Extract ROIs using AAL atlas

465 views
Skip to first unread message

Maria

unread,
Dec 1, 2021, 3:57:22 AM12/1/21
to HCP-Users
Dear HCP experts,


Hello, I'm a newbie who just started using HCP data and wb_command.
I have a question regarding utilizing the AAL atlas to extract ROIs from dlabel.nii format file.

Data: task/resting-state fMRI
Goal: To extract averaged value from each ROIs for each volume
         (i.e., the output dimension would be volumes x the number of ROIs (e.g., 1200 x 116))
Tried process:
1) wb_command -volume-label-import './aal_roi_atlas.nii.gz' ' ' './aal.nii.gz'
2) wb_command -volume-label-to-surface-mapping './aal.nii.gz' './#.R.midthickness.32k_fs_LR.surf.gii' './aal_R_label.gii'
3) wb_command -volume-label-to-surface-mapping './aal.nii.gz' '#.L.midthickness.32k_fs_LR.surf.gii' './aal_L_label.gii'
4) wb_command -cifti-create-dense-from-template './#.aparc.a2009s.32k_fs_LR.dlabel.nii' './aal116.dlabel.nii' -label CORTEX_LEFT './aal_L.label.gii' -label CORTEX_RIGHT './aal_R_label.gii'

Question:
May I ask you the proper method to extract ROIs utilizing another NIFTI format (which unit is voxel) atlas?
The reason is, I assumed that every subject would get the same number of ROIs, such as 116.
However, when I conducted the above sequence, I got the different number of ROIs from different subjects.

In addition, I also tried to conduct warping (written below) the ROIs into individual subject space, but it also returns different numbers of ROIs for different subjects.
applywarp --interp=nn -i <ROIs.nii.gz> -r ${StudyFolder}/${Subject}/MNINonLinear/ROIs/Atlas_ROIs.2.nii.gz -o <ROIs.2.nii> 


I would be thankful if you could help me with this issue.


Thank you,
Mari

‍허다운[ 대학원박사과정재학 / 인공지능학과 ]

unread,
Dec 1, 2021, 10:23:27 AM12/1/21
to hcp-...@humanconnectome.org
Dear HCP experts,


Hello, I'm a newbie who just started using HCP data and wb_command.
I have a question regarding utilizing the AAL atlas to extract ROIs from dlabel.nii format file.

Data: task/resting-state fMRI
Goal: To extract averaged value from each ROIs for each volume
         (i.e., the output dimension would be volumes x the number of ROIs (e.g., 1200 x 116))
Tried process:
1) wb_command -volume-label-import './aal_roi_atlas.nii.gz' ' ' './aal.nii.gz'
2) wb_command -volume-label-to-surface-mapping './aal.nii.gz' './#.R.midthickness.32k_fs_LR.surf.gii' './aal_R_label.gii'
3) wb_command -volume-label-to-surface-mapping './aal.nii.gz' '#.L.midthickness.32k_fs_LR.surf.gii' './aal_L_label.gii'
4) wb_command -cifti-create-dense-from-template './#.aparc.a2009s.32k_fs_LR.dlabel.nii' './aal116.dlabel.nii' -label CORTEX_LEFT './aal_L.label.gii' -label CORTEX_RIGHT './aal_R_label.gii'

Question:
May I ask you the proper method to extract ROIs utilizing another NIFTI format (which unit is voxel) atlas?
The reason is, I assumed that every subject would get the same number of ROIs, such as 116.
However, when I conducted the above sequence, I got different number of ROIs from different subjects.


May I ask you what is wrong with the process?

Thank you in advance, and I hope you have a nice day.


Sincerely,
DaWoon



Glasser, Matt

unread,
Dec 1, 2021, 10:37:27 AM12/1/21
to hcp-...@humanconnectome.org

The biggest problem I see below is that you are using the AAL atlas.  That has little to do with the neurobiological organization of the brain.  Here is a better parcellation:

 

https://balsa.wustl.edu/file/3VLx

 

This should hopefully be back up soon, as there seems to be some IT issues where at Wash U.  I don’t see an immediate issue with your commands otherwise except that you probably need to use a enclosing approach for volume to surface mapping. 

 

Matt.

--
You received this message because you are subscribed to the Google Groups "HCP-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hcp-users+...@humanconnectome.org.
To view this discussion on the web visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CALPODGHgeiWijG3AUrS5eYLMcUiTLD9bZVnasDeqm9nshKWjfw%40mail.gmail.com.

 


The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

LEE

unread,
Dec 2, 2021, 1:29:11 AM12/2/21
to HCP-Users, glas...@wustl.edu
I have a problem with utilizing the Yeo17 atlas to extract ROIs.

After following the above sequence, I got a dlabel file.         Yeo17 (NIFTI format) -> Yeo17.dlabel.nii (CIFTI format)

However I got different number of ROIs from different subjects when I finally use 
wb_command -cifti-parcellate './rfMRI_REST1_LR_Atlas.dtseries.nii' './Yeo17.dlabel.nii' COLUMN './Yeo17.ptseries.nii'
to extract ROIs.

question:
1. Is there any problem with Yeo17 atlas?
2. How can I get Yeo17.dlabel.nii that can make the same dimension for each subject   (i.e., the output dimension would be volumes x the number of ROIs (e.g., 1200 x 114))


2021년 12월 2일 목요일 오전 12시 37분 27초 UTC+9에 glas...@wustl.edu님이 작성:

Glasser, Matt

unread,
Dec 2, 2021, 5:42:05 AM12/2/21
to LEE, HCP-Users

See my point about volume to surface mapping.  The Yeo atlas was computed on the surface, however, so there should be no need to do any of the below stuff. 


Matt.

Maria

unread,
Dec 5, 2021, 10:38:48 PM12/5/21
to HCP-Users, glas...@wustl.edu
My apologies for the late reply.
Thank you for the detailed explanation.
It helped a lot for me to understand.

Thank You,
DaWoon


2021년 12월 2일 목요일 오전 12시 37분 27초 UTC+9에 glas...@wustl.edu님이 작성:

The biggest problem I see below is that you are using the AAL atlas.  That has little to do with the neurobiological organization of the brain.  Here is a better parcellation:

Reply all
Reply to author
Forward
0 new messages