spinocerebellar
• 7 - WM Right_ventral-spinocerebellar
• 8 - WM Left_rubrospinal
• 9 - WM Right_rubrospinal
• 10 - WM Left_lateral-reticulospinal
• 11 - WM Right_lateral-reticulospinal
• 12 - WM Left_spinal-lemniscus-spinothalamic
• 13 - WM Right_spinal-lemniscus-spinothalamic
• 14 - WM Left_spino-olivary
• 15 - WM Right_spino-olivary
• 16 - WM Left_ventrolateral-reticulospinal
• 17 - WM Right_ventrolateral-reticulospinal
• 18 - WM Left_lateral-vestibulospinal
• 19 - WM Right_lateral-vestibulospinal
• 20 - WM Left_ventral-reticulospinal
• 21 - WM Right_ventral-reticulospinal
• 22 - WM Left_ventral-corticospinal
• 23 - WM Right_ventral-corticospinal
• 24 - WM Left_tectospinal
• 25 - WM Right_tectospinal
• 26 - WM Left_medial-reticulospinal
• 27 - WM Right_medial-reticulospinal
• 28 - WM Left_medial-longitudinal-fasciculus
• 29 - WM Right_medial-longitudinal-fasciculus
• 30 - anterior-fissure
• 31 - septum
• 32 - GM Left_medial-ventral-horn
• 33 - GM Left_central-ventral-horn
• 34 - GM Left_lateral-ventral-horn
• 35 - GM Right_medial-ventral-horn
• 36 - GM Right_central-ventral-horn
• 37 - GM Right_lateral-ventral-horn
• 38 - GM Left_dorsal-intermediate-zone
• 39 - GM Left_ventral-intermediate-zone
• 40 - GM Right_dorsal-intermediate-zone
• 41 - GM Right_ventral-intermediate-zone
• 42 - GM Left_reticular-formation
• 43 - GM Left_dorsomarginal-nucleus
• 44 - GM Left_fasciculus-dorsolateralis
• 45 - GM Right_reticular-formation
• 46 - GM Right_dorsomarginal-nucleus
• 47 - GM Right_fasciculus-dorsolateralis
• 48 - central-canal
• 49 - CSF contour

Combined labels

• 50 - GM Left_ventral-horn, 32:34
• 51 - GM Right_ventral-horn, 35:37
• 52 - GM Left_intermediate-zone, 38:39
• 53 - GM Right_intermediate-zone, 40:41
• 54 - GM Left_dorsal-horn, 42:44
• 55 - GM Right_dorsal-horn, 45:47
• 56 - spinal-cord, 0:48
• 57 - white-matter, 0:31
• 58 - gray-matter, 32:48
• 59 - dorsal-columns, 0:3
• 60 - lateral-funiculi, 4:13
• 61 - ventral-funiculi, 14:29

The intensities of each label in atlas/AMU7T_50_labels.nii.gz are detailed here, and the parcellation can be visualized with this Lookup table in FSLeyes and ITK-SNAP

Usage: Registration of AMU7T to Subject space

For an optimal registration of the AMU7T, it is recommended to work with a WM mask that covers the dorsal horns up to the CSF (see the the animation below).

Optimal parameters for register AMU7T to subject space

sct_register_to_template -i image.nii.gz -s image_wm.nii.gz -l landmarks.nii.gz -c t1 -v 1 -s-template-id 4  -t ../template_AMU7T  -param step=1,type=imseg,algo=centermassrot,rot_method=pcahog:step=2,type=seg,algo=bsplinesyn,slicewise=0,metric=MeanSquares,samplStrategy=None,samplPercent=0.2,iter=2,smooth=1,rot_method=pcahog:step=3,type=seg,algo=syn,metric=MeanSquares,shrink=2,dof=Tz_Rz_Sz,slicewise=1,iter=20 -ref subject

Atlas registration to subject space

sct_apply_transfo -i template_AMU7T/atlas/AMU7T_50_labels.nii.gz -d image.nii.gz -w warp_template2anat.nii.gz -o labels_AMU7T2anat.nii.gz -x nn

Other uses with SCT commands

sct_warp_template -d image.nii.gz -w warp_template2anat.nii.gz  -t template_AMU7T
sct_extract_metric -i image.nii.gz -f label/atlas  -method wa -o wa_T1.csv -z 2:17 -perslice 1

Python script for registrater AMU7T template to subject space

python template_AMU7T/code/AMU7T_registration.py --path_image image.nii.gz --mask_wm image_wm.nii.gz --landmarks landmarks.nii.gz --path_template_AMU7T template_AMU7T/ --path_output OUTPUT

Related issues

#13 #4 #PR14

How to cite

Le Troter, A., Laines Medina, N., Mchinda, S., Cohen-Adad, J., & Callot, V. (June 2023). AMU7T : a 3D qT1 and T2*w high-resolution in vivo template with refined white and gray matter parcellation dedicated to 7T spinal cord MR analyses [Oral presentation - p. 0569]. ISMRM & ISMRT annual Meeting & Exhibition, Toronto, ON, Canada.