Hey all,
Something I realized but have not had to use so far is that you have the
DICOM images of your Siemens MoCo series, I believe that the realignment
parameters resulting from realignment applied by the scanner are stored
in the header.
You can use the that the function dicm2nii to extract them automatically
in a json file:
http://uk.mathworks.com/matlabcentral/fileexchange/42997-dicom-to-nifti-converter--nifti-tool-and-viewer
Just make sure you call the following line in your script before you
call the dicm2nii function:
setpref('dicm2nii_gui_para', 'save_json', true);
You might wanna plot those realign parameter first. If you they look
"fine", you could potentially use them in your GLM.
As you mentioned, what is not clear: what type of realignment is used to
generate the MoCo series.
Also but someone corrects me on this: I think that using the MoCo series
sort of prevents you from using the "realign and unwarp" from SPM.
> The other problem is that I only have MoCo series and when I check the
> realign parameter from SPM process, I can see nearly great motions
> (0.5 to 1mm on translation and 2mm on rotation).
> I though that MoCo could do a better job...
That's really strange. Stupid question: you get those extreme values
within a run or between runs?
Hope this helps
Best
Remi
|