Hi Pia,
> I have attached *con images from the same data set, once done on the
> smwc1* and once on the swc1*... does the grey ghost around the
> smwc1*. make sense (check top image)? The smwc1* images of each
> single subject look fine.
I think the grey ghost just corresponds to very low values close to zero - but
not quite zero. I'm assuming that you didn't use any grey matter masking.
The modulated data are saved as floating point, whereas the unmodulated data
are saved as unsigned bytes. These remain in their original formats after
smoothing. The black areas correpond to where there was an exact zero in one
or more of the images. There would be more of these for the unsigned byte
data (because values less than about 1/512 would be rounded down to zero).
For the floating point data, the extent of the grey ghost is the same size as
the width of the non-zero parts of the Gaussian function used to do the
smoothing.
All the best,
-John
|