Hi, Jun:
Thanks a lot for the detailed reply. But I am not sure if I did anything
wrong. The resulting int16 image does not look right. I attached the two
fig files for one plane of the brain (I read in the file and display in
matlab). Does this look right to you? Thanks!
Xue
Quoting Jun Ding <[log in to unmask]>:
> Hi, Xue,
>
> Yes, you can use the SPMt.img file. The type of that
> file is 'float'. There is a easy way in SPM to convert
> the 'float' type to 'int' as you want. On the
> interface of SPM2, there is a button 'ImCalc', click
> on that one,
> choose the file you want to work on, then you will
> need to input the 'Output filename'. After that, you
> will be asked to input the 'Evaluated Function'. Input
> 'i1', and then the type 'int16' file will be generated
> under that name you choose.
>
> Jun
> --- "HUA,XUE" <[log in to unmask]> wrote:
>
> > Thanks for the answer. I am sorry if I confused you
> > that I am hacking the
> > code. I want to generate a Pseudo-T map using
> > another image rendering
> > software called BrainSuite2 which read in analyze
> > format files. I prefer
> > using uint16 format. So I was trying to see if I
> > could read in the file
> > and convert to uint16, then use it for display.
> > Actually, could I just
> > use the SPMt.img directly? Does it contain the
> > Pseudo-T statistic if I
> > setup the program using variance smoothing? Thanks!
> >
> > Xue
> >
> >
> > Quoting Thomas E Nichols <[log in to unmask]>:
> >
> > > Xue,
> > >
> > > > I am using SnPM and I have a quick question with
> > SnPMt.mat
> > > > file. According to the comment, SnPMt.mat
> > contains a 1xS matrix of
> > > > the statistic of interest (either t or pseudo-t)
> > supplied for all S
> > > > voxels at locations XYZ. I assume this the
> > statistic for the correct
> > > > permutation. The matrix size for the file is 1 x
> > 1923595. This does
> > > > not conform to my image matrix size
> > (199x199x199=7880599) I fed into
> > > > the program. Does the SPM program chop off some
> > of the peripheral
> > > > regions? How could I translate the result back
> > into my original
> > > > image matrix? Thanks a lot for your time and
> > help!
> > >
> > > S is the number of voxels within the brain mask,
> > which will generally
> > > be quite smaller than the total number of possible
> > voxels in the 3D
> > > volume.
> > >
> > > To work with SnPMt use the XYZ variable. XYZ is a
> > 3xS matrix, giving
> > > you the x,y,z location of each column of SnPMt.
> > >
> > > Hope this helps! Happy hacking!
> > >
> > > -Tom
> > >
> > >
> > > -- Thomas Nichols --------------------
> > Department of
> > > Biostatistics
> > > http://www.sph.umich.edu/~nichols
> > University of Michigan
> > > [log in to unmask] 1420
> > Washington Heights
> > > -------------------------------------- Ann
> > Arbor, MI 48109-2029
> > >
> > >
> >
>
>
|