 |
 |
Dear Claire, I'm trying to run eddy with slice-to-volume correction, but for a few subjects, there is an issue with the corrected DWI- some voxels have been cut out of some of the superior slices (I've attached a screenshot). This didn't happen when I ran eddy without slice-to-volume correction. This is the eddy command I ran that caused the issue: eddy_cuda9.1 --imain=dwi_unringed.nii.gz --mask=brain_mask.nii.gz --index=index.txt --acqp=acqparams.txt --bvecs=dwi.bvec --bvals=dwi.bval --topup=field --repol --out=dwi_eddy_out --mporder=8 --slspec=slspec.txt --s2v_niter=10 --s2v_lambda=1 --s2v_interp=trilinear I'm using a single band acquisition and I created the slspec.txt from a json file from dcm2niix. What might be causing the problem? Is there anything I can change in the eddy command to fix it? I don’t think that is is necessarily something wrong per se. Missing data like that means that there is at least one volume in which that part of the brain wasn’t acquired (because the subject moved that part out of the FOV). Whenever data is missing for a voxel in any of the volumes, eddy will mask that voxel out in all volumes. You see the problem in the top slices, and it is to be expected that if any part of the brain is going to move out of the FOV it will be the top or bottom slices. But with single band data and slice-to-vol there is the added complication that there is very little information in a slice that only has a 100 or so brain voxels. It is therefore very difficult to work out exactly where it goes, so the precision/reliability of the estimates decreases as you go to the top or bottom slices where there are only a few intracerebral voxels. So it is not impossible that the “missing data” isn’t actually missing, but rather just appears to be missing because eddy hasn’t been able to work out exactly where is should go. Unfortunately this is a genuine problem with single-band and slice-to-volume (it is much easier with multi-band, and the higher the multi-band factor the easier). You can test to increase the --s2v_lambda to for example 10. That means that you put a stronger prior on movement being “smooth” over time, which decreases the risk of eddy getting the end slices completely wrong but which also gives a little less flexibility in putting the other slices right in the case of very sudden movement. Jesper Many thanks, Claire Claire Kelly Research Assistant Victorian Infant Brain Studies (VIBeS) Murdoch Children's Research Institute The Royal Children's Hospital, 50 Flemington Road Parkville, Victoria 3052 Australia This e-mail and any attachments to it (the "Communication") are, unless otherwise stated, confidential, may contain copyright material and is for the use only of the intended recipient. If you receive the Communication in error, please notify the sender immediately by return e-mail, delete the Communication and the return e-mail, and do not read, copy, retransmit or otherwise deal with it. Any views expressed in the Communication are those of the individual sender only, unless expressly stated to be those of Murdoch Children’s Research Institute (MCRI) ABN 21 006 566 972 or any of its related entities. MCRI does not accept liability in connection with the integrity of or errors in the Communication, computer virus, data corruption, interference or delay arising from or in respect of the Communication. ________________________________ To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 <eddy_slice-to-vol-error.png> ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1