On 4/13/06, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote:

how did you manually create the brain.mgz files? These must be intensity
normalized (with mri_normalize) first, otherwise nothing will work. The
white matter should be close to or equal to 110 everywhere in the
brain.mgz volume.
On Thu, 13 Apr 2006, Jenni Pacheco wrote:

> Hi Tom,
>
> What kind of scans are you working with?  How were they collected?  Have you
> averaged more than one acquisition?
>
> Jenni
>
>  _____
>
> From: Tom Bayer [mailto:tabayer@gmail.com]
> Sent: Thursday, April 13, 2006 7:41 PM
> To: Jenni Pacheco
> Subject: Re: [Freesurfer] aseg.auto.mgz
>
>
> autorecon2 ran all the way through. I first noticed a problem because the
> wm.mgz file is only a narrow band of voxels and its distorted.  The
> brain.mgz file is missing some areas of white matter.   aseg.auto.mgz is
> missing a hemisphere, and aseg.mgz is very similar in shape to wm.mgz, which
> is an irregularly shaped blob that occupies about 1/4 of the coronal
> section, and is roughly centered in the image.
>
> I am fairly certain that the failure occured in the step generating
> aseg.auto.mgz or one of the steps just prior, because nu_noneck.mgz looks
> good.
>
> I created brainmask.mgz and brainmask.auto.mgz using manual thresholding and
> they are exactly to my specifications.  Strangely, the output file brain.mgz
> still has problems.  It looks exactly as it did before manual thresholding.
>
> on a previous post to the list, it was suggested that default thresholding
> took over because brainmask.mgz and brainmask.auto.mgz do not match.  I
> assure you that they do match.  I cop[ied the mask that I manually created
> so that both files match exactly.
>
> -tom
>
>
> On 4/13/06, Jenni Pacheco <jpacheco@nmr.mgh.harvard.edu> wrote:
>
> Hi Tom,
>
> We're going to need a little more info.  Did autorecon2 finish?  Did it
> fail part way through?  When you say everything has problems, what kind of
> problems?
>
> go here to see the types of information we need in order to best help you:
> https://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
>
> thanks,
> --
> Jenni Pacheco, M.A.
> Massachusetts General Hospital
> Athinoula A. Martinos Center for Biomedical Imaging
> Bldg. 149, 13th Street, CNY 2301
> Charlestown, MA 02129
> jpacheco@nmr.mgh.harvard.edu
>
> On Thu, 13 Apr 2006, Tom Bayer wrote:
>
>> I'm having problems with the output of recon-all -autorecon2.  proceeding
>> down the recon-all dev table, I checked the outputs of each step.
>> everything looks good until aseg.auto.mgz  everything after has problems.
>> What kind of manual intervention do I need to do for this?
>>
>> -Tom
>>
>
>
>
>