Hi Doug, The second command below should copy your edits from the finalsurf manedit in cross to the long directory. You can try that command manually with a fresh volume. If it really does not carry over, can you send me the two volumes (edited one from cross and the unedited one from long) so that I can take a look. Maybe your edits are wrong (deletes are not zero, but adifferent number, I think 1).

Not sure what consequences it has to copy to brainmask but I would not recommend it.

Martin

Sent via my smartphone, please excuse brevity.

-------- Original message -------- From: Douglas Merkitch Douglas_Merkitch@rush.edu Date:06/05/2014 9:51 PM (GMT+01:00) To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] brain.finalsurfs.manedit.mgz file edits & longitudinal processing

Hello Martin and FreeSurfer experts,

I've checked the recon-all.log from the 2nd tp of one of the longitudinal subject directories (1004_1.long.1004_base) and found the following 4 lines using your search criteria:

1. \n mri_mask -transfer 255 -keep_mask_deletion_edits brainmask.auto.mgz brainmask_1004_base.mgz brainmask.auto.mgz \n 2. \n mri_mask -transfer 255 -keep_mask_deletion_edits -xform /Users/glennstebbins/Desktop/harlow_mount/JenG/ResearchAssistant/data/freesurfer_v5.1/1004_base/mri/transforms/1004_1_to_1004_base.lta brain.finalsurfs.mgz /Users/glennstebbins/Desktop/harlow_mount/JenG/ResearchAssistant/data/freesurfer_v5.1/1004_1/mri/brain.finalsurfs.manedit.mgz brain.finalsurfs.mgz \n 3. \n mri_mask -transfer 255 -keep_mask_deletion_edits -xform /Users/glennstebbins/Desktop/harlow_mount/JenG/ResearchAssistant/data/freesurfer_v5.1/1004_base/mri/transforms/1004_1_to_1004_base.lta wm.mgz /Users/glennstebbins/Desktop/harlow_mount/JenG/ResearchAssistant/data/freesurfer_v5.1/1004_1/mri/wm.mgz wm.mgz \n 4. \n mri_mask -transfer 255 -keep_mask_deletion_edits brain.finalsurfs.mgz brain.finalsurfs.manedit.mgz brain.finalsurfs.mgz \n

Does any of this stand out to you?

In visually comparing the brain.finalsurfs.manedit.mgz files between the cross and long, the edits do not appear to transfer to the long as has happened previously. However, when the brain.finalsurfs.manedit.mgz file is copied to the brianmask.mgz file (in the base and cross time points), and the longitudinal processes stream is run, the edits now contained in the brainmask.mgz files do seem to transfer.

Can you think of any complications (other than cerebellar) that would arise from copying the brain.finalsurfs.manedit.mgz file (where edits were made) to the brainmask.mgz in the cross-sectional and base directories and rerunning the longitudinal processing stream from scratch?   Thanks for all your help!

Doug

Doug Merkitch Neurological Sciences Rush University Medical Center Phone: (312) 563-3853 Fax: (312) 563-4660 Email: Douglas_Merkitch@rush.edu

On May 19, 2014, at 3:52 PM, Martin Reuter mreuter@nmr.mgh.harvard.edu wrote:

Hi Doug,

Usually I edit dura etc in the brainmask and only touch finalsurfs for cerebellum edits. But your way should also work (at least I don't know right now why it should not).

Anyway, I looked at recon-all and in the mask BFS step, it does copy the edits over from the cross into the long (if it's working as it should, see line 2565). That command should be printed in your log file of the long run and you can search and double-check that it is there. (mri_mask -transfer 255 -keep_mask_deleting_edits -xform .....)

You can also compare the brain_finalsurfs between cross and long in each time point to make sure the edits did get transferred. Be aware that these two images are not registered. The cross is in the original space, the long in the same space as the base and all other time points. So you may have to search around to find the same exact place.

