Print

Print
Hi - if you register two FA maps directly to the T1 then you can apply those transforms directly.  It is possible that you may get more accurate/consistent results by registering the two FA images to eachother first (or to the halfway space), averaging, and estimating a single transform to the T1, and then you would need to combine the relevant transforms together to get from native FA images to the T1, etc.

Steve.



On 11 Feb 2013, at 18:28, Brett Harton wrote:

Hi All,

I have DTI data from 2 time points and I want to draw several ROIs that are quite small, so I need to use the subject's T1 to draw them on.  I initially did the halfway transform for the 2 DTI data sets and drew larger ROIs directly onto the FA map.  My question is: if I register both FA maps (from the 2 time points) to the subject's T1 (the cleaner T1 of the 2 T1s collected) using the 'secondary images to apply transform' option in the FLIRT GUI, will the same transform be applied to both FA maps so that I do not need to run the halfway transform on the 2 FA maps?  Basically, I'm wondering if registering both FA maps to the same T1 takes care of putting them into the same space for analysis.

Hopefully this makes sense...thanks for any input!

Brett



---------------------------------------------------------------------------
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
---------------------------------------------------------------------------