I was referring to the resulting combined act and deact map, but the same
effect can be seen in all w*_mean-maps.
BR,
Tuomo
On Thu, 10 Jul 2008, Mark Woolrich wrote:
> Which file were you looking at for the probmap?
>
> Activation class is w2_mean.
> Deactivation class is w3_mean.
> Non-activation class is w1_mean.
>
> Also worth noting that using the --timingon will slow the execution
> time quite a bit.
>
> Cheers, Mark.
>
> ----
> Dr Mark Woolrich
> EPSRC Advanced Research Fellow University Research Lecturer
>
> Oxford University Centre for Functional MRI of the Brain (FMRIB),
> John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
>
> Tel: (+44)1865-222782 Homepage: http://www.fmrib.ox.ac.uk/~woolrich
>
>
>
>
> On 10 Jul 2008, at 09:30, Tuomo Starck wrote:
>
> > Hi,
> >
> > We faced unexpected thresholding results from the nonspatial mixture
> > modelling applied for z-score maps. There are hardly any other
> > values in the
> > probmap than ones, thousands of them, and some tens of zeros and
> > additionally very few values between 0-1. Gamma-Gaussian fits look
> > ok (attch
> > mmfit.png) and mm command usage should be ok (mm --mask --sdf --
> > timingon
> > --ns -V --logdir).
> >
> > So in the end there is not much thresholding done for our z-score
> > maps as
> > mixture modelling for some reason assigns for almost every gamma-
> > fitted
> > voxel a value of 1 in the probmap. Something seems to go wrong, any
> > help
> > greatly appreciated!
> >
> > BR,
> > Tuomo Starck
> >
> > <mmfit.PNG>
>
>
|