Dear João,
these images are the modulated images, where the modulation with the volume changes (due to spatial normalization) may lead to signal values > 1. However, in addition to Christophers answer it might be interesting to note that the segmentations of the VBM8 toolbox are based on a partial volume estimation (PVE) model. This PVE additionally models mixed classes between two tissue classes (GM/WM and GM/CSF) and the interpretation is that a particular voxel with a value of e.g. 0.8 (without modulation) consists of 80% GM and 20% of either WM or CSF (depending on the localization of this voxel). Thus, it is not a pure probability that this voxel belongs to GM.
The effect of the PVE is mostly obvious in the basal ganglia, where large partial volume effects occur (e.g. thalamus or putamen).
Regards,
Christian
--
____________________________________________________________________________
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]
http://dbm.neuro.uni-jena.de
On Mon, 4 Oct 2010 00:48:30 +0100, João Duarte <[log in to unmask]> wrote:
>But if I display one modulated normalized image (m0wrp1... for
>example) there are voxels with intensity higher than 1, so it can't be
>a probability value...
>
>2010/10/3 Watson, Christopher <[log in to unmask]>:
>> If you're asking about the c1, c2, c3 images, then the signal intensity represents the probability that the voxel is a member of that particular tissue class.
>> ________________________________________
>> From: SPM (Statistical Parametric Mapping) [[log in to unmask]] On Behalf Of Jo�o Duarte [[log in to unmask]]
>> Sent: Sunday, October 03, 2010 11:36 AM
>> To: [log in to unmask]
>> Subject: [SPM] Segmentation signal intensity
>>
>> Dear SPM community,
>>
>> I was wondering about how to interpret the intensity signal in the
>> segmented images of GM, WM and CSF. What's the range of the signal
>> intensity values in the GM image, for example? It depends on the range
>> of the raw image or does it have a specific interval?
>>
>> And what does it mean to have a voxel with higher intensity signal
>> than another voxel, in the GM image? Does it mean that voxel has
>> higher gray matter volume if it is a non-linear only modulated
>> normalized GM image?
>>
>> Thanks for any help
>>
>> Jo�o Duarte
>>
|