Print

Print
Thank you very much for the response.

Just a follow-up: to best of my understanding, if my image is going to be resliced anyway at the next step of the preprocessing, it is preferable not to reslice it before. The question is then why in tutorial they use estimate & reslice for realignment, if I anyway after that there is slice time correction. Do the operations like slice time correction and smoothing reslice the image ?

On Tue, Jan 8, 2019 at 11:04 PM Angstadt, Mike <[log in to unmask]> wrote:
Yes, estimate just changes the image transform in the header without modifying the data portion of the image. However, the header is how SPM (and other tools) determine how to position the image when displaying it, and consequently how it relates to other images. So a T1 image that has been Coregister-Estimated to something else will definitely be positioned differently from an unmodified version of that T1 if you are using a viewer that takes the transform into account (like SPM's Check Reg). If instead you load the voxelwise data for those images and directly compare, the voxel values should be identical, it's just that the position of each voxel has changed when the image is oriented based on the transformation.

-Mike
________________________________________
From: SPM (Statistical Parametric Mapping) <[log in to unmask]> on behalf of John Gelburg <[log in to unmask]>
Sent: Tuesday, January 8, 2019 3:14:38 PM
To: [log in to unmask]
Subject: [SPM] coregister (estimate) in SPM12: is the source image actually jiggled / rotated?

Dear list,

Consider the face example from the SPM manual where T1 is used as the source image and mean EPI as the reference. What happens after running coregister (estimate)? The T1 should be realigned to mean EPI. I was thinking that, the estimate option does not actually rotates the image, but only changes its header. If so, I should not expect to see the actual change in T1 orientation (e.g., if I overlay my functional activations they should not be aligned). But if now I segment the T1, the coregistration information from the header will be taken into account. To complicate the things, when I run checkreg to original T1 and the T1 after co-registration the values seem not to be identical. So, the T1 image might have changed after coregister (estimate).

I would appreciate if someone can clarify the things to me.

Many thanks,
John
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues