External Email - Use Caution
Dear FreeSurfer experts,
I have two questions on subject IDs in FreeSurfer output.
1) is it possible to change to subject identifier in the FreeSurfer output after recon-all has been run?
Background: I would like to upload data pre-processed with FreeSurfer to a consortium server. The subject ID is a random identifier, but the upload guidelines say this identifier must only occur in certain file types (Excel files, .log files, and some more, but not in any other files).
I noticed that a lot of the output files produced by FreeSurfer contain the subject identifier somewhere, including various binary files, so I guess there is no easy way to change it. But maybe there is? Like rerunning a part of recon-all?
2) If not, I could rename the source NIFTI files (e.g., from the identifiers to something like 'subject001', 'subject002', ...) and re-process everything. This would take some computational time for the > 500 subjects, but it would be okay I guess.
But a large number of the subjects have manual edits applied already. Is there a way to keep the manual edits? E.g., I thought maybe I could rename the individual directories of the edited subjects, leave the edited files in there, and run recon-all again. But what would happen? Will the new output files have the renamed ID from the directory name (and thus the recon-all command line), or will they still use the old ID (from the header of the existing files)? Or will the existing, edited files which contain a different ID be ignored during the new run because the IDs do not match, and thus the existing edits would have no effect?
All the best,
Tim
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Hi Tim, can you send a list of files that have the identifier. For volumes and surfaces, it might be as easy as running mri_convert/mris_convert using the same file as input and output.
On 1/13/2020 5:54 AM, Tim Schäfer wrote:
External Email - Use CautionDear FreeSurfer experts,
I have two questions on subject IDs in FreeSurfer output.
- is it possible to change to subject identifier in the FreeSurfer output after recon-all has been run?
Background: I would like to upload data pre-processed with FreeSurfer to a consortium server. The subject ID is a random identifier, but the upload guidelines say this identifier must only occur in certain file types (Excel files, .log files, and some more, but not in any other files).
I noticed that a lot of the output files produced by FreeSurfer contain the subject identifier somewhere, including various binary files, so I guess there is no easy way to change it. But maybe there is? Like rerunning a part of recon-all?
- If not, I could rename the source NIFTI files (e.g., from the identifiers to something like 'subject001', 'subject002', ...) and re-process everything. This would take some computational time for the > 500 subjects, but it would be okay I guess.
But a large number of the subjects have manual edits applied already. Is there a way to keep the manual edits? E.g., I thought maybe I could rename the individual directories of the edited subjects, leave the edited files in there, and run recon-all again. But what would happen? Will the new output files have the renamed ID from the directory name (and thus the recon-all command line), or will they still use the old ID (from the header of the existing files)? Or will the existing, edited files which contain a different ID be ignored during the new run because the IDs do not match, and thus the existing edits would have no effect?
All the best,
Tim
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
External Email - Use Caution
Hi Douglas,
sure, you can get two example files here:
wget http://rcmd.org/tmp/brain.mgz wget http://rcmd.org/tmp/lh.BA1_exvivo.label
The brain volume contains the ID because it stores the full path to the tairach xfm (which includes the folder name, so the ID). You can check by:
mri_info brain.mgz | grep talair
(In this case the ID is 'tim').
Without mri_info, you can get it as well:
mv brain.mgz brain.gz gunzip -c brain.gz | strings | grep talair
The second file is an ASCII label from the same subject. You can get the ID by running:
head -n 1 lh.BA1_exvivo.label
There are some other files which contain the ID, these are just 2 examples.
I am currently in contact with the people running the consortium server and I am not really sure whether the rule that the ID must not be in the files makes any sense at all. Maybe I can get around this.
Best,
Tim
On January 17, 2020 at 12:23 AM "Greve, Douglas N.,Ph.D." DGREVE@mgh.harvard.edu wrote:
Hi Tim, can you send a list of files that have the identifier. For volumes and surfaces, it might be as easy as running mri_convert/mris_convert using the same file as input and output.
On 1/13/2020 5:54 AM, Tim Schäfer wrote:
External Email - Use CautionDear FreeSurfer experts,
I have two questions on subject IDs in FreeSurfer output.
- is it possible to change to subject identifier in the FreeSurfer output after recon-all has been run?
Background: I would like to upload data pre-processed with FreeSurfer to a consortium server. The subject ID is a random identifier, but the upload guidelines say this identifier must only occur in certain file types (Excel files, .log files, and some more, but not in any other files).
I noticed that a lot of the output files produced by FreeSurfer contain the subject identifier somewhere, including various binary files, so I guess there is no easy way to change it. But maybe there is? Like rerunning a part of recon-all?
- If not, I could rename the source NIFTI files (e.g., from the identifiers to something like 'subject001', 'subject002', ...) and re-process everything. This would take some computational time for the > 500 subjects, but it would be okay I guess.
But a large number of the subjects have manual edits applied already. Is there a way to keep the manual edits? E.g., I thought maybe I could rename the individual directories of the edited subjects, leave the edited files in there, and run recon-all again. But what would happen? Will the new output files have the renamed ID from the directory name (and thus the recon-all command line), or will they still use the old ID (from the header of the existing files)? Or will the existing, edited files which contain a different ID be ignored during the new run because the IDs do not match, and thus the existing edits would have no effect?
