Hello Matthew,
Could this problem be obviated another way by using an image without
a great deal of susceptibility artifact (e.g., fast SE, or SPGR)
instead of one's GE EPI images to "Determine Parameters" of spatial
normalization (which is what many people do, of course)? And,
wouldn't this be a better choice than masking one's EPIs because
then one would determine normalization parameters from the
subject's full brain image? The susceptibility artifact would still
be there, but it would not interact with the normalization.
I'd be very interested to hear what you think about this.
Thanks,
Eric
Quoting Matthew Brett <[log in to unmask]>:
> Hi,
>
> > In our brain scans, we see a significant drop-out in the
> orbital
> > frontal cortex (OFC) , which results in images with very less
> coverage
> > in the frontal regions. When we normalize those images to the
> SPM
> > standard template, the resulting images look just like the
> template with
> > all the frontal areas present!?!?!?!?!
>
> You may want to have a look at this page:
>
> http://imaging.mrc-cbu.cam.ac.uk/imaging/MaskedEpiNormalization
>
> Best,
>
> Matthew
>
|