[HCP-Users] Estimation of surface area

[Paul, Subhadip]

Dear HCP users,

HCP  has released thickness, surface area values based on DK (Desikan-Killiany) 
atlas in a csv file. I would like to extract those values using Destrieux 
atlas. To verify, whether am I following the correct procedure, I have 
extracted regional cortical thickness data using MATLAB from the HCP directory 
(/HCP_1200/100206/MNINonLinear/Native/) using DK atlas and compared with HCP 
released values. The values are same. Now, I am using connectome-workbench to 
estimate regional surface area using DK atlas. However, the extracted value 
does not match with HCP provided value.

To extract the area of left bankssts area (key 1) from 
/HCP_1200/100206/MNINonLinear/Native/, I am using the following connectome 
workbench commands:

wb_command -gifti-label-to-roi 100206.L.aparc.native.label.gii roi.func.gii 
-key 1

wb_command -metric-weighted-stats roi.func.gii -sum -area-surface 
100206.L.midthickness.native.surf.gii -show-map-name

Could you please tell me whether am I using the correct commands? When I used 
*pial*.surf.gii or *white*.surf.gii instead of *midthickness*.surf.gii file in 
the second command, still the extracted value does not match with the HCP 
released value.

Once I figure out the correct approach, I would replace the *aparc.*.label.gii 
file in the first command, with *aparc.a2009s*.label.gii to extract surface 
area using Destrieux atlas.

Thanking you,

Subhadip



Subhadip Paul

King's College London

16 De Crespigny Park, London SE5 8AF

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] Estimation of surface area

[Timothy Coalson]

Those commands look correct, however MNI space may not be the correct place
to make these measurements, as the registration distorts the subject
anatomy to match a template (and the template happens to be significantly
larger than the average subject).  I suspect we do not actually recompute
thickness in MNI space, so a thickness file in the MNINonLinear folder may
simply be a copy from the T1w folder.  Similarly, I would expect that our
released values for surface area are measured in T1w space, which is a
rigidly aligned, distortion corrected scanner space, matching the true
shape of the subject's brain.

Note that with cifti files, you can use -cifti-parcellate to do these
per-area summaries for all areas at once, rather than one at a time.  You
can use -cifti-convert -to-text to dump the data values after that (and
-file-information will tell you the order of areas in the parcellated file).

Tim


On Wed, Apr 18, 2018 at 1:07 PM, Paul, Subhadip 
wrote:

> Dear HCP users,
>
> HCP  has released thickness, surface area values based on DK
> (Desikan-Killiany) atlas in a csv file. I would like to extract those
> values using Destrieux atlas. To verify, whether am I following the correct
> procedure, I have extracted regional cortical thickness data using MATLAB
> from the HCP directory (/HCP_1200/100206/MNINonLinear/Native/) using DK
> atlas and compared with HCP released values. The values are same. Now, I am
> using connectome-workbench to estimate regional surface area using DK
> atlas. However, the extracted value does not match with HCP provided value.
>
> To extract the area of left bankssts area (key 1) from
> /HCP_1200/100206/MNINonLinear/Native/, I am using the following
> connectome workbench commands:
>
> wb_command -gifti-label-to-roi 100206.L.aparc.native.label.gii
> roi.func.gii -key 1
>
> wb_command -metric-weighted-stats roi.func.gii -sum -area-surface
> 100206.L.midthickness.native.surf.gii -show-map-name
>
> Could you please tell me whether am I using the correct commands? When I
> used *pial*.surf.gii or *white*.surf.gii instead of *midthickness*.surf.gii
> file in the second command, still the extracted value does not match with
> the HCP released value.
>
> Once I figure out the correct approach, I would replace the
> *aparc.*.label.gii file in the first command, with *aparc.a2009s*.label.gii
> to extract surface area using Destrieux atlas.
>
> Thanking you,
>
> Subhadip
>
>
> *Subhadip Paul*
>
> King's College London
>
> 16 De Crespigny Park, London
> SE5
> 8AF
> 
>
>
> ___
> HCP-Users mailing list
> HCP-Users@humanconnectome.org
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] Estimation of surface area

[Glasser, Matthew]

That is correct.  MNI space IS NOT the correct place to measure areas, volumes, 
thicknesses.

