Dear Freesurfer developers,
I have a few analyse images which are oriented incorrectly that are part of a larger dataset. When viewed in tkmedit, what is supposed to be coronal is actually displayed as axial and the sagittal is flipped. In these cases the Talairach transform fails spectacularly.
To try and fix the problem I used mri_info to determine the orientation information of my good files (LSA) and of the ones that aren't working (LAS), then used mri_convert with the --in_orientation and out_orientation flags to change the dimensions in analyse format. The analyze images look fine in MRIcro and the orig.mgz looks fine in tkmedit, but yet again the Talairach fails in the same way. I've also tried this by reorienting the images using MRIcro but this has the same result.
When you view the analyze images using GWRaw MRI viewer http://www.wideman-one.com/gw/brain/orientation/orientterms.htm (supposed to read the image file without header information), they look the same, but I'm a bit unsure of this program.
I'm really at a bit of a loss to try and fix this problem now so any help or suggestions would be really appreciated!
Many thanks,
Dominic
Melbourne Neuropsychiatry Centre
Hi Dominic,
do you have access to any format previous to the analyze one? It doesn't preserve orientation info, and all bets are off once you go through it.
cheers Bruce On Mon, 2 Nov 2009, Dominic Brendon Dwyer wrote:
Dear Freesurfer developers,
I have a few analyse images which are oriented incorrectly that are part of a larger dataset. When viewed in tkmedit, what is supposed to be coronal is actually displayed as axial and the sagittal is flipped. In these cases the Talairach transform fails spectacularly.
To try and fix the problem I used mri_info to determine the orientation information of my good files (LSA) and of the ones that aren't working (LAS), then used mri_convert with the --in_orientation and out_orientation flags to change the dimensions in analyse format. The analyze images look fine in MRIcro and the orig.mgz looks fine in tkmedit, but yet again the Talairach fails in the same way. I've also tried this by reorienting the images using MRIcro but this has the same result.
When you view the analyze images using GWRaw MRI viewer http://www.wideman-one.com/gw/brain/orientation/orientterms.htm (supposed to read the image file without header information), they look the same, but I'm a bit unsure of this program.
I'm really at a bit of a loss to try and fix this problem now so any help or suggestions would be really appreciated!
Many thanks,
Dominic
Melbourne Neuropsychiatry Centre
Hi Bruce,
Thanks for your quick reply. Unfortunately the images are more than 10 years old and have only been stored on our systems in analyze format. I'm trying to source the original DAT tapes, but thought I'd try to fix the problem without them. In the main I was puzzled that tkmedit reads the converted image ok, but the talairach fails as if the orientation information was still wrong. If there are other things I could try it would be great to know, otherwise I'll just cross my fingers and hope for the tapes.
Thanks very much.
Dominic
-----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Monday, 2 November 2009 11:20 PM To: Dominic Brendon Dwyer Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] orientation problem resulting in talairach failure
Hi Dominic,
do you have access to any format previous to the analyze one? It doesn't
preserve orientation info, and all bets are off once you go through it.
cheers Bruce On Mon, 2 Nov 2009, Dominic Brendon Dwyer wrote:
Dear Freesurfer developers,
I have a few analyse images which are oriented incorrectly that are
part
of a larger dataset. When viewed in tkmedit, what is supposed to be coronal is actually displayed as axial and the sagittal is flipped. In these cases the Talairach transform fails spectacularly.
To try and fix the problem I used mri_info to determine the
orientation
information of my good files (LSA) and of the ones that aren't working (LAS), then used mri_convert with the --in_orientation and out_orientation flags to change the dimensions in analyse format. The analyze images look fine in MRIcro and the orig.mgz looks fine in tkmedit, but yet again the Talairach fails in the same way. I've also tried this by reorienting the images using MRIcro but this has the
same
result.
When you view the analyze images using GWRaw MRI viewer http://www.wideman-one.com/gw/brain/orientation/orientterms.htm (supposed to read the image file without header information), they
look
the same, but I'm a bit unsure of this program.
I'm really at a bit of a loss to try and fix this problem now so any help or suggestions would be really appreciated!
Many thanks,
Dominic
Melbourne Neuropsychiatry Centre
Hi Dominic
the big problem is that you'll never know if you have a left/right swap.
sorry :< Bruce On Tue, 3 Nov 2009, Dominic Brendon Dwyer wrote:
Hi Bruce,
Thanks for your quick reply. Unfortunately the images are more than 10 years old and have only been stored on our systems in analyze format. I'm trying to source the original DAT tapes, but thought I'd try to fix the problem without them. In the main I was puzzled that tkmedit reads the converted image ok, but the talairach fails as if the orientation information was still wrong. If there are other things I could try it would be great to know, otherwise I'll just cross my fingers and hope for the tapes.
Thanks very much.
Dominic
-----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Monday, 2 November 2009 11:20 PM To: Dominic Brendon Dwyer Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] orientation problem resulting in talairach failure
Hi Dominic,
do you have access to any format previous to the analyze one? It doesn't
preserve orientation info, and all bets are off once you go through it.
cheers Bruce On Mon, 2 Nov 2009, Dominic Brendon Dwyer wrote:
Dear Freesurfer developers,
I have a few analyse images which are oriented incorrectly that are
part
of a larger dataset. When viewed in tkmedit, what is supposed to be coronal is actually displayed as axial and the sagittal is flipped. In these cases the Talairach transform fails spectacularly.
To try and fix the problem I used mri_info to determine the
orientation
information of my good files (LSA) and of the ones that aren't working (LAS), then used mri_convert with the --in_orientation and out_orientation flags to change the dimensions in analyse format. The analyze images look fine in MRIcro and the orig.mgz looks fine in tkmedit, but yet again the Talairach fails in the same way. I've also tried this by reorienting the images using MRIcro but this has the
same
result.
When you view the analyze images using GWRaw MRI viewer http://www.wideman-one.com/gw/brain/orientation/orientterms.htm (supposed to read the image file without header information), they
look
the same, but I'm a bit unsure of this program.
I'm really at a bit of a loss to try and fix this problem now so any help or suggestions would be really appreciated!
Many thanks,
Dominic
Melbourne Neuropsychiatry Centre
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Dominic Brendon Dwyer