On Thu, 3 Jul 2008 14:25:57 +0200, Christophe Phillips <[log in to unmask]> wrote:
>Just my 2EUR cents,
> From my (little) experience, spm5 segmentation (unified framework) is
>superior to spm2's one. :-)
>BUT spm5 segmentation is much more sensitive (in a bad way) to the
>original orientation of the image to segment, i.e. an image that is
>easily processed with spm2 without the need for manual realignment may
>NOT be so within spm5. :-( Changing the origin and orientation of the
>image such that it's closer to the MNI space, always solved the problem
>for me. (*)
>BTW, Dartel needs the files produced by spm5's segmentation to work.
>(*) on my "wish list" of tools for spm: I would love to be able to
>interactively reorient one image within the "checkreg" display...
This wish is easy to fulfill. Simply copy the attached file in the spm5/spm_orthviews folder and
use the right mouse button to find the AC-setting tool. As all spm_orthviews plugins this plugin
will only work with the checkreg function and not with the display tool.
Christian Gaser, Ph.D.
Assistant Professor of Computational Neuroscience
Department of Psychiatry
Friedrich-Schiller-University of Jena
Jahnstrasse 3, D-07743 Jena, Germany
Tel: ++49-3641-934752 Fax: ++49-3641-934755
e-mail: [log in to unmask]
>chetelat a écrit :
>> We also had several problems of segmentation with SPM5, so that for several
>> analyses we still used SPM2. However, we've tried The Dartel toolbox and it
>> seems good, there were less misclassification problem and findings were much
>> more like those obtained with SPM2. This was for gray matter, but it may be
>> the same for CSF...
>> Hope this helps
>> -----Message d'origine-----
>> De : SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] De la
>> part de Herve Lemaitre
>> Envoyé : mardi 1 juillet 2008 23:05
>> À : [log in to unmask]
>> Objet : [SPM] SPM5 segmentation
>> Dear John and SPM list,
>> Using the unified segmentation in SPM5 (with the more recent updates
>> installed), we noticed that for a large proportion of subjects (30%), voxels
>> belonging to the skull were misclassified into the CSF class (see example
>> - Top left: normalized MRI.
>> - Top right: segmented CSF map using SPM2 optimized VBM.
>> - Bottom left: SPM5 segmented CSF map.
>> As a result, the CSF volume estimation for this subject is 705 cm3 instead
>> of 442 cm3 which seems more reasonable using the optimized VBM with SPM2.
>> Has anyone else noticed that and are there any ways to improve this result?
>> Thanks in advance for your suggestions,
>> Herve Lemaitre
>> Herve Lemaitre PhD
>> Clinical Brain Disorders Branch, NIMH
>> Building 10, Room 3C-108
>> 10 Center Drive, MSC 1384
>> Bethesda, MD 20892-1384
>> Phone: (301) 594 9922
>> Fax: (301) 451 4269
>> [log in to unmask]
>Christophe Phillips, Ir, PhD
>FRS-FNRS. Research Associate
>Lecturer in Applied sciences
>Cyclotron Research Centre, B30
>University of Liege, Sart Tilman
>4000 Liege, Belgium
>Tel: +32 4 366 2316 (secr.)
> +32 4 366 2366
>Fax: +32 4 366 2946
>email: [log in to unmask]
>web : http://www.ulg.ac.be/crc