Peace,

Matt.

From: 
>
 on behalf of Timothy Coalson >
Date: Wednesday, April 18, 2018 at 1:25 PM
To: "Paul, Subhadip" >
Cc: "hcp-users@humanconnectome.org" 
>
Subject: Re: [HCP-Users] Estimation of surface area

Those commands look correct, however MNI space may not be the correct place to 
make these measurements, as the registration distorts the subject anatomy to 
match a template (and the template happens to be significantly larger than the 
average subject).  I suspect we do not actually recompute thickness in MNI 
space, so a thickness file in the MNINonLinear folder may simply be a copy from 
the T1w folder.  Similarly, I would expect that our released values for surface 
area are measured in T1w space, which is a rigidly aligned, distortion 
corrected scanner space, matching the true shape of the subject's brain.

Note that with cifti files, you can use -cifti-parcellate to do these per-area 
summaries for all areas at once, rather than one at a time.  You can use 
-cifti-convert -to-text to dump the data values after that (and 
-file-information will tell you the order of areas in the parcellated file).

Tim


On Wed, Apr 18, 2018 at 1:07 PM, Paul, Subhadip 
> wrote:

Dear HCP users,

HCP  has released thickness, surface area values based on DK (Desikan-Killiany) 
atlas in a csv file. I would like to extract those values using Destrieux 
atlas. To verify, whether am I following the correct procedure, I have 
extracted regional cortical thickness data using MATLAB from the HCP directory 
(/HCP_1200/100206/MNINonLinear/Native/) using DK atlas and compared with HCP 
released values. The values are same. Now, I am using connectome-workbench to 
estimate regional surface area using DK atlas. However, the extracted value 
does not match with HCP provided value.

To extract the area of left bankssts area (key 1) from 
/HCP_1200/100206/MNINonLinear/Native/, I am using the following connectome 
workbench commands:

wb_command -gifti-label-to-roi 100206.L.aparc.native.label.gii roi.func.gii 
-key 1

wb_command -metric-weighted-stats roi.func.gii -sum -area-surface 
100206.L.midthickness.native.surf.gii -show-map-name

Could you please tell me whether am I using the correct commands? When I used 
*pial*.surf.gii or *white*.surf.gii instead of *midthickness*.surf.gii file in 
the second command, still the extracted value does not match with the HCP 
released value.

Once I figure out the correct approach, I would replace the *aparc.*.label.gii 
file in the first command, with *aparc.a2009s*.label.gii to extract surface 
area using Destrieux atlas.

Thanking you,

Subhadip



Subhadip Paul

King's College London

16 De Crespigny Park, London 

 SE5 
8AF

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users


___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] surface with and without MSMAll

[Xinyang Liu]

Ok, I've made clear about the reason. The T1w image opened from the workbench 
tutorial data has also been registered to the MNI space. Therefore, the MNI 
surface matches well with this T1w_restore.nii.gz image in MNI space, and the 
T1w surface matches well with the T1_acpc_dc_restore.nii.gz image. 


Problem solved. Sorry to disturb with such a hurried asking.
Wish everyone a good day. :)


Best regards,
Xinyang



在 2018-04-18 16:00:36，"Xinyang Liu"  写道：

Dear experts,


Hi. Sorry that I made a mistake in my last email. 
The "100307.L.white.32k_fs_LR.surf.gii" was provided from the MNI folder in the 
workbench tutorial data, and the MSMAll file came from T1w folder. There is 
only tiny difference from the MSMAll registration.


Therefore, the difference actually lies in the surfaces from the MNI and T1w 
folder.  It turns out that the surfaces with the same name from these two 
folders differ a lot in their shapes (please see the new screenshot attached), 
which I didn't expect to be so obvious. We can see that the MNI surfaces have 
better alignment with individual brain structure, not T1w. Why does that 
happen? In this way, should we change to choose the surface in MNI folder 
instead of T1w to produce surface ROI in fibre tracking?


