I have .mesh file and a corresponding MRI file in Analyze format. Mesh was generated with Brainvisa 3.1 or upper.
When loading the .mesh with the new public 3.1 release, the mesh and MRI are OK. But when loading the same files with the old anatomist 3.0.2 the mesh is not registered with the MRI (a rotation and maybe a translation pb too).
Since I use software that assumes the mesh an the mri are in the old coordinate system convention I'd like to know what transformations I need to apply the mesh and/or MRI to have registered files within Anatomist 3.0.2
Thanks
Coordinate system change between Anatomist 3.0.2 and 3.1
-
riviere
- Site Admin
- Posts: 1361
- Joined: Tue Jan 06, 2004 12:21 pm
- Location: CEA NeuroSpin, Saint Aubin, France
- Contact:
Re: Coordinate system change between Anatomist 3.0.2 and 3.1
Hi,
The internal coordinates system has not changed, *but*:
- the way aims/anatomist interpret SPM normalized images especially has changed by default: read http://brainvisa.info/doc/aimsdata-3.1/ ... ormat.html do normalized SPM images are likely to be loaded flipped compared as before if you stay with the default configuration. This is an important change that we tried to make apparent in the changelogs.
- transformations to foreign referentials that come with images (either explicitely like in NIFTI format, or implicitely like in SPM normalized images) have a bit changed: when loaded in Anatomist, transformations to the SPM MNI template referential are now going to the actual SPM referential, not a flipped version in our Aims orientation like it did before. I think it will generally not impact users since you normally compare images providing transformations to the same referential, whatever it is. But if you have checked 'apply SPM/Nifti referential/transformation information to loaded objects' (or something like this), then it is possible that the SPM volume has such a transformation (if it is normalized, typically), but the mesh doesn't. Or more precisely, in your case, both do have information (if they have been produced by the T1 pipeline of Brainvisa), but Anatomist 3.1 correctly interpret is, whereas Anatomist 3.0.2 did not interpret such information for meshes.
I guess it's something like that...
Denis
The internal coordinates system has not changed, *but*:
- the way aims/anatomist interpret SPM normalized images especially has changed by default: read http://brainvisa.info/doc/aimsdata-3.1/ ... ormat.html do normalized SPM images are likely to be loaded flipped compared as before if you stay with the default configuration. This is an important change that we tried to make apparent in the changelogs.
- transformations to foreign referentials that come with images (either explicitely like in NIFTI format, or implicitely like in SPM normalized images) have a bit changed: when loaded in Anatomist, transformations to the SPM MNI template referential are now going to the actual SPM referential, not a flipped version in our Aims orientation like it did before. I think it will generally not impact users since you normally compare images providing transformations to the same referential, whatever it is. But if you have checked 'apply SPM/Nifti referential/transformation information to loaded objects' (or something like this), then it is possible that the SPM volume has such a transformation (if it is normalized, typically), but the mesh doesn't. Or more precisely, in your case, both do have information (if they have been produced by the T1 pipeline of Brainvisa), but Anatomist 3.1 correctly interpret is, whereas Anatomist 3.0.2 did not interpret such information for meshes.
I guess it's something like that...
Denis
Re: Coordinate system change between Anatomist 3.0.2 and 3.1
OK I finally found my way out of this.
I convert my analyze file to GIS/IMA with Brainvisa 3.1 AimsFileConvert
and converted the outputed GIS/IMA volume to analyze with Brainvisa 3.0.2 AimsFileConvert
the resulting analyze volume is valid within anatomist 3.0.2 !
Thx for your help.
I convert my analyze file to GIS/IMA with Brainvisa 3.1 AimsFileConvert
and converted the outputed GIS/IMA volume to analyze with Brainvisa 3.0.2 AimsFileConvert
the resulting analyze volume is valid within anatomist 3.0.2 !
Thx for your help.