Hi - no, motion correction is done on the whole 3D volume with an affine transformation, so as you discovered it's not responsible for whole-slice effects like this.

What you are seeing is normally due to scanner problems - something to do with the timings of the slice acquisitions.

Steve.




On 23 Jan 2013, at 20:05, Andrew Yung wrote:

Hi there,

I am trying to do ICA analysis on rat brain experiencing silent seizure activity in a model of absence epilepsy.  I've run the data through MELODIC and it finds 17-20 components.  However, when I look at the spatial maps of the components, I see that many of the ICs basically show complete activation of all brain pixels belonging to a single slice.  I have 8 slices in my imaging volume, and 8 IC maps show whole-brain activation in individual slices.

I was thinking perhaps it was due to the motion correction, since I assume that the correction is done on a per-slice basis.  However, when I turn off MCFLIRT I get the same behaviour.

Would you have any thoughts as to what might be causing this apparent artifact?

Thanks for your help!

-andrew



---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------