Look forward to your reply. Thanks.


Best regards,
Xinyang



At 2018-04-18 14:59:10, "Xinyang Liu"  wrote:

Dear HCP experts,


Hi. In previous discussions about using surface ROIs in fibre tracking, 
surfaces in ${StudyFolder}/${Subject}/T1w/fsaverage_LR32k with MSMAll was 
recommended.


However, I just found that the surface files with and without MSMAll 
registration differ a lot within individuals. I put a screenshot attached, 
which includes "100307.L.white_MSMAll.32k_fs_LR.surf.gii" and 
"100307.L.white.32k_fs_LR.surf.gii". From the white surface outlines, we can 
see that the surface without MSMAll registration 
(100307.L.white.32k_fs_LR.surf.gii) has better alignment with the brain GM/WM 
structure.


Could you please tell me the reason why we should use MSMAll surface in 
individual fibre tracking? Thank you very much.


Best regards,
Xinyang





 

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users




【网易自营|30天无忧退货】爱上书写：施华洛世奇制造商星空原色水晶笔，限时仅29元>>




 

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users
___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

[HCP-Users] Native space surface curvature available?

[Jacob Miller]

Hi all,

I am seeking to perform an anatomical analysis on the HCP dataset which
requires the surfaces and cortical folding (gyri/sulci curvature) for each
individual subject, in native space. This information is available in a
typical FreeSurfer recon-all analysis. However, in the HCP data (namely, in
$subject/T1w/Native/) the surfaces do not appear to have curvature
information. Was this information released in the data from the modified
HCP FreeSurfer pipelines, and I'm missing where this may be stored? Or, are
the gyru/sulci curvature for each subject not available? If so, is there a
way to obtain these curvatures without running an entire recon-all pipeline
on each native T1 scan?

Please let me know if I may be mistaken! I've only just begun trying to use
this great dataset.

Best,
Jacob

-- 

Jacob Miller
Graduate Student, D'Esposito Lab
Helen Wills Neuroscience Institute (HWNI)
University of California, Berkeley

https://despolab.berkeley.edu/

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] Native space surface curvature available?

[Harms, Michael]

Hi,
The entirety of the original FS output is available at 
${subject}/T1w/${subject}, which you can obtain via the “Structural Extended” 
packages, or as part of “Connectome-in-a-Box” (no longer taking new orders, but 
perhaps you already have one) or available on S3.

Cheers,
-MH

--
Michael Harms, Ph.D.
---
Associate Professor of Psychiatry
Washington University School of Medicine
Department of Psychiatry, Box 8134
660 South Euclid Ave.Tel: 314-747-6173
St. Louis, MO  63110  Email: mha...@wustl.edu

From:  on behalf of Jacob Miller 

Date: Wednesday, April 18, 2018 at 9:04 PM
To: "hcp-users@humanconnectome.org" 
Subject: [HCP-Users] Native space surface curvature available?

Hi all,

I am seeking to perform an anatomical analysis on the HCP dataset which 
requires the surfaces and cortical folding (gyri/sulci curvature) for each 
individual subject, in native space. This information is available in a typical 
FreeSurfer recon-all analysis. However, in the HCP data (namely, in 
$subject/T1w/Native/) the surfaces do not appear to have curvature information. 
Was this information released in the data from the modified HCP FreeSurfer 
pipelines, and I'm missing where this may be stored? Or, are the gyru/sulci 
curvature for each subject not available? If so, is there a way to obtain these 
curvatures without running an entire recon-all pipeline on each native T1 scan?

Please let me know if I may be mistaken! I've only just begun trying to use 
this great dataset.

Best,
Jacob

--

Jacob Miller

Graduate Student, D'Esposito Lab

Helen Wills Neuroscience Institute (HWNI)

University of California, Berkeley

https://despolab.berkeley.edu/

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users


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.

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users