> we would like to do VBM analysis on a group of patients and controls, but
> we have acquired only 2D spin echo images with 4mm slice thickness
> (0.9x0.9x4mm spatial resolution) of the whole brain volume.
> Would these images be appropriate for VBM, assuming that of course we have
> the same images for patients and controls?
The images would be appropriate, providing that the images could be segmented
well. The 4mm slice thickness means that there is likely to be a lot of
partial volume voxels (that contain a combination of different tissue
classes, such as GM+WM or GM+CSF). The SPM99 segmentation model does not
account for partial volume, so the results may not be as good. There are a
few pieces of segmentation software out there that can cope with partial
volume, so someone may like to write in and tell us about them (Steve?).
> Which smoothing should we use?
I generally suggest about 12mm FWHM for grey matter. The optimal amount
depends on registration accuracy, noise and the volumes of structural
differences you want to detect.
> Is it necessary to do the Jacobian modulation?
I prefer to see Jacobian modulation. Other people don't.
> Can somebody please explain to us how we would practictally do it?
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0007&L=spm&P=R16328
> We have tried so far to do segmentation according to Good and the analysis
> looks ok, but the data (gray and white matter fractions) are different than
> those reported in the paper by Good published on Neuroimage 2001.
> I guess that would depend on the type of images.
>
> Last question: is it necessary to do maximum inhomegeneity correction?
It depends how much bias your images contain. We found that including bias
correction didn't have a negative impact on segmentations of perfectly
uniform (simulated) images. However, if you have perfectly uniform images,
then you may get better results by not including bias correction. (SPM99
Bias correction actually introduces bias into uniform images. I have
identified the cause of this problem and am working on getting a fix out.)
Best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|