Hi Doug,
Ah, I found the cause myself, which might be useful for others. It happens already in the first step, before the ones I listed: in mri_convert. I do:
mri_convert struct.nii 001.mgz
And it flips. This I can see because I converted the mgz back (mri_convert 001.mgz orig.nii) and then it's indeed reversed (and shifted, rotated, all rigid body transformations one can think of). Why does this happen? Is it normal for nii-files? I see that I can specify the orientation in mri_convert by doing mri_convert -in_orientation and then a three letter specification. I chose PSR and now the changes are far less severe (only a small shift still) and most importantly: no flip anymore. Do I now need to reconvert my all my surfaces? I don't hope, but I'm at least glad I have this one fixed.
best,
Frank
----- "Douglas N Greve" greve@nmr.mgh.harvard.edu schreef:
Van: "Douglas N Greve" greve@nmr.mgh.harvard.edu Aan: "Frank Leoné" f.leone@donders.ru.nl Cc: freesurfer@nmr.mgh.harvard.edu Verzonden: Woensdag 18 januari 2012 18:34:31 Onderwerp: Re: [Freesurfer] Freesurfer flips brains?
Hi Frank, we'll need a lot more detail. Are you doing all of this in FreeSurfer? What are your command lines? How are you evaluating whether volumes are flipped? doug
Léoné wrote:
Dear all,
I must be missing something really obvious, but it seems my brains
are flipped. What I do:
- I have a highres structural scan, nii file format
- I convert it to a surface using recon-all
- Coregister the highres structural to the functional scan, so it is
in the voxel size and orientation as the functional scans
- I try to bbregister the coregistered and resliced T1 with the
surface
But then: hey, it doesn't fit, my low-res T1 is flipped compared to
my y surface, hence to my original T1. So is it indeed true that my T1 might have been flipped? If so, does it make a wrong assumption on neurological vs radiological convention, or what is going on?
Thanks in advance,
best, Frank
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
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
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.
Hi Frank,
I think we still need to know. What software are you using to view the volumes?
D
On Thu, Jan 19, 2012 at 10:59 AM, Léoné, F.T.M. (Frank) f.leone@fcdonders.ru.nl wrote:
Hi Doug,
Ah, I found the cause myself, which might be useful for others. It happens already in the first step, before the ones I listed: in mri_convert. I do:
mri_convert struct.nii 001.mgz
And it flips. This I can see because I converted the mgz back (mri_convert 001.mgz orig.nii) and then it's indeed reversed (and shifted, rotated, all rigid body transformations one can think of). Why does this happen? Is it normal for nii-files? I see that I can specify the orientation in mri_convert by doing mri_convert -in_orientation and then a three letter specification. I chose PSR and now the changes are far less severe (only a small shift still) and most importantly: no flip anymore. Do I now need to reconvert my all my surfaces? I don't hope, but I'm at least glad I have this one fixed.
best,
Frank
----- "Douglas N Greve" greve@nmr.mgh.harvard.edu schreef:
Van: "Douglas N Greve" greve@nmr.mgh.harvard.edu Aan: "Frank Leoné" f.leone@donders.ru.nl Cc: freesurfer@nmr.mgh.harvard.edu Verzonden: Woensdag 18 januari 2012 18:34:31 Onderwerp: Re: [Freesurfer] Freesurfer flips brains?
Hi Frank, we'll need a lot more detail. Are you doing all of this in FreeSurfer? What are your command lines? How are you evaluating whether volumes are flipped? doug
Léoné wrote:
Dear all,
I must be missing something really obvious, but it seems my brains
are flipped. What I do:
- I have a highres structural scan, nii file format
- I convert it to a surface using recon-all
- Coregister the highres structural to the functional scan, so it is
in the voxel size and orientation as the functional scans
- I try to bbregister the coregistered and resliced T1 with the
surface
But then: hey, it doesn't fit, my low-res T1 is flipped compared to
my y surface, hence to my original T1. So is it indeed true that my T1 might have been flipped? If so, does it make a wrong assumption on neurological vs radiological convention, or what is going on?
Thanks in advance,
best,
Frank _______________________________________________ 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
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Dan,
Sure. It was in my earlier e-mail of today: I use tksurfer, see below. For this last test, I used the SPM display function though, that now shows them to both have the same orientation, so appears to be okay. Or am I still missing something?
best,
Frank
Hi Douglas,
Ah, makes sense, sorry for that.
