Dear Markus
There is no generally accepted recipe for how to do spatial preprocessing, as there are different ways to combine the different steps.
Generally, you do not need to reslice your images in any of the linear processing steps, i.e. realignment or coregistration, as the transformations are stored in the mat-files (SPM99, SPM2) respectively in the NIFTI-headers (SPM5). Hence SPM knows about the previous steps and a reslicing (which would degrade image quality due to interpolation) is not necessary.
Conseqently the only time, where a reslicement is necessary is the spatial normalisation, which is non-linear and cannot be stored with the images.
Also, be aware of the fact, that there are residual (non-linear) differences between the EPIs and your T1 image, which are not corrected during co-registration. As your goal is to register the functional data as good as possible into the template space, you might want to use the following procedure:
- Realign the EPIs and create a mean EPI
(Realign & Reslice, Reslice: "mean only")
- Cogregister the T1 to the mean EPI
(Source T1, target mean EPI)
- Normalise the mean EPI using the unified segmentation algorithm in SPM5, which avoides the need to use a specific EPI template but brings your data nto MNI/ICBM space
- Apply the derived normalisation-parameters to all your images (mean EPI, individual EPIs, coregistered T1)
- Smoth your data, e.g. using an 8mm kernel, which is usually sufficient for fMRI data
Go on with the stats, including the realignment parameters as confounds in the design matrix. Also, if you include time derivatives for your regressors as well, there is no real need for slice time correction, which is why I did not include it in the "recipe" outlined above.
Hope ths helps
Simon
----- Original Message -----
From: Markus Burgmer <[log in to unmask]>
Date: Tuesday, January 9, 2007 12:11 pm
Subject: [SPM] manual
> dear list.
>
> is there a good manual about the preprocessing procedures in spm?
>
> i would like to have information about kinds of questions like:
>
> - do i get better results in normalisation with coreg my epi on the t1
> image and normalize the latter with the t1 template (there is a mail in the
> spm list archives, i think it was john ashburners, who prefers normalization with the mean epi image,
> but it was according spm99)?
>
> or - do i have to estimate & resclice images in the coreg step or is it
> enough to just estimate (by the way, reslicing "blows" up my single epi
> images to a size of the t1 image (160mb) and therefore it takes hours).
>
> the spm5 manual is no real help, because it just describes the single
> features, but gives no advise what do do. the typical related papers are no
> real help either, because their are overloaded with mathematical
> terms and procedures, i do not understand.
>
> so anything around for a dummy who just want to drive a car without being a mechanic?
>
> thanks for the help.
>
> markus
>
-----------------------------------------
Simon Eickhoff
Institut for Medicine (IME); Research Center Jülich
Leo-Brandt-Str. 5; 52425 Juelich, Germany
Phone + 49 2461-61-5219 / Fax + 49 2461-61-2820
e-mail: [log in to unmask] / [log in to unmask]
|