 |
 |
Amy, > So...before considering the first level, I like statr from the beginning. > Say from preprocessing step... Yes, that makes sense :) > Usually, the steps which I perform are > 1. Segmentation ... of what? > 2. Normalize: Write (This actually takes ur template to MNI which I > don't want... So how to handle this step?) ... as I said, omit it here. Not sure what you mean by "template to MNI space" as usually a template is in normalized (e.g., MNI) space to begin with. > 3. ReALIGN ANd unwarp ... which I would do in native space anyway. > 4. Coregistration: Estimate ... you can happily do that in native space, too. > 5. Normalize:Write ... again? I guess this pertains to the functional images. In this case, you would probably select the con-images from your first level statistics. > 6.Smooth Yes. > What changes needs to be done in my preprocessing steps... > I actually want to do fmRI MRS correlation,. In that case you may want to optimize your processing for resting state processing in any case, see the several hundred papers addressing this, as well as the toolboxes specifically designed for this purpose. > I am attaching a paper where they used native space processing for that... > Please go thru the page no. 4 of this article > and kindly let me knw how to do it... No. You asked me a question to which I gave you an answer (or rather, an opinion), in the hope that it would be helpful. I am not going to go through papers with you. Cheers, Marko > On Fri, Oct 17, 2014 at 11:41 AM, Marko Wilke > <[log in to unmask] > <mailto:[log in to unmask]>> wrote: > > Hi Amy, > > I'd like to do an fMRI experiment wherein, instead of taking the > subject template to MNI template, I need them in their native > space. MAy > I know how to perform such experiment? > > > This will actually be easier on the first level as you simply skip > the step of spatial normalization :) It becomes slightly more tricky > afterwards but not overly so: instead of normalizing all EPI images, > you now only normalize the con-images and perform second-level > statistics on these. One issue to be aware of is that more extensive > interpolation schemes (such as higher level b-splines in spm) will > lead to out-of-brain voxels not being absolutely 0, which may > interfere with mask generation. You can use an explicit mask to that > effect. > > Incidentally, I think this is actually the preferred approach: > mathematically, they should yield pretty much equivalent results but > they don't. The issue, of course, is interpolation, and without > having explored this exhaustively, I believe that one can make the > case that it is better "to interpolate" the results (generated from > non-interpolated raw data) than "to interpolate" the raw data. But > that is only my 2 cents. > > Cheers, > Marko > > -- > ______________________________________________________ > PD Dr. med. Marko Wilke > Facharzt für Kinder- und Jugendmedizin > Leiter, Experimentelle Pädiatrische Neurobildgebung > Universitäts-Kinderklinik > Abt. III (Neuropädiatrie) > > Marko Wilke, MD, PhD > Pediatrician > Head, Experimental Pediatric Neuroimaging > University Children's Hospital > Dept. III (Pediatric Neurology) > > Hoppe-Seyler-Str. 1 > D - 72076 Tübingen, Germany > Tel. +49 7071 29-83416 > Fax +49 7071 29-5473 > [log in to unmask] > <mailto:[log in to unmask]> > > http://www.medizin.uni-__tuebingen.de/kinder/epn/ > <http://www.medizin.uni-tuebingen.de/kinder/epn/> > ______________________________________________________ > > -- ____________________________________________________ PD Dr. med. Marko Wilke Facharzt für Kinder- und Jugendmedizin Leiter, Experimentelle Pädiatrische Neurobildgebung Universitäts-Kinderklinik Abt. III (Neuropädiatrie) Marko Wilke, MD, PhD Pediatrician Head, Experimental Pediatric Neuroimaging University Children's Hospital Dept. III (Pediatric Neurology) Hoppe-Seyler-Str. 1 D - 72076 Tübingen, Germany Tel. +49 7071 29-83416 Fax +49 7071 29-5473 [log in to unmask] http://www.medizin.uni-tuebingen.de/kinder/epn/ ____________________________________________________