Hi Veronica,
au contraire, it does prevent from biasing your results...
Please have a look at the 2006 Brain paper by Dorsaint-Pierre or the 2001 Cerebral Cortex paper by Watkins: using an asymmetric template can enhance the asymmetries of the original data set, but the symmetrical template bypasses this problem.
Cheers,
Gwenaelle
> De: Veronica Witte <[log in to unmask]>
> Objet: [FSL] VBM template generation - why symmetrically?
> À: [log in to unmask]
> Date: Mercredi 2 février 2011, 20h32
> Hi,
> we read the fsl-vbm scripts and wondered why grey matter
> templates which derive from the subjects will be symmetrical
> (by right/left-flipping and averaging them)?
> \$FSLDIR/bin/fslmerge -t template_4D_GM \$mergelist
> \$FSLDIR/bin/fslmaths template_4D_GM -Tmean template_GM
> \$FSLDIR/bin/fslswapdim template_GM -x y z
> template_GM_flipped
> \$FSLDIR/bin/fslmaths template_GM -add template_GM_flipped
> -div 2
>
> In doing so and registering individual GM images to this
> symmetrical template, we think that hemisphere-dependent
> information might get lost, and this might bias results - ?
> Or is there a specific reason for the right/left-flipping?
> Thank you in advance for your help!
> best, Veronica
--------------------------------------------------------------------
Gwenaëlle Douaud, PhD
FMRIB Centre, University of Oxford
John Radcliffe Hospital, Headington OX3 9DU Oxford UK
Tel: +44 (0) 1865 222 523 Fax: +44 (0) 1865 222 717
www.fmrib.ox.ac.uk/~douaud
--------------------------------------------------------------------
|