Anyway, assuming edits do get through, the finalsurfs is then used in the mris_make_surfaces step. In long, this step is different from cross, as it takes the surfaces from the base as initialization and then fine tunes them. So you also have to check the surfaces in the base are all OK. That is essential as long takes it from there.

Best, Martin

On 05/16/2014 12:29 PM, Douglas Merkitch wrote: Dear Martin and FreeSurfer experts,

Thanks for your response.

I have just finished re-running some subjects from scratch in 5.3 and it appears that the edits from brain.finalsurfs.manedit.mgz are not being applied to the longitudinal subject directories.

Just to clarify, I have been doing ALL of the pial surface edits in the brain.finalsurfs.manedit.mgz, not just the pial surface edits that have extended into the cerebellum. The rationale for doing this is mainly continuity, as the previous research assistant made all pial surface edits in this way because of a response she had received from the FS forum saying that there shouldn't be any problem with performing all the pial surface edits in the brain.finalsurfs.manedit.mgz file.

Is there something in the FreeSurfer longitudinal processing code that would treat edits near the cerebellum of the brain.finalsurfs.manedit.mgz file differently from those made in other parts of the brain (e.g. lateralorbitofrontal, dura, etc.,)?

Can you think of anything else that would be causing this or anything that I could try to get the longitudinal processing stream to incorporate the edits from the brain.finalsurfs.manedit.mgz file?

Thanks for your help!

Doug Merkitch Neurological Sciences Rush University Medical Center Phone: (312) 563-3853 Fax: (312) 563-4660 Email: Douglas_Merkitch@rush.edu

Begin forwarded message:

From: Martin Reuter mreuter@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] brain.finalsurfs.manedit.mgz file edits & longitudinal processing Date: May 2, 2014 4:20:17 PM CDT To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu Reply-To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu

Hi Doug,

so you 1. edited the cross sectionals and surfaces look fine (make sure you edit the brain.finalsurfs.manedit.mgz file, which you create as a copy of the brain.finalsurfs.mgz) 2. checked/edited the base and surfaces look fine too 3. ran the longitudinal from scratch (w/o any edits)

and then in the long, the surface looks not good?

Edits should automatically be copied from cross->long (if the manedit file exist). Surfaces are taken from the base to initialize longitudinal processing. That is why both step1 and step2 above are important. Step3 is a good idea to make sure that the longs are not polluted by some earlier edits. For example edits from cross are only copied over if the long has no manedit file, else only the edits in the long will be used. (I think this also answers some of the questions in the archive mail 1 you cite below, the archive mail 2 is a question about cross sectional processing, but I think there the problem is that brainmask.finalsurfs.mgz was edited directly, not the manedit file, so edits probably get overwritten).

I also have the feeling that we fixed something about the editing in longitudinal processing after 5.1, but cannot find it documented on the wiki (so not sure). You may want to try to run this with 5.3 if it does not work with 5.1

Best, Martin

On 05/02/2014 10:45 AM, Douglas Merkitch wrote: Dear Freesurfer experts,

I have a question regarding longitudinal edits. Specifically the brain.finalsurfs.manedit.mgz file and the longitudinal processing stream in Freesufer 5.1 on a Mac (freesurfer-i686-apple-darwin9.8.0-stable5-20110525).

The problem that I am having is that the edits saved in the brain.finalsurfs.manedit.mgz file do not carry over when processing longitudinally.

Previous posts have addressed this topic, but have not (to my knowledge) solved the issue. Please see the threads from the Freesurfer archive below for a brief context:

1. http://www.mail-archive.com/freesurfer%40nmr.mgh.harvard.edu/msg27695.html

2. http://www.mail-archive.com/freesurfer%40nmr.mgh.harvard.edu/msg33471.html

Please let me know if more information is needed to understand the issue.

Any insight/ideas regarding this issue would be greatly appreciated!

Thanks,

Doug

Doug Merkitch Neurological Sciences Rush University Medical Center Phone: (312) 563-3853 Fax: (312) 563-4660 Email: Douglas_Merkitch@rush.edu

_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer