External Email - Use Caution
Dear FreeSurfer-Experts,
I’m in need of your experience to help me with a decision in my current project:
I want to use certain areas, that were identified by structural graph theoretical analyses based on the FreeSurfer parcellation and DTI data, as Regions of interest in SPM. We already have a script that helps with the transfer and creation of ROIs however it is still unclear what’s the best way to create the ROIs. Here are a few of my ideas and pros and cons regarding these:
1. Parcel the MNI T1 standard brain used in SPM with FreeSurfer and create the ROIs from the identified regions from that PRO: direct use of the standard space, that is later used in SPM’s group level analysis CON: SPM tends to be peculiar in giving information about their standard template, so I am unsure if the one I found in the spm12-folder is the correct one to use
2. Extract the ROIs of each participant individually and then use SPM to normalize and averaging/summing these PRO: Ending up in same space as the SPM analysis CON: identified regions are (at the moment) not individually identified, although they could be; normalization is difficult with just ROIs ALTERNATIVE: use FreeSurfer not on individual “raw” T1 images but on already normalized T1 images and then do the ROIs individually per participant.
Probably, I am also forgetting another possibility. I am looking forward to hearing your ideas and comments. Thank you very much in advance, Best wishes, Falko
I would probably just use the 1st option.
On 3/30/2022 5:20 AM, Falko Mecklenbrauck wrote:
External Email - Use Caution
Dear FreeSurfer-Experts,
I’m in need of your experience to help me with a decision in my current project:
I want to use certain areas, that were identified by structural graph theoretical analyses based on the FreeSurfer parcellation and DTI data, as Regions of interest in SPM. We already have a script that helps with the transfer and creation of ROIs however it is still unclear what’s the best way to create the ROIs. Here are a few of my ideas and pros and cons regarding these:
- Parcel the MNI T1 standard brain used in SPM with FreeSurfer and
create the ROIs from the identified regions from that PRO: direct use of the standard space, that is later used in SPM’s group level analysis CON: SPM tends to be peculiar in giving information about their standard template, so I am unsure if the one I found in the spm12-folder is the correct one to use
- Extract the ROIs of each participant individually and then use SPM
to normalize and averaging/summing these PRO: Ending up in same space as the SPM analysis CON: identified regions are (at the moment) not individually identified, although they could be; normalization is difficult with just ROIs ALTERNATIVE: use FreeSurfer not on individual “raw” T1 images but on already normalized T1 images and then do the ROIs individually per participant.
Probably, I am also forgetting another possibility. I am looking forward to hearing your ideas and comments. Thank you very much in advance, Best wishes, Falko
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas N. Greve -
Falko Mecklenbrauck