From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Jesper Andersson
Sent: Wednesday, October 05, 2016 5:41 AM
To: [log in to unmask]
Subject: Re: [FSL] topup and eddy multishell different TE

I’m trying to figure out how to preprocess a multi-shell diffusion dataset. For each shell, the diffusion weighted images (A>>P) and the b0 images (P>>A) were collected with the same echo time. Between shells, the echo time was not fixed (64, 76, 83, 94 ms respectively for each of the 4 shells).  But for all shells, the total EPI readout time was the same.

 

There’ll be different T2 weighting between different shells.  Is it possible to perform a joint eddy current correction with all shells?  If so, what’s the best way to carry out the TOPUP step, and is it possible to use the results as eddy input?

as long as there has been no re-shimming between the acquisition of the different shells, my experience is that this works well. I have done this many times on data acquired by the HCP during the piloting phase. The echo time doesn’t enter in to the models used by either topup or eddy.

Eduardo, it would be interesting to hear your experience and what it is that fails when trying to use all shells in the correction?

Qiuyun, there is no principal difference between this case and the case we describe on the EDDY help page. Run topup with your P->A b0 image and an A->P b0 image from one of the shells. I suggest using the first b0 from the shell that was acquired closest in time to the P->A b0. Make sure that the first volume in the 4D file you feed into topup is the A->P b0 image. Concatenate all the shells, doing this for data, bvals and bvecs alike. Make sure that the shell you took the b0 from for topup is the first shell in your concatenated data set. Then just run eddy in the usual way.

But please note that for your next step, the modelling of the diffusion signal, you need to address the problem of differing echo time for the different shells. Maybe Stam or Saad has some suggestions for how to do that?

Jesper

 

Any suggestions will be highly appreciated.  Thanks a lot in advance!

 

Best,

Qiuyun

 

 

The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.