Hi
mainly because you don't want the large amount of smoothing that people
typically do in pre-stats. If you want to get rid of residual motion,
try temporal filtering and mc first, run ICA on that and then use
spatial filtering to blur.
cheers
christian
On 21 May 2004, at 15:48, Jennifer Bramen wrote:
> Hi Everyone,
>
> I was wondering why using the raw data, and not the filterered_func
> data created after prestats, is
> suggested if you want to use MELODIC to de-noise. It seems to me that
> you would want to
> remove any residual effects of motion AFTER motion correction, not
> before.
>
> Thanks!
>
> Jen
>
>
> Jennifer Bramen
> UCLA Brain Mapping Center
> 660 Charles Young Dr. So
> Los Angeles, CA 90095
>
--
Christian F. Beckmann
Oxford University Centre for Functional
Magnetic Resonance Imaging of the Brain,
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
Email: [log in to unmask] - http://www.fmrib.ox.ac.uk/~beckmann/
Phone: +44(0)1865 222782 Fax: +44(0)1865 222717
|