Notice that the first row of the -nifti-information -print-matrix output is
labeled with a 0, and that the count of the output rows is 59412.  It is
zero-indexed.  For extracting the data in a file in text form, you may also
want to look at -cifti-convert -to-text.

I believe the data in the medial wall is excluded from even the label file,
despite the label table of that file including keys for medial wall, so I
suspect a different label file was used for the parcellation.  Someone else
will have to provide the answer here, as I am not familiar with the details
of the parcellated files we have released.

As a side note, you can generate a label file that is mostly equivalent
(will have different colors and exclude unused labels, may have different
label keys) to the label file used to create any parcellated cifti file
with the -cifti-parcel-mapping-to-label command.

Tim


On Sun, Feb 7, 2016 at 6:19 PM, Ari Pinar <[email protected]> wrote:

> Hi HCP users,
>
> I am currently working with rs-fMRI data and have some questions regarding
> the parcellation scheme label file provided by HCP, specifically the file
> named "aparc.a2009s.32k_fs_LR.dlabel.nii"
>
> Firstly, when using the two workbench commands:
>
> "wb_command -file-information 109123.aparc.a2009s.32k_fs_LR.dlabel.nii"
> and
> "wb_command -nifti-information -print-matrix
> 109123.aparc.a2009s.32k_fs_LR.dlabel.nii"
>
> I get a discrepancy between the dimensions of the image file between these
> two commands (59412 using -file-information vs. 59411 with
> -nifti-information). I am thinking this may be related to the manner in
> which these commands utilise 0/1-based indexing, but clarification on this
> would be appreciated.
>
> Secondly, using the -file-information command on this file "
> 109123.aparc.a2009s.32k_fs_LR.dlabel.nii" seems to output a list of
> regions/labels (150 unique labels in total), however when I extract
> timeseries from the HCP provided ICA-FIX cleaned data, the output is
> timeseries from 148 unique regions. I would like to know whether any
> regions have been excluded (medial wall perhaps?), which would account for
> this difference?
>
> Thanks for your assistance.
>
> Best,
>
> Ari
>
> _______________________________________________
> HCP-Users mailing list
> [email protected]
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to