JISCMail - FSL Archives

I'm surprised by a couple of things here. One is that your differences seem huge (32 kHz!) as we normally only have a few hundred Hz difference. Or is that just the extreme range and most values are nothing like this?

The second thing is your dwell time of 0.032 seconds (or 32 ms). Is this correct? I am much more used to seeing values around 0.5 ms, so again this seems orders of magnitude different from normal.

Everything else seems sensible.

You might want to double check that you do not have a non-zero mean in your fieldmap too, as that will just add a global translation that you probably don't want.

From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of "Caspar M. Schwiedrzik" <[log in to unmask]>
Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
Date: Friday, 6 November 2015 21:31
To: "[log in to unmask]" <[log in to unmask]>
Subject: [FSL] FUGUE with Bruker 7T data

Hi!

I am trying to undistort some EPI data that we acquired on a Bruker 7T using FUGUE.

The main problem that I have is that the output I get from FUGUE seems scrambled.

Specifically, if I do not use a mask, I get only noise. If I do use a mask, I get only zeros where the brain should be, and a nice black rectangle of zeros in all but the coronal view.

The fieldmap I have is apparently a map of differences in Hz, along with a magnitude map.

The range of differences is from about -32000 to about +32000, everything in float format.

I first multiply the fieldmap by 6.28 to get to radians per sec.

I then align the epi and the magnitude map and apply the alignment to the scaled phase difference map.

The resulting alignment looks good.

Finally, I run fugue with the following flags:

-v

-i epi.nii

--loadfmap=fieldmap.nii.gz

--unwarpdir=y

-u epi_unwarp.nii.gz

--mask=epi_mask.nii

--dwell=0.032

Is there any obvious mistake in these steps?

Has anyone successfully undistorted Bruker EPI (or DTI) data using FUGUE?

Thanks, Caspar