Hi,
I have two questions--one regarding checking whether my general approach to preprocessing is correct, and the other regarding the the coregistration step that I am currently stuck on. I am new to SPM and have been seeing various approaches to these questions, but I'm not sure which options would work best in conjunction:
1) First, my basic prepreocessing question: does the following order of prepreocessing steps make sense, for a functional MRI project involving 155 subjects, using SPM 12b (or SPM 8, if that would be better for these steps):
-slice timing (completed)
-realignment: estimate and reslice (completed)
-coregistering: estimate only (or, should I do estimate and reslice? I thought to skip it as I read that excessive reslicing is not advised)
-segmenting (I have been told this is better than just normalizing--would this completely replace the normalizing step?)
-smoothing
Any thoughts on whether this make sense or whether I am missing any steps/options would be helpful.
----
2) Second question on coregistration step, which I am stuck on:
Currently, I have finished the realignment step for my functional images and want to coregister them with the T1s (or, if it is better, coregister the T1s with the functional images). Is one of the following options more recommended--or, what advantages/disadvantages do they have?
option 1: for each subject, coregistering their mean functional realigned image (mean raf image) to their mean T1 image. (that is, set the reference as the subject's mean structural image, and the source as the subject's mean functional image)
option 2: for each subject, coregistering their mean T1 image to their mean functional (mean raf) image (that is, set the reference as the subject's mean functional image, and the source image as the subject's mean structural)
(OR, is it advisable to do something else entirely, such as:-
-use additional "other images"
-use something other than the MEAN subject images in either T1s or functionals or both in considering these options (e.g. coregistering the mean t1 image with ALL the functionals for each subject?)
-use a mean file produced at a different step for the functionals rather than the mean *raf file for each subject?
------
Any advice would be very, very appreciated!
Thanks,
-Victoria
|