Hi Bruce,

The T1 images are oriented correctly, but the values are messed up somehow--I'm not sure why. I've attached a screenshot with the problem subject on the left and a normal subject on the right. I've also attached the talairach_with_skull.lta file, which looks normal. Is it possible that recon-all is messing up prior to the watershed?

Thanks!

On Fri, Mar 15, 2013 at 12:17 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote:
it seems like the watershed failed and nothing got created after that. You need to track down why. Usually it's either the talaraich that failed or the volume is oriented incorrectly. Can you bring the T1.mgz up in tkmedit and confirm that the directions it shows are anatomically correct? Then look at the transforms/talairach_with_skull.lta


On Fri, 15 Mar 2013, Hayley Dorfman wrote:

Hi Bruce,

Thanks for your quick reply! I looked in the mri subfolder for this
participant, and many necessary files do not exist for them. These are the
only files that were created in their mri subfolder:

mri_nu_correct.mni.log
orig
rawavg.mgz
talairach_with_skull.log
nu.mgz
orig.mgz
T1.mgz
transforms

The talairach transform looks fine (see attached talairach.xfm file), but as
I mentioned above, there are no brain.mgz or brainmask.mgz files to check.
Additionally, the orig and rawaverage files look like they have proper
contrast, but the nu.mgz has a strange "halo" above the brain and the T1.mgz
is completely blown out (screenshots attached). The brain is also not in the
center of the field of view--could this be affecting skull stripping or the
nu file?

Thanks for your help!


On Thu, Mar 14, 2013 at 6:27 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu>
wrote:
      Hi Hayley

      the LINE SEARCH messages can be ignored, but not the watershed
      one. The watershed error means the skull stripping has failed,
      possibly because of a talairach normalization problem. Have you
      checked the talairach.xfm and the brain.mgz? That's where I
      would start

      cheers
      Bruce


      On Thu, 14 Mar 2013, Hayley Dorfman wrote:

            Hello,

            I'm having trouble with recon-all. For 1 of 49
            subjects, it completes with
            the following error:

            mri_watershed Error: 
             GLOBAL region of the brain empty !

            Other errors are found in the log which read:

             IFLAG= -1  LINE SEARCH FAILED. SEE DOCUMENTATION OF
            ROUTINE MCSRCH ERROR
            RETURN OF LINE SEARCH: INFO= 6 POSSIBLE CAUSES:
            FUNCTION OR GRADIENT ARE
            INCORRECT OR INCORRECT TOLERANCES

            This subject's surf files are also empty.

            I found this thread from two years ago mentioning a
            similar issue, but a
            solution was not available:
            http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg18863.html

            Has there been any progress? I've attached my log
            file.

            Any help is much appreciated!

            -Hayley






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.




--
Hayley




--
Hayley