> Hi Gwenaelle,
>
> Thanks for the email. Just one or two things I'm not
> too sure about. Say I
> divide the colourbar's intensity by 20 to produce a new
> colourbar with the
> range 0-5; how can I be certain that the significance value
> at a certain
> intensity in my tstat map is the same as the corresponding
> colour in the
> colour bar? It looks about right but I'm not sure..
> Secondly, setting the
> min of my colourbar at 1.8, would that be done using the -l
> lower threshold
> option within fslstats?
>
> Thanks again,
> Ronan
>
> 2009/2/27 Gwenaëlle DOUAUD
> <
[log in to unmask]>
>
> > Hi Ronan,
> >
> > the min value is a cutoff threshold in fslview, but
> the max one is not, it
> > just "saturates" the image (with my own
> non-English way of saying it!). If
> > the stats image you present are showing t-value say
> >1.8 and "saturated" at
> > 5, then all you need to do is something like:
> > fslmaths colourbar -div 20 new_colourbar and in
> fslview, set the min of
> > your new_colourbar at 1.8. You will then have a
> colourbar of min 1.8 and max
> > 5..
> > Alternatively, you can just add/place the numbers
> "1.8" at the bottom of
> > your colourbar in your figure and "5" at the
> top, that will do!
> >
> > Cheers,
> > Gwenaelle
> >
> >
> > --- En date de : Ven 27.2.09, Ronan Joseph Kelly
> <
[log in to unmask]> a écrit
> > :
> >
> > > De: Ronan Joseph Kelly <
[log in to unmask]>
> > > Objet: [FSL] Colour Bar Significance Values
> > > À:
[log in to unmask]
> > > Date: Vendredi 27 Février 2009, 17h43
> > > Dear FSL experts,
> > >
> > > I have finished conducting a FSL-VBM analysis and
> have just
> > > produced my
> > > corrected tstat maps highlighting statistically
> significant
> > > voxel clusters
> > > between groups. The completed maps are produced
> with a
> > > colour gradient,
> > > highlighting the thresholded range of t values
> applied in
> > > the study
> > > (approximately t = 1.8 and above). My question
> concerns the
> > > use of colour
> > > bars to highlight different significance values
> in my maps.
> > > The 'colour
> > > significance bar' I have been trying to use
> is
> > > colourbar.nii.gz posted
> > > earlier this year in the FSL archives. However,
> the
> > > significance range in
> > > this is 0 - 100, so how do I threshold this to
> correspond
> > > to the values
> > > associated with the maps in my study? I've
> been trying
> > > to do it manually but
> > > no luck so far. Any ideas anyone?
> > >
> > > Many thanks,
> > > Ronan
> > >
> > > --
> > > Ronan Kelly
> > > Neuroinflammatory Research Group
> > > Trinity College Institute of Neuroscience
> > > University of Dublin, Trinity College
> > > Dublin 2
> > > Ireland
> >
> >
> >
> >
> >
>
>
> --
> Ronan Kelly
> Neuroinflammatory Research Group
> Trinity College Institute of Neuroscience
> University of Dublin, Trinity College
> Dublin 2
> Ireland