Re: [Freesurfer] Using fieldmap to correct EPI distortions

Hi Doug,

So I am using a command like this:

epidewarp.fsl --mag ~/MonkeyfMRI/oz_fparc/fieldmap/003/fid_mag.nii --dph ~/MonkeyfMRI/oz_fparc/fieldmap/004/fid_phasediff.nii --exf ~/MonkeyfMRI/oz_fparc/bold/template.nii.gz --exfdw ~/MonkeyfMRI/oz_fparc/bold/b0dc.nii.gz --vsm vsm.nii.gz --tediff 2.46 --esp 0.31 --sigma 2 --perev

(note that instead of --o b0dc.nii.gz, I wrote --exfdw b0dc.nii.gz because this command apparently does not have --o option)

Before running the above command, I have registered the mag and phasediff volumes to the template volume and I have generated new volumes (fid_mag and fid_phasediff) that are already in register with the template.

I have two steps remaining in preproc-sess, which are register-sess and rawfunc2surf-sess. Should I use both -b0dc and -vsm flags for these two commands? I assume that when I use the above flag(s), my EPI fmc volumes will be dewarped first, and then they will get registered to the anatomical and the surface, right? So I hope that when I view/check the result of registration with tkregister2 (for possible manual interventions), I will see the dewarped template ...?

Finally what is the effect of specifying -b0dc in mkanalysis-sess?

Thanks, Reza

Hi,

I have a basic question, but the answer can be useful for others as well.

I am doing a customized preprocessing of monkey functional data in FS-FAST v5.3. After motion correction and before functional-anatomical registration, I want to use fieldmap data to correct spatial distortions of EPI data. So:

My fieldmap data: dicom files from a fieldmap scan (there are actually two series of dicom files for that scan)

My EPI data: fmc.nii.gz files in run folders

Now I have three questions:

1. How can I unpack the fieldmap data? An example command ...

I think you get two series out, one magnitude and one phase, just convert those to nifti

2. How can I use FS-FAST's epidewarp.fsl command for my dataset? Again an

example command would be very useful.

something like epidewarp.fsl --mag mag.nii --dph phase.nii --exf template.nii --o b0dc.nii ... You have to add --tediff and --esp, but those depend on you sequence Put the output (b0dc.nii) into session/bold (or session/rest, or where ever your fmri data are located)

3. How will the undistorted volumes be used later on in the FS-FAST

analysis stream?

When you run preproc-sess and mkanalysis-sess add -b0dc to the command line

Your help would be very much appreciated.

Thanks, Reza