I am having trouble with one of my skull strips, and getting very extreme results. When I re-run the skullstrip using different watershed values, anything at 56% or below takes 75%+ of the brain out, and as soon as I jump to 57% or higher, it leaves on almost all of the skull, except for maybe 5% of it. Is this indicative of a more complicated problem, and what is my next step (other than manually taking off the skull), if any? Thanks
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix info is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing your orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting very extreme results. When I re-run the skullstrip using different watershed values, anything at 56% or below takes 75%+ of the brain out, and as soon as I jump to 57% or higher, it leaves on almost all of the skull, except for maybe 5% of it. Is this indicative of a more complicated problem, and what is my next step (other than manually taking off the skull), if any? Thanks _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
I am currently in the process of manual editing...just curious though, what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix info is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing your orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting very extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out, and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking off
the
skull), if any? Thanks _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
it helps detect cases in which the deformation is too big (e.g. when the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious though, what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix info is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing your orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting very extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out, and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking off
the
skull), if any? Thanks _______________________________________________ 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
I just finished my 4 hour manual editing...went to file>save volume as...then saved it to where it defaulted me, which was the mri folder of this subject. When i pull up my brainmask, it is giving me the same one that it had before i made any changes. I hope i didnt do something wrong and lost all the edits. There is 256 COR files though now in my mri folder for this subject I noticed. Maybe these are the edited slices? Just let me know how to get my final brainmask and how to proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g. when the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix
info
is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing
your
orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting
very
extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out,
and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking
off
the
skull), if any? Thanks _______________________________________________ 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
sounds like you saved it in COR format, which is what will happen if you specify a destination folder instead of a file name (e.g. brainmask.mgz). Try loading that in tkmedit to make sure it's got your edits, then you can convert it to .mgz with mri_convert.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save volume as...then saved it to where it defaulted me, which was the mri folder of this subject. When i pull up my brainmask, it is giving me the same one that it had before i made any changes. I hope i didnt do something wrong and lost all the edits. There is 256 COR files though now in my mri folder for this subject I noticed. Maybe these are the edited slices? Just let me know how to get my final brainmask and how to proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g. when the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix
info
is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing
your
orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting
very
extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out,
and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking
off
the
skull), if any? Thanks _______________________________________________ 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
The brainmask.mgz file is the one that should be edited (and saved), so select that filename to save. Then just continue with autorecon2 (and it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag to recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save volume as...then saved it to where it defaulted me, which was the mri folder of this subject. When i pull up my brainmask, it is giving me the same one that it had before i made any changes. I hope i didnt do something wrong and lost all the edits. There is 256 COR files though now in my mri folder for this subject I noticed. Maybe these are the edited slices? Just let me know how to get my final brainmask and how to proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g. when the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix
info
is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing
your
orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting
very
extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out,
and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking
off
the
skull), if any? Thanks _______________________________________________ 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
but you'll need to convert the COR file you saved to brainmask.mgz if you want to recover your edits (4 hours of manual editing!).
Bruce
On Wed, 28 Jun 2006, Nick Schmansky wrote:
The brainmask.mgz file is the one that should be edited (and saved), so select that filename to save. Then just continue with autorecon2 (and it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag to recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save volume as...then saved it to where it defaulted me, which was the mri folder of this subject. When i pull up my brainmask, it is giving me the same one that it had before i made any changes. I hope i didnt do something wrong and lost all the edits. There is 256 COR files though now in my mri folder for this subject I noticed. Maybe these are the edited slices? Just let me know how to get my final brainmask and how to proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g. when the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the fix
info
is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing
your
orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote:
I am having trouble with one of my skull strips, and getting
very
extreme results. When I re-run the skullstrip using different
watershed
values, anything at 56% or below takes 75%+ of the brain out,
and
as
soon as I jump to 57% or higher, it leaves on almost all of the
skull,
except for maybe 5% of it. Is this indicative of a more
complicated
problem, and what is my next step (other than manually taking
off
the
skull), if any? Thanks _______________________________________________ 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
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Alright I converted the COR files to mgz fine, and saved the volume as brainmask.mgz. All looks good. But before I run -autorecon2, am I going to run into problems because I only have a brainmask.mgz and not a brainmask.auto.mgz? (I deleted the old brainmask.auto.mgz because it was the old volumes before I corrected them manually). I just want to make sure I am good to run -autorecon2 before I go ahead and begin it.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
but you'll need to convert the COR file you saved to brainmask.mgz if you want to recover your edits (4 hours of manual editing!).