All the best,
Tim
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
For the mgz volumes, you should be able to do something like mri_add_xform_to_header -c /new/path/to/xfm brain.mgz
For the labels you could just use something like sed s/oldsubjectname/newsubjectname
On 1/17/2020 3:48 AM, Tim Schäfer wrote:
External Email - Use CautionHi Douglas,
sure, you can get two example files here:
wget http://rcmd.org/tmp/brain.mgz wget http://rcmd.org/tmp/lh.BA1_exvivo.labelThe brain volume contains the ID because it stores the full path to the tairach xfm (which includes the folder name, so the ID). You can check by:
mri_info brain.mgz | grep talair(In this case the ID is 'tim').
Without mri_info, you can get it as well:
mv brain.mgz brain.gz gunzip -c brain.gz | strings | grep talairThe second file is an ASCII label from the same subject. You can get the ID by running:
head -n 1 lh.BA1_exvivo.labelThere are some other files which contain the ID, these are just 2 examples.
I am currently in contact with the people running the consortium server and I am not really sure whether the rule that the ID must not be in the files makes any sense at all. Maybe I can get around this.
Best,
Tim
On January 17, 2020 at 12:23 AM "Greve, Douglas N.,Ph.D." DGREVE@mgh.harvard.edu wrote:
Hi Tim, can you send a list of files that have the identifier. For volumes and surfaces, it might be as easy as running mri_convert/mris_convert using the same file as input and output.
On 1/13/2020 5:54 AM, Tim Schäfer wrote:
External Email - Use CautionDear FreeSurfer experts,
I have two questions on subject IDs in FreeSurfer output.
- is it possible to change to subject identifier in the FreeSurfer output after recon-all has been run?
Background: I would like to upload data pre-processed with FreeSurfer to a consortium server. The subject ID is a random identifier, but the upload guidelines say this identifier must only occur in certain file types (Excel files, .log files, and some more, but not in any other files).
I noticed that a lot of the output files produced by FreeSurfer contain the subject identifier somewhere, including various binary files, so I guess there is no easy way to change it. But maybe there is? Like rerunning a part of recon-all?
- If not, I could rename the source NIFTI files (e.g., from the identifiers to something like 'subject001', 'subject002', ...) and re-process everything. This would take some computational time for the > 500 subjects, but it would be okay I guess.
But a large number of the subjects have manual edits applied already. Is there a way to keep the manual edits? E.g., I thought maybe I could rename the individual directories of the edited subjects, leave the edited files in there, and run recon-all again. But what would happen? Will the new output files have the renamed ID from the directory name (and thus the recon-all command line), or will they still use the old ID (from the header of the existing files)? Or will the existing, edited files which contain a different ID be ignored during the new run because the IDs do not match, and thus the existing edits would have no effect?
All the best,
Tim
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
External Email - Use Caution
Hi Doug,
thanks for the mri_add_xform_to_header idea, I didn't know about that option!
Best,
Tim
On January 21, 2020 at 4:24 PM "Greve, Douglas N.,Ph.D." DGREVE@mgh.harvard.edu wrote:
For the mgz volumes, you should be able to do something like mri_add_xform_to_header -c /new/path/to/xfm brain.mgz
For the labels you could just use something like sed s/oldsubjectname/newsubjectname
On 1/17/2020 3:48 AM, Tim Schäfer wrote:
External Email - Use CautionHi Douglas,
sure, you can get two example files here:
wget http://rcmd.org/tmp/brain.mgz wget http://rcmd.org/tmp/lh.BA1_exvivo.labelThe brain volume contains the ID because it stores the full path to the tairach xfm (which includes the folder name, so the ID). You can check by:
mri_info brain.mgz | grep talair(In this case the ID is 'tim').
Without mri_info, you can get it as well:
mv brain.mgz brain.gz gunzip -c brain.gz | strings | grep talairThe second file is an ASCII label from the same subject. You can get the ID by running:
head -n 1 lh.BA1_exvivo.labelThere are some other files which contain the ID, these are just 2 examples.
I am currently in contact with the people running the consortium server and I am not really sure whether the rule that the ID must not be in the files makes any sense at all. Maybe I can get around this.
Best,
Tim
On January 17, 2020 at 12:23 AM "Greve, Douglas N.,Ph.D." DGREVE@mgh.harvard.edu wrote:
Hi Tim, can you send a list of files that have the identifier. For volumes and surfaces, it might be as easy as running mri_convert/mris_convert using the same file as input and output.
On 1/13/2020 5:54 AM, Tim Schäfer wrote:
External Email - Use CautionDear FreeSurfer experts,
I have two questions on subject IDs in FreeSurfer output.
- is it possible to change to subject identifier in the FreeSurfer output after recon-all has been run?
Background: I would like to upload data pre-processed with FreeSurfer to a consortium server. The subject ID is a random identifier, but the upload guidelines say this identifier must only occur in certain file types (Excel files, .log files, and some more, but not in any other files).
I noticed that a lot of the output files produced by FreeSurfer contain the subject identifier somewhere, including various binary files, so I guess there is no easy way to change it. But maybe there is? Like rerunning a part of recon-all?
- If not, I could rename the source NIFTI files (e.g., from the identifiers to something like 'subject001', 'subject002', ...) and re-process everything. This would take some computational time for the > 500 subjects, but it would be okay I guess.
But a large number of the subjects have manual edits applied already. Is there a way to keep the manual edits? E.g., I thought maybe I could rename the individual directories of the edited subjects, leave the edited files in there, and run recon-all again. But what would happen? Will the new output files have the renamed ID from the directory name (and thus the recon-all command line), or will they still use the old ID (from the header of the existing files)? Or will the existing, edited files which contain a different ID be ignored during the new run because the IDs do not match, and thus the existing edits would have no effect?
All the best,
Tim
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany
freesurfer@nmr.mgh.harvard.edu
-
Greve, Douglas N.,Ph.D. -
Tim Schäfer