Print

---

[log in to unmask]" type="cite"> Hi Rich,

>>So are you suggesting I co-register my mean functional image to the EPI template, skip segmentation, then run 'Normalize:Estimate & Write' with the EPI template file as the template image under the Estimation options?
Yes, with the exception that you do not have to coregister the mean functional to the EPI template as long as the origin (and orientation) of your functionals does not differ too much from MNI-space (origin close to the AC). You can check this with the 'Checkreg'-utility. You´d then run 'Normalize:Estimate & Write' with the mean functional as the source image (to estimate the transform to the EPI-template) and as 'images to write' you´d select all the realigned functionals for this subject/session. This was the usual normalise procedure several years ago and was replaced by the unified segmentation in SPM5. I didn´t know it is possible to run unified segmentation directly on the functionals as suggested by Eva.

Best,
Michel

> Subject: Re: [SPM] normalization without anatomy
> From: [log in to unmask]
> Date: Wed, 18 Aug 2010 11:11:24 +1000
> To: [log in to unmask]
>
> Hi Michel,
>
> Thanks for your help. I'm a bit of a beginner and have just been following the SPM5 manual tutorial whereby I co-register my anatomical image to the mean functional, segment the structural image and then use the _sn.mat file generated as the parameter file to 'Normalize:Write' the functional images. So are you suggesting I co-register my mean functional image to the EPI template, skip segmentation, then run 'Normalize:Estimate & Write' with the EPI template file as the template image under the Estimation options?
>
> I appreciate the help here . .
>
> Rich
>
>
> On 18/08/2010, at 10:51 AM, michel grothe wrote:
>
> > Hello Rich,
> > in this case you can (or have to) normalise the functionals directly to the EPI-template provided by SPM using the old-fashioned Normalise button. This procedure may be less acurate and is generally not recommended anymore. However, I guess you´d then have to normalise all your functionals this way to have consistent preprocessing on your images. Don´t know if a more accurate spatial normalisation (unified segmentation or DARTEL) would be worth omitting those 3 subjects...
> >
> > Hope this helps,
> > Michel
> >
> > > Date: Wed, 18 Aug 2010 09:55:35 +1000
> > > From: [log in to unmask]
> > > Subject: [SPM] normalization without anatomy
> > > To: [log in to unmask]
> > >
> > > Dear SPM List,
> > >
> > > We have collected functional and anatomical scans on 30 people in a study on emotional regulation, and I wish to preprocess the functional images for a standard 2nd-level analysis as described in the SPM manual. However it turns out we are missing the anatomical scans for 3 people we would normally use in the coregistration, segmentation and normalization steps. My question is it possible to normalize the functional images from these 3 people without their anatomical scans?
> > >
> > > Ta,
> > >
> > > Rich
>

--
Mag. Dr. Eva-Maria Seidel
Research Associate
Department of Clinical, Biological, and Differential Psychology
Faculty of Psychology, University of Vienna

http://psychologie.univie.ac.at/klinische-psychologie


Liebiggasse 5, A-1010 Wien
T +43-1-4277-47895
F +43-1-4277-47899
[log in to unmask]