Bruce
On Wed, 28 Jun 2006, Nick Schmansky wrote:
The brainmask.mgz file is the one that should be edited (and
saved), so
select that filename to save. Then just continue with autorecon2
(and
it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag
to
recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save
volume
as...then saved it to where it defaulted me, which was the mri
folder of
this subject. When i pull up my brainmask, it is giving me the
same one
that it had before i made any changes. I hope i didnt do
something
wrong and lost all the edits. There is 256 COR files though now
in my
mri folder for this subject I noticed. Maybe these are the
edited
slices? Just let me know how to get my final brainmask and how
to
proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g.
when
the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
First have a look at the troubleshooting wiki pages at:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
In particular, Subject 1 has a skull strip problem, and the
fix
info
is here:
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
I'm guessing you have seen this page since you have already attempted one type of fix (adjusting the watershed parameters). The alternative fix is the slice-by-slice manual editing.
You could also try adding the wsatlas flag:
recon-all -s (subject) -skullstrip -wsatlas
which will use an atlas to help with the skull-strip.
Does the contrast of your image look good? That is, comparing
your
orig.mgz to that of the sample subject 'bert' included with freesurfer?
Nick
On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote: > I am having trouble with one of my skull strips, and getting
very
> extreme results. When I re-run the skullstrip using
different
watershed > values, anything at 56% or below takes 75%+ of the brain out,
and
as > soon as I jump to 57% or higher, it leaves on almost all of
the
skull, > except for maybe 5% of it. Is this indicative of a more complicated > problem, and what is my next step (other than manually taking
off
the > skull), if any? > Thanks > _______________________________________________ > 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
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
I'm not sure what will happen. Safest to just recreate it:
cd subject/mri mri_watershed T1.mgz brainmask.auto.mgz
dasoscia@buffalo.edu wrote:
Alright I converted the COR files to mgz fine, and saved the volume as brainmask.mgz. All looks good. But before I run -autorecon2, am I going to run into problems because I only have a brainmask.mgz and not a brainmask.auto.mgz? (I deleted the old brainmask.auto.mgz because it was the old volumes before I corrected them manually). I just want to make sure I am good to run -autorecon2 before I go ahead and begin it.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
but you'll need to convert the COR file you saved to brainmask.mgz if you want to recover your edits (4 hours of manual editing!).
Bruce
On Wed, 28 Jun 2006, Nick Schmansky wrote:
The brainmask.mgz file is the one that should be edited (and
saved), so
select that filename to save. Then just continue with autorecon2
(and
it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag
to
recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save
volume
as...then saved it to where it defaulted me, which was the mri
folder of
this subject. When i pull up my brainmask, it is giving me the
same one
that it had before i made any changes. I hope i didnt do
something
wrong and lost all the edits. There is 256 COR files though now
in my
mri folder for this subject I noticed. Maybe these are the
edited
slices? Just let me know how to get my final brainmask and how
to
proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g.
when
the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
I am currently in the process of manual editing...just curious
though,
what exatly does adding an atlas do to the skullstrip process?
Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu:
>First have a look at the troubleshooting wiki pages at: > > > >
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
>In particular, Subject 1 has a skull strip problem, and the > >
fix
info
>is >here: > > > >
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
>I'm guessing you have seen this page since you have already >attempted >one type of fix (adjusting the watershed parameters). The >alternative >fix is the slice-by-slice manual editing. > >You could also try adding the wsatlas flag: > > recon-all -s (subject) -skullstrip -wsatlas > >which will use an atlas to help with the skull-strip. > >Does the contrast of your image look good? That is, comparing > >
your
>orig.mgz to that of the sample subject 'bert' included with >freesurfer? > >Nick > > >On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote: > > >>I am having trouble with one of my skull strips, and getting >> >>
very
>>extreme results. When I re-run the skullstrip using >> >>
different
>watershed > > >>values, anything at 56% or below takes 75%+ of the brain out, >> >>
and
>as > > >>soon as I jump to 57% or higher, it leaves on almost all of >> >>
the
>skull, > > >>except for maybe 5% of it. Is this indicative of a more >> >> >complicated > > >>problem, and what is my next step (other than manually taking >> >>
off
>the > > >>skull), if any? >>Thanks >>_______________________________________________ >>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
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
or in your case dave, just:
cp brainmask.mgz brainmask.auto.mgz
since you manually edited brainmask.mgz
On Thu, 2006-06-29 at 14:18 -0400, Doug Greve wrote:
I'm not sure what will happen. Safest to just recreate it:
cd subject/mri mri_watershed T1.mgz brainmask.auto.mgz
dasoscia@buffalo.edu wrote:
Alright I converted the COR files to mgz fine, and saved the volume as brainmask.mgz. All looks good. But before I run -autorecon2, am I going to run into problems because I only have a brainmask.mgz and not a brainmask.auto.mgz? (I deleted the old brainmask.auto.mgz because it was the old volumes before I corrected them manually). I just want to make sure I am good to run -autorecon2 before I go ahead and begin it.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
but you'll need to convert the COR file you saved to brainmask.mgz if you want to recover your edits (4 hours of manual editing!).
Bruce
On Wed, 28 Jun 2006, Nick Schmansky wrote:
The brainmask.mgz file is the one that should be edited (and
saved), so
select that filename to save. Then just continue with autorecon2
(and
it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag
to
recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save
volume
as...then saved it to where it defaulted me, which was the mri
folder of
this subject. When i pull up my brainmask, it is giving me the
same one
that it had before i made any changes. I hope i didnt do
something
wrong and lost all the edits. There is 256 COR files though now
in my
mri folder for this subject I noticed. Maybe these are the
edited
slices? Just let me know how to get my final brainmask and how
to
proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
it helps detect cases in which the deformation is too big (e.g.
when
the surface deforms all the way into the cerebellum), and can thus recover from e.g. cerebellum chopping.
On Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote:
> I am currently in the process of manual editing...just curious > though,
> what exatly does adding an atlas do to the skullstrip process? > > Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu: > > > > First have a look at the troubleshooting wiki pages at: > > > > > >
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
> > In particular, Subject 1 has a skull strip problem, and the > >
fix
info
> > is > > here: > > > > > >
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
> > I'm guessing you have seen this page since you have already > > attempted > > one type of fix (adjusting the watershed parameters). The > > alternative > > fix is the slice-by-slice manual editing. > > > > You could also try adding the wsatlas flag: > > > > recon-all -s (subject) -skullstrip -wsatlas > > > > which will use an atlas to help with the skull-strip. > > > > Does the contrast of your image look good? That is, comparing > > your
> > orig.mgz to that of the sample subject 'bert' included with > > freesurfer? > > > > Nick > > > > > > On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote: > > > > > I am having trouble with one of my skull strips, and getting > > > very
> > > extreme results. When I re-run the skullstrip using > > >
different
> > watershed > > > > > values, anything at 56% or below takes 75%+ of the brain out, > > > and
> > as > > > > > soon as I jump to 57% or higher, it leaves on almost all of > > >
the
> > skull, > > > > > except for maybe 5% of it. Is this indicative of a more > > > > > complicated > > > > > problem, and what is my next step (other than manually taking > > > off
> > the > > > > > skull), if any? > > > Thanks > > > _______________________________________________ > > > 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
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
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
Oh, don't do that! recon-all will compare the auto with the non-auto. If they are the same then it will re-compute the auto and simply copy to the non-auto, and your changes will be lost. If they are different, then it will recompute the auto but will not propagate it to the non-auto. Make sense?
