help
47 Subscribers 
help > BrainNet Viewer for structural connectomes
Showing 1-2 of 2 posts
Jan 3, 2023 05:01 PM | rosella trò
BrainNet Viewer for structural connectomes
Hello!
I am a PostDoc Researcher working on pediatric neuroimaging at University of Genoa. First of all, I would like to thank you for providing such a useful tool for connectome visualization. SInce I am new to it, I wanted to make sure to apply all the steps properly.
My goal is representing both individual and consensus connectomes for 2 groups of subjects: controls and patients.
What I do right now is:
- as surface file, merge lh.pial and rh.pial from freesurfer/surf folder for each subject (and for representing the consensus connectome, take a random subject belonging to the group)
- as a node file, I use BrainNet to generate it: "BrainNet_GenCoord('aparc+aseg.nii','subject.node')
where aparc+aseg is taken from freesurfer/mri for each subject (and for representing the consensus connectome, take a random subject belonging to the group)
- as a edge file, I load the ".csv" file obtained from Mrtrix3 for each subject, after thresholding and binarizing it in matlab (and for representing the consensus connectome, take a the binarized consensus ".txt" file generated with Brain Connectivity Matlab Toolbox)
I simply wanted to make sure what I do (both for individual subjects and for consensus) is correct. The resulting connectome I obtain for one example subject is attached in the figures, and I can observe that:
1) nodes do not seem to be perfectly overlaid on the surface (maybe that is just because cerebellum is included in the parcellation, thus causing some nodes to appear outside the pial surface)
2) connections between nodes are "strangely" distributed (i.e., not symmetrical).
In the figures, I selected all nodes and edges with no exclusion/thresholding criteria.
Is the procedure I follow correct and the result thus only depends on the input data, or am I doing something wrong within BNV?
Any hints would be really appreciated.
thanks in advance,
Rosella
I am a PostDoc Researcher working on pediatric neuroimaging at University of Genoa. First of all, I would like to thank you for providing such a useful tool for connectome visualization. SInce I am new to it, I wanted to make sure to apply all the steps properly.
My goal is representing both individual and consensus connectomes for 2 groups of subjects: controls and patients.
What I do right now is:
- as surface file, merge lh.pial and rh.pial from freesurfer/surf folder for each subject (and for representing the consensus connectome, take a random subject belonging to the group)
- as a node file, I use BrainNet to generate it: "BrainNet_GenCoord('aparc+aseg.nii','subject.node')
where aparc+aseg is taken from freesurfer/mri for each subject (and for representing the consensus connectome, take a random subject belonging to the group)
- as a edge file, I load the ".csv" file obtained from Mrtrix3 for each subject, after thresholding and binarizing it in matlab (and for representing the consensus connectome, take a the binarized consensus ".txt" file generated with Brain Connectivity Matlab Toolbox)
I simply wanted to make sure what I do (both for individual subjects and for consensus) is correct. The resulting connectome I obtain for one example subject is attached in the figures, and I can observe that:
1) nodes do not seem to be perfectly overlaid on the surface (maybe that is just because cerebellum is included in the parcellation, thus causing some nodes to appear outside the pial surface)
2) connections between nodes are "strangely" distributed (i.e., not symmetrical).
In the figures, I selected all nodes and edges with no exclusion/thresholding criteria.
Is the procedure I follow correct and the result thus only depends on the input data, or am I doing something wrong within BNV?
Any hints would be really appreciated.
thanks in advance,
Rosella
Jan 4, 2023 09:01 AM | rosella trò
RE: BrainNet Viewer for structural connectomes
Hi all,
this is just to update you that I have tried visualizing the same example subject with Mrtrix3, and it seems input data are correct. As you can observe from the attached picture, aparc+aseg image excludes cerebellum and thus some nodes are outside it. The surface I use with BNV is created from this segmentation, so I think the outcome I obtain is expected.
Thank you in advance for any suggestion/confirmation.
Best regards,
Rosella
this is just to update you that I have tried visualizing the same example subject with Mrtrix3, and it seems input data are correct. As you can observe from the attached picture, aparc+aseg image excludes cerebellum and thus some nodes are outside it. The surface I use with BNV is created from this segmentation, so I think the outcome I obtain is expected.
Thank you in advance for any suggestion/confirmation.
Best regards,
Rosella