Dear Jesper,
Thank you very much for the reply. I made sure I didn't run applytopup between topup and eddy and the same parameter given to --out for topup was fed in to the --topup for eddy. It seems the --imain parameter of eddy only accepts one file name (unlike applytopup
where --imain can take a list of files), so I concatenated the two diffusion data sets (total volume is 2*N) with opposite PE, and also the bvecs, bvals like you suggested (bvecs is 3 by 2*N, bvals is 1 by 2*N). The indices in index.txt is N times of 1 followed
by N times of 3, corresponding to the 1st and 3rd line in acqparams.txt (there are two volumes of b0 image from both PE). If I run it like this in terminal, I get the following error message:
eddy: msg=These data do not support least-squares resampling
Am I missing something in the parameters? I am running FSL 5.0.8.
Regards,
Hua
From: "Jesper Andersson" <
[log in to unmask]>
To: [log in to unmask]
Sent: Tuesday, March 10, 2015 6:09:42 AM
Subject: Re: [FSL] How to use --resamp option in Eddy
Dear Hua,
I want to use topup + eddy for susceptibility and eddy current correction. In the help info of eddy, it says there's an optional argument --resamp which selects the resampling method between jac and lsr. I wanted to use the lsr option, but apparently in
this case the input images need to be organized in a different way. What is the correct command line call to run eddy with the --resamp=lsr option? (assuming we already have the result from topup)
first of all, you should NOT have resampled your images with the topup results (i.e. not run applytopup) prior to running eddy on the data. Just run topup and then feed the name you gave as a parameter to —out as input to —topup when running eddy.
You data should be arranged such that the bvals and bvecs are “repeated”, i.e. the first half of the bvecs should be identical to the second half. The second half of the data should be acquired with opposing PE compared to the first half.
Jesper