I do most of the work in SPM, so:
- I have a highres structural scan, nii file format - I convert it to a surface using recon-all - Coregister the highres structural to the functional scan, so it is in the voxel size and orientation as the functional scans (SPM) - I try to bbregister the coregistered and resliced T1 with the surface using something like: bbregister --s S7_FS --mov ~/Data/Exp1_Saccades_phase/S7_FS/mri/orig/flipped_rmasked_str_S2_seler20090508.nii --init-spm --reg ~/Data/Exp1_Saccades_phase/S7_FS/mri/orig/register.partial.auto.dat --t1 and view it with vglrun tkregister2 --mov /home/action/fraleo/Data/Exp1_Saccades_phase/S7_FS/mri/orig/flipped_rmasked_str_S2_seler20090508.nii --reg /home/action/fraleo/Data/Exp1_Saccades_phase/S7_FS/mri/orig/register.partial.auto.dat --surf
Here you see that I already use a flipped structural scan. If I do it like this, bbregister converges to a nice and correct solution (everything aligns). If I don't, tkregister2 showms unregistered struc and func, while it is obvious it would be roughly aligned if you would flip either of them.
In addition/in general: What would be the recommended way to do the coregistration etc. if you do all you analyses in SPM? Would it for example be better to coregistrate and reslice all functionals with the structural, so they are automatically aligned with the surface? Now I have a structural which is coregistered with the funcional and I realign that one, but that makes for one extra step where it can go wrong. Don't think this is related to the flip, but I also have a few alignment problems for other subjects.
Thanks a lot in advance,
best,
Frank ----- "Dan Wakeman" dgwakeman@gmail.com schreef:
Van: "Dan Wakeman" dgwakeman@gmail.com Aan: "F.T.M. Léoné (Frank)" f.leone@donders.ru.nl Cc: freesurfer@nmr.mgh.harvard.edu Verzonden: Donderdag 19 januari 2012 17:09:28 Onderwerp: Re: [Freesurfer] Freesurfer flips brains?
Hi Frank,
I think we still need to know. What software are you using to view the volumes?
D
On Thu, Jan 19, 2012 at 10:59 AM, Léoné, F.T.M. (Frank) f.leone@fcdonders.ru.nl wrote:
Hi Doug,
Ah, I found the cause myself, which might be useful for others. It
happens already in the first step, before the ones I listed: in mri_convert. I do:
mri_convert struct.nii 001.mgz
And it flips. This I can see because I converted the mgz back
(mri_convert 001.mgz orig.nii) and then it's indeed reversed (and shifted, rotated, all rigid body transformations one can think of). Why does this happen? Is it normal for nii-files? I see that I can specify the orientation in mri_convert by doing mri_convert -in_orientation and then a three letter specification. I chose PSR and now the changes are far less severe (only a small shift still) and most importantly: no flip anymore. Do I now need to reconvert my all my surfaces? I don't hope, but I'm at least glad I have this one fixed.
best,
Frank
----- "Douglas N Greve" greve@nmr.mgh.harvard.edu schreef:
Van: "Douglas N Greve" greve@nmr.mgh.harvard.edu Aan: "Frank Leoné" f.leone@donders.ru.nl Cc: freesurfer@nmr.mgh.harvard.edu Verzonden: Woensdag 18 januari 2012 18:34:31 Onderwerp: Re: [Freesurfer] Freesurfer flips brains?
Hi Frank, we'll need a lot more detail. Are you doing all of this
in
FreeSurfer? What are your command lines? How are you evaluating whether volumes are flipped? doug
Léoné wrote:
Dear all,
I must be missing something really obvious, but it seems my
brains
are flipped. What I do:
- I have a highres structural scan, nii file format
- I convert it to a surface using recon-all
- Coregister the highres structural to the functional scan, so it
is
in the voxel size and orientation as the functional scans
- I try to bbregister the coregistered and resliced T1 with the
surface
But then: hey, it doesn't fit, my low-res T1 is flipped compared
to
my y surface, hence to my original T1. So is it indeed true that my
T1
might have been flipped? If so, does it make a wrong assumption on neurological vs radiological convention, or what is going on?
Thanks in advance,
best,
Frank _______________________________________________ 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
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
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.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Dan Wakeman -
Léoné, F.T.M. (Frank)