Dear Bruce and FS experts
I'm sending attached 2 images, 1) one with the original subject, without any transformation (just from NIfTY format to mgz format), and 2) the T1.mgz created during the recon-all step, note that the original image has the background with a higher intensity, could be the background the problem to the recon-all? is there any way to put the background to zero?
Best
Gabriel
Date: Tue, 22 Dec 2009 10:51:16 -0500 From: fischl@nmr.mgh.harvard.edu To: gabriellbk@hotmail.com Subject: RE: [Freesurfer] recon-all problem
Hi Gabriel,
can you post this type of question to the list so others can answer? Send along a snapshot of the T1.mgz also so we can take a look Bruce
On Tue, 22 Dec 2009, Gabriel Go.Es. wrote:
Thanks Bruce
I have some more doubts, when you say the "talairach" what are you exacly refering to?, I've checked all the txt's named talairach, and where they say intensity is equal to 1; about the T1 mgz, it apperas almost all in withe, I guess that could be the main problem, looks like the intensity its bad, I've also checked the nu.mgz, orig.mgz and rawavg.mgz, and all of them look so withe, but the rawavg.mgz which not so withe as the others but withe too. I guess the intensity from my original image is not good, how can i fix it to make the recon-all?
Regards
Gabriel
Date: Tue, 22 Dec 2009 09:25:46 -0500 From: fischl@nmr.mgh.harvard.edu To: gabriellbk@hotmail.com CC: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] recon-all problem
check the talairach and the intensity normalization (T1.mgz) to make sure they both succeeded. One or the other may need to be corrected.
cheers, Bruce On Tue, 22 Dec 2009, Gabriel Go.Es. wrote:
Dear Fs experts
I'm trying to run a recon-all on one subject but FS gives me the following error message:
#@# Skull Stripping Mon Dec 21 17:25:55 CET 2009 Linux localhost.localdomain 2.6.9-34.ELsmp #1 SMP Thu Mar 9 06:23:23 GMT 2006 x86_64 x86_64 x86_64 GNU/Linux
recon-all exited with ERRORS at Mon Dec 21 18:11:04 CET 2009
I take a look in the log trying to fing a solution and say:
****************TEMPLATE DEFORMATION****************
second estimation of the COG coord: x=127,y=128, z=127, r=9985 iterations GLOBAL CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 20574 RIGHT_CER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 99 LEFT_CER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 126 RIGHT_BRAIN CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 936 LEFT_BRAIN CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 576 OTHER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 18837 mri_watershed Error: GLOBAL region of the brain empty ! Linux localhost.localdomain 2.6.9-34.ELsmp #1 SMP Thu Mar 9 06:23:23 GMT 2006 x86_64 x86_64 x86_64 GNU/Linux
recon-all exited with ERRORS at Mon Dec 21 18:11:04 CET 2009
what can I do to keep with the recon-all, could be something wrong in the image?
Best
Gabriel
Sólo hay un loro experto en Windows 7 en todo el mundo. Y vive en Sietes ¡Cónocelo! http://www.sietesunpueblodeexpertos.com/
Date una vuelta por Sietes y conoce el pueblo de los expertos en Windows 7 http://www.sietesunpueblodeexpertos.com/
_________________________________________________________________ Sólo hay un loro experto en Windows 7 en todo el mundo. Y vive en Sietes ¡Cónocelo! http://www.sietesunpueblodeexpertos.com/
Hi Gabriel,
what was done to these datasets before the "original" image? They look like they are already skull-stripped, and are they really embedded in some type of bright background?
Bruce On Tue, 22 Dec 2009, Gabriel Go.Es. wrote:
Dear Bruce and FS experts
I'm sending attached 2 images, 1) one with the original subject, without any transformation (just from NIfTY format to mgz format), and 2) the T1.mgz created during the recon-all step, note that the original image has the background with a higher intensity, could be the background the problem to the recon-all? is there any way to put the background to zero?
Best
Gabriel
Date: Tue, 22 Dec 2009 10:51:16 -0500 From: fischl@nmr.mgh.harvard.edu To: gabriellbk@hotmail.com Subject: RE: [Freesurfer] recon-all problem
Hi Gabriel,
can you post this type of question to the list so others can answer? Send along a snapshot of the T1.mgz also so we can take a look Bruce
On Tue, 22 Dec 2009, Gabriel Go.Es. wrote:
Thanks Bruce
I have some more doubts, when you say the "talairach" what are you exacly refering to?, I've checked all the txt's named talairach, and where they say intensity is equal to 1; about the T1 mgz, it apperas almost all in withe, I guess that could be the main problem, looks like the intensity its bad, I've also checked the nu.mgz, orig.mgz and rawavg.mgz, and all of them look so withe, but the rawavg.mgz which not so withe as the others but withe too. I guess the intensity from my original image is not good, how can i fix it to make the recon-all?
Regards
Gabriel
Date: Tue, 22 Dec 2009 09:25:46 -0500 From: fischl@nmr.mgh.harvard.edu To: gabriellbk@hotmail.com CC: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] recon-all problem
check the talairach and the intensity normalization (T1.mgz) to make sure they both succeeded. One or the other may need to be corrected.
cheers, Bruce On Tue, 22 Dec 2009, Gabriel Go.Es. wrote:
Dear Fs experts
I'm trying to run a recon-all on one subject but FS gives me the following error message:
#@# Skull Stripping Mon Dec 21 17:25:55 CET 2009 Linux localhost.localdomain 2.6.9-34.ELsmp #1 SMP Thu Mar 9 06:23:23 GMT 2006 x86_64 x86_64 x86_64 GNU/Linux
recon-all exited with ERRORS at Mon Dec 21 18:11:04 CET 2009
I take a look in the log trying to fing a solution and say:
****************TEMPLATE DEFORMATION****************
second estimation of the COG coord: x=127,y=128, z=127, r=9985 iterations GLOBAL CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 20574 RIGHT_CER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 99 LEFT_CER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 126 RIGHT_BRAIN CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 936 LEFT_BRAIN CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 576 OTHER CSF_MIN=1, CSF_intensity=2, CSF_MAX=3 , nb = 18837 mri_watershed Error: GLOBAL region of the brain empty ! Linux localhost.localdomain 2.6.9-34.ELsmp #1 SMP Thu Mar 9 06:23:23 GMT 2006 x86_64 x86_64 x86_64 GNU/Linux
recon-all exited with ERRORS at Mon Dec 21 18:11:04 CET 2009
what can I do to keep with the recon-all, could be something wrong in the image?
Best
Gabriel
Sólo hay un loro experto en Windows 7 en todo el mundo. Y vive en Sietes ¡Cónocelo! http://www.sietesunpueblodeexpertos.com/
Date una vuelta por Sietes y conoce el pueblo de los expertos en Windows 7 http://www.sietesunpueblodeexpertos.com/
Sólo hay un loro experto en Windows 7 en todo el mundo. Y vive en Sietes ¡Cónocelo! http://www.sietesunpueblodeexpertos.com/
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Gabriel Go.Es.