> Date: Mon, 9 Sep 2013 15:02:48 +0100
> From: [log in to unmask]
> Subject: Re: [FSL] DTI processing/preprocessing 2nd take
> To: [log in to unmask]
>
> Hi Alain
>
> Saad replied to your post a few days ago:
>
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1309&L=FSL&F=&S=&P=48424
>
> Cheers
> Stam
>
>
> On 9 Sep 2013, at 10:54, Alain Imaging wrote:
>
> > Sorry for the repost but I'd really like to have some opinion in this matter.
> >
> > Dear all, I recently started to work on DTI images for the first time.
> > What I want to do is use the BPM toolbox for performing correlation between some PET based images, grey matter density images and diffusion parameters as FA and MD.
> > To do so I need to register FA and MD maps of each subject in the standard MNI space (the same in which I already registered PET images and GM density images).
> > I'd like to know if the pipeline I used is good enough, if there is something to ameliorate or something to completely change.
> > 1) I merge all the DTI images (once for each direction)
> > 2)I perform eddy current correction (using eddy_current rather than topup&eddy because I have not two images acquired with different pahse encoding)
> > 3) fit the tensor using DTI fit
> > 4)register FA map onto the subject's T1 using flirt, 12 DOF, nearest neighbor interpolation, correlation ratio cost function;
> > 5) registering the subject's T1 onto the 2mm MNI template (flirt+fnirt)
> > 6)applying the warping from step 5 to the FA map registered onto the T1
> > 7)applying the warping from step 5 AND the affine matrix from step 4 to MD map.
> >
> > At the end I have the FA and MD maps of each subject in the standard space.
> > While these FA and MD maps "nicely" overlap with my other modalities in the standard space, they look rather strange. In particular at the border there are some "spike" and "blurriness". I'm pretty happy with the result but I wonder if there is something I can do better.
> >
> > Another question regard the smoothing of FA and MD maps: all my other modalities are smoothed at 8x8x8. I think I should smooth the FA and MD maps too, in order to perform analysis with the other modalities, but somehow I feel unconfortable in smoothing FA and MD maps. Should I smooth or not ?
> >
> > Sorry for the wall of text, I hope in some answers,
> > Best
> > Alain
> >