doug
Nick Schmansky wrote:
or in your case dave, just:
cp brainmask.mgz brainmask.auto.mgz
since you manually edited brainmask.mgz
On Thu, 2006-06-29 at 14:18 -0400, Doug Greve wrote:
I'm not sure what will happen. Safest to just recreate it:
cd subject/mri mri_watershed T1.mgz brainmask.auto.mgz
dasoscia@buffalo.edu wrote:
Alright I converted the COR files to mgz fine, and saved the volume as brainmask.mgz. All looks good. But before I run -autorecon2, am I going to run into problems because I only have a brainmask.mgz and not a brainmask.auto.mgz? (I deleted the old brainmask.auto.mgz because it was the old volumes before I corrected them manually). I just want to make sure I am good to run -autorecon2 before I go ahead and begin it.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
but you'll need to convert the COR file you saved to brainmask.mgz if you want to recover your edits (4 hours of manual editing!).
Bruce
On Wed, 28 Jun 2006, Nick Schmansky wrote:
The brainmask.mgz file is the one that should be edited (and
saved), so
select that filename to save. Then just continue with autorecon2
(and
it will use your brainmask.mgz). It will not overwrite your brainmask.mgz unless you specify the aptly named '-clean-bm' flag
to
recon-all.
On Wed, 2006-06-28 at 15:00 -0400, dasoscia@buffalo.edu wrote:
I just finished my 4 hour manual editing...went to file>save
volume
as...then saved it to where it defaulted me, which was the mri
folder of
this subject. When i pull up my brainmask, it is giving me the
same one
that it had before i made any changes. I hope i didnt do
something
wrong and lost all the edits. There is 256 COR files though now
in my
mri folder for this subject I noticed. Maybe these are the
edited
slices? Just let me know how to get my final brainmask and how
to
proceed with running -autorecon2 on it properly. Thanks.
Quoting Bruce Fischl fischl@nmr.mgh.harvard.edu:
>it helps detect cases in which the deformation is too big (e.g. > > >
when
>the >surface deforms all the way into the cerebellum), and can thus >recover from >e.g. cerebellum chopping. > >On >Wed, 28 Jun 2006 dasoscia@buffalo.edu wrote: > > > > >>I am currently in the process of manual editing...just curious >> >> >> >though, > > > >>what exatly does adding an atlas do to the skullstrip process? >> >>Quoting Nick Schmansky nicks@nmr.mgh.harvard.edu: >> >> >> >> >>>First have a look at the troubleshooting wiki pages at: >>> >>> >>> >>> >>>
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TroubleshootingData
>>>In particular, Subject 1 has a skull strip problem, and the >>> >>> >>>
fix
>info > > > >>>is >>>here: >>> >>> >>> >>> >>>
https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/SkullStripFix
>>>I'm guessing you have seen this page since you have already >>>attempted >>>one type of fix (adjusting the watershed parameters). The >>>alternative >>>fix is the slice-by-slice manual editing. >>> >>>You could also try adding the wsatlas flag: >>> >>> recon-all -s (subject) -skullstrip -wsatlas >>> >>>which will use an atlas to help with the skull-strip. >>> >>>Does the contrast of your image look good? That is, comparing >>> >>> >>> >your > > > >>>orig.mgz to that of the sample subject 'bert' included with >>>freesurfer? >>> >>>Nick >>> >>> >>>On Wed, 2006-06-28 at 10:09 -0400, dasoscia@buffalo.edu wrote: >>> >>> >>> >>>>I am having trouble with one of my skull strips, and getting >>>> >>>> >>>> >very > > > >>>>extreme results. When I re-run the skullstrip using >>>> >>>> >>>>
different
>>>watershed >>> >>> >>> >>>>values, anything at 56% or below takes 75%+ of the brain out, >>>> >>>> >>>> >and > > > >>>as >>> >>> >>> >>>>soon as I jump to 57% or higher, it leaves on almost all of >>>> >>>> >>>>
the
>>>skull, >>> >>> >>> >>>>except for maybe 5% of it. Is this indicative of a more >>>> >>>> >>>> >>>complicated >>> >>> >>> >>>>problem, and what is my next step (other than manually taking >>>> >>>> >>>> >off > > > >>>the >>> >>> >>> >>>>skull), if any? >>>>Thanks >>>>_______________________________________________ >>>>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
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
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
dasoscia＠buffalo.edu -
Doug Greve -
Nick Schmansky