Dear Martin and Bruce,
thank you very much for your detailed answers! This helps a lot!
Best,
Janosch
Am 15.01.2014 um 21:53 schrieb Martin Reuter mreuter@nmr.mgh.harvard.edu:
Hi Janosch,
there is no fixed naming convention for subjectids from our side, only recommendations. If you like you can call the first time point of subject1 'mike' and the second time point of the same subject 'peter' , we don't care, but it would be confusing.
We recommend a convention where you have
subjectid_tp1 subjectid_tp2
or subjectid01 ... or subjectidA subjectidB
etc.
that makes things easier, but is not necessary.
For cross sectional processing you can place these wherever you want and process a subject_tp at a time in a different location if you like. For longitudinal processing you need to create a single directory where all of these are symlinked (we only need read access). You can do this on a per-subject-basis or for all subjects which is often simpler. The -base run then creates additional directories that may look like subjectid or subjectid_base or subjectid_template (depending on how you name it).
The -long runs finally create additional directories that look like subjectid_tp1.base.subjectid (or ...base.subjectid_tempalte again depending on how you named the base).
If you want (for further postprocessing scripts) you can create a different directory structure similar to what you did for the cross sectional runs and symlinking the *.long.* into that structure, but any processing needs to be done in the single directory that has all time points and base and long runs.
Hope that helps, Best, Martin
On 01/15/2014 01:40 PM, Bruce Fischl wrote:
I'll leave this for Martin Bruce
On Wed, 15 Jan 2014, Janosch Linkersdörfer wrote:
Hi Bruce,
thank you very much for your answer!
Am 15.01.2014 um 14:33 schrieb Bruce Fischl fischl@nmr.mgh.harvard.edu:
Hi Janosch
it will certainly make your life easier if you don't change the names.
So the subject names are saved in other places/files than the folder name?
Symlinking is fine if you keep the same name, but particularly for the longitudinal runs the names have meaning and are stored in files like the tps one which stores the timepoints that went into the base.
OK, so if I understand you correctly, the folder structure for the initial processing does not matter as I can collect different subjects from different places in one $SUBJECTS_DIR for doing (cross-sectional or longitudinal) stats by either copying or symlinking their subject folders.
But the name of the symlink has to be the same as the one used in the initial processing, so I would have to include the time point in the subject name even in the initial processing?
BTW, does one have to use your naming scheme, i.e., "tp1SUBJECTID" or is this flexible, e.g., "tp0_SUBJECTID", "SUBJECTID_tp1", "2011__SUBJECTID", etc.?
Thanks a lot,
Janosch
cheers Bruce On Wed, 15 Jan 2014, Janosch Linkersdörfer wrote:
Hi,
sorry if my question was to basic, but I would really appreciate if somebody could give me some insight into whether the folder organization and the subject name used in the first run of recon-all have to be stable over all all following processing steps or if one can, e.g., change the names of the subject folders and/or symlink them to another location (with another name) and continue with further processing.
Thanks,
Janosch
Am 13.01.2014 um 13:34 schrieb Janosch Linkersdörfer notes4januz@googlemail.com:
Hi all,
I have 4 years of structural scans from children and 2 from adults. I would like to analyze the data both cross-sectionally and longitudinally. In the recommendations for the longitudinal analysis, it says one should process all images in one folder with the time point in the subjectID/subject folder, e.g.,
.../data/tp1_subj001 .../data/tp2_subj001 .../data/tp3_subj001 .../data/tp4_subj001 .../data/tp1_subj002 ...
For the cross-sectional analyses, I would like to organize the analyses in folders for subject age and year, e.g.,
.../data/children/tp1/subj001 .../data/children/tp1/subj002 .../data/children/tp2/subj001 .../data/children/tp2/subj002 ... .../data/adults/tp1/adult_subj001 .../data/adults/tp1/adult_subj002 .../data/adults/tp2/adult_subj001 .../data/adults/tp2/adult_subj001
Is it possible to organize the data this way and to symlink the individual folders into another folder for the longitudinal analysis, i.e.,
.../longitudinal_analysis/tp1_subj001 -> .../data/children/tp1/subj001 .../longitudinal_analysis/tp2_subj001 -> .../data/children/tp2/subj001 ...
Additionally, are there any considerations regarding renaming/moving subject folders (as long as anything inside a subject folder remains unchanged), i.e. are there any hard links or similar that would break further processing?
Thanks a lot!
Janosch
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Martin Reuter, Ph.D. Assistant in Neuroscience - Massachusetts General Hospital Instructor in Neurology - Harvard Medical School MGH / HMS / MIT
A.A.Martinos Center for Biomedical Imaging 149 Thirteenth Street, Suite 2301 Charlestown, MA 02129
Phone: +1-617-724-5652 Email:
mreuter@nmr.mgh.harvard.edu
reuter@mit.edu
Web : http://reuter.mit.edu