Hi Eric,
> We've been trying to undrestand to what extent the resolution of the prior
> probability maps affects the outcome of the segmentation.
I think this is an intersting question and I shall look forward to
other's explaining it... however, I think it would be vital to know what
software version you used, SPM2 or spm5?
> segmentations on the same brain using the VBM5 toolbox (by Christian
> Gaser) and a pediatric prior probability maps (courtesy of Cincinnati
> Children's Hospital Medical Center) at 1x1x1mm resolution, then with the
> same prior maps at 2x2x2mm (created out of the first one using
> spm_reslice).
One could argue that the interpolation of this step (which is likely
made worse by the smoothness?) would contribute to the difference in
quality. If you are interested to look into this further, e-mail me
off-list and I will lokk if I can find the unsmoothed version here
somewhere.
> We got rather significant differences in the segmentation
> results, a sample of those results is attached to this post
I must say I find these hard to judge as they may be partial volume
effects brought about by slightly different positions... especially the
gyrus you marked that looks so different: if the one image is just a tad
more tilted than the other you could get effects like this. Have you
tried coregistering the two results and then do a difference image?
> We couldn't even tell the difference
> between the 2 prior maps.
Agein, perhaps a difference image in imcalc (i1-i2) would help to shed a
light on this.
Best,
Marko
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|