 |
 |
Hi, I notice that the bad case you showed had images that were all brain extracted whereas the good case showed non-brain for some of the crucial images (such as the fieldmap to structural registration). Is it possible that you’ve not provided both the non-brain-extracted and brain-extracted images in some cases? Or maybe the file names aren’t set correctly in some cases according to the rule of only differing by “_brain” at the end. This naming is indicated on the FSL Wiki in the following section from the FEAT User Guide (under Pre-Stats): "FEAT will also look for the original non-brain-extracted fieldmap magnitude in the same directory as the brain-extracted image that you input into the GUI, this non-brain-extracted magnitude image should have the same filename as the brain-extracted version, with the only difference being the "_brain" at the end of the brain-extracted version. For example, fieldmap_mag and fieldmap_mag_brain, but not something like fieldmap_mag and fieldmap_mag_brain_v1, as the latter ends with "_brain_v1" and not just "_brain" and so this would cause a problem.” So I would carefully check that all the necessary brain-extracted and non-brain-extracted inputs were correctly named and specified. That is my best guess as to what is going wrong. All the best, Mark From: Andreas Bartsch <[log in to unmask]<mailto:[log in to unmask]>> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]<mailto:[log in to unmask]>> Date: Tuesday, 21 April 2015 19:48 To: "[log in to unmask]<mailto:[log in to unmask]>" <[log in to unmask]<mailto:[log in to unmask]>> Subject: Re: [FSL] B0 unwarping misalignment Hi, you need to get the PE right. Are you sure that this is not, for example, LR in these cases? There is a bug / feature in the patient protocol copy that may cause this. Cheers, Andreas Am 21.04.15 11:48 schrieb "Inês Violante" unter <[log in to unmask]<mailto:[log in to unmask]>>: Dear Andreas, In order to solve the misalignment issue we tried the following: - fslreorient2std both the T1 anatomical image and the gradient field map images - Try to run Melodic with each phase encoding direction to see whether one of it will solve the issue None of this helped. Any ideas? Thank you very much, ines On Mon, Apr 20, 2015 at 9:11 PM, Andreas Bartsch <[log in to unmask]<mailto:[log in to unmask]>> wrote: Well, at least in the bad case the distortion correction actually seems to have introduced additional distortions in AP - from what I can tell in the low res jpeg. So something is definitely not right here. Cheers, Andreas Am 20.04.15 18:21 schrieb "Inês Violante" unter <[log in to unmask]<mailto:[log in to unmask]>>: Dear Andreas, When you mentioned a bad z-shift do you think that the phase enconding direction for these subjects could be different? Thanks ines On Mon, Apr 20, 2015 at 5:14 PM, Li, Lucia M <[log in to unmask]<mailto:[log in to unmask]>> wrote: Dear Andreas, Thank you for replying. Yes, we use the BBR option in the registration step. Lucia ________________________________________ From: FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] on behalf of Andreas Bartsch [[log in to unmask]<mailto:[log in to unmask]>] Sent: Monday, April 20, 2015 4:36 PM To: [log in to unmask]<mailto:[log in to unmask]> Subject: Re: [FSL] B0 unwarping misalignment Hi, this looks like bad z-shift. Do you use FSL's latest version, i.e. BBR? Cheers, Andreas Am 20.04.15 15:31 schrieb "Lucia M Li" unter <[log in to unmask]<mailto:[log in to unmask]>>: Dear FSL users & experts, We were hoping to get some help on an issue we've had with our unwarping step (run as part of running a single-session melodic). We noticed that for a couple of people in our study, the "fieldmap to highres" registration looks seriously displaced - see link: https://dl.dropboxusercontent.com/u/22466029/FieldmapAlignment/Missali gn me nt.jpg The majority of our scans have registered very well however - see link: https://dl.dropboxusercontent.com/u/22466029/FieldmapAlignment/GoodAli gn me nt.png We were wondering if anyone could advise on what might have caused this displacement in a few people only (scanned using the same scanner/ parameters, fieldmaps prepared using the same parameters & unwarped with the same parameters)? We were also wondering what effect this has on the final data (the final func to highres looks ok) and what we can do to correct it? Many thanks in advance for your help. Kind regards, Lucia