Hi SPM experts,
I'm a beginner of SPM and hope someone can clarify my concept. In the fMRI preprocessing pipeline, I was taught to, after realignment, perform coregister (using mean EPI images as "reference", T2 images as "source image," and T1 images as "other image.") in a single subject's data. My understanding is that this with match T2 and T1 images with the mean realigned and resliced EPI images.
In normalization, I specified EPI.nii as the "template image," mean EPI images as the "source image," and the realigned EPI images (those entitled rxxx) as "images to write." The matching of mean EPI images to EPI.nii will produce the estimated parameters (the file called xxx.sn), and rxxx images will be matched with EPI.nii according to the parameters in xxx.sn file.
Is my understanding correct?
If yes, If my interest is just to get an average group activation map using one-sample t-test, is it necessary to perform coregister? (Because I think I can spare the time to collect T2 images.) What is the reason to match T2 and T1 images with the mean EPI images?
Thanks a lot.
Mike
|