Dear Kelvin:
>
> In our fMRI experiment, we have an oblique axial cutting for both
> multiphase EPI image and a corresponding T1 image (all not cover the whole
> brain). When the T2 EPI is used for normalization, the result is
> different from when I use T1. I assume that T1 would produce a more
> accurate normalization.
If you're trying to normalize the T1 image then you should use the T1
template. Also, in general the T1 image because it is a better scan will
produce a more accurate normalization than the T2-EPI scan.
Why don't you get a T1 that covers the whole brain? It shouldn't take that
much more time and might produce a better normalization than a partial T1
volume. You then have to make sure you specify the origin of the T1 and EPI
images correctly.
>
> I'm thinking what's the process to use T1 as normalization standard and
> wrap the T2 in the same way rather than use T2 as normalization
> standard. Is the correct procedure:
> 1) realign and reslice T1 & T2 sets
> 2) normalize T1
> 3) coregister realigned T2 (object) to realigned T1 (target)
> 4) statistical analysis on coregistered T2
> 5) overlay statistical maps with normalized T1
I don't think the above is quite right. Here are the steps we use.
1) realign EPI dataset and just produce a mean image (don't reslice the
other images).
2) coregister T1 to mean of realigned EPI images.
3) normalize the T1.
4) apply the normalization parameters from the T1 to the EPI images.
5) now stats and T1 will be in normalized space.
If you want to use the T1 template as a standard at your site you can change
this in spm_defaults. Otherwise choose the proper template when you setup
the normalization.
regards,
Darren
-------------------------------------------------------------------
Darren R. Gitelman, M.D. E-mail: [log in to unmask]
Cognitive Neurology and Voice: (312) 908-9023
the Alzheimer's Disease Center Fax: (312) 908-8789
WWW: http://www.brain.nwu.edu
Northwestern Univ., 320 E. Superior St., Searle 11-470, Chicago, IL 60611
-------------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|