Hi Falk,
They are in mm. One reason that the distances are larger than one might expect when using reconMNI is that in addition to the shape differences, these distances also include the residual differences in the position of the structure after affine registration. This is by design, as it allows first to be sensitive to both shape and pose differences using reconMNI. If you are interested in discriminating between the two, you might consider running using reconNative+rigidAlign as well as this removes the position differences.
Hope that helps!
Best,
Eelke
> On 21 Nov 2015, at 17:25, Falk Mancke <[log in to unmask]> wrote:
>
> Dear FSL Experts,
>
> I am having a quick question concerning the 4D file of the vertex analysis using first_utlis together with the ReconMNI option.
> I hope I understand it correctly that these values (one image per subject) represent the signed perpendicular distance from the average surface of the particular group under study. In our case the values range more or less between +/- 2.5.
>
> My question is what is the unit of these values? Are these milimeters (would be a surprisingly large range then) or are these already Z-scores, i.e. standard deviation units. If the latter is true what have they been originally?
>
> Kind regards and thank you very much in advance for your help!
>
> Falk
|