Hi Matt… Juan is working with me at Columbia to get the topups working with the HCP pipeline. The echo spacing is the same in both scans, as is the geometry we believe, though we still need to confirm that. The warpfield images have no information in them that I can tell, as changing the scaling of the display doesn't reveal anything. Also the shimming should be the same for all scans. The spin echo and gradient echo images look like they have the same distortions after they've been rigidly realigned, so we're a little puzzled as to why the topup procedure is failing. Any help would be appreciated. Thanks!
_________________________ gaurav patel gauravpa...@gmail.com www.neurofreak.net > On May 22, 2015, at 3:00 PM, Glasser, Matthew <glass...@wusm.wustl.edu> wrote: > > It helps to reply to the previous e-mail chain so that all of the info on a > question is in one place. > > Are you sure that those warp fields don’t have data in them, just a bad > scaling of the display? I don’t think the issue is with the topup > configuration (we’ve never had to change that, despite using it on scanners > with different field strengths). > > It would be good to check on what I suggested in the previous response. > While it is undoubtably better to use a proper SBRef generated from the MB > sequence itself, it should still be possible to do what you are doing. It is > important that using a rigid registration of the MB to the SBRef results in a > very precise alignment (the images should have exactly the same distortion). > > Another question, do you know that your scanner used the same shim for all > the scans? > > Peace, > > Matt. > > From: Juan Sanchez <sanch...@nyspi.columbia.edu> > Date: Friday, May 22, 2015 at 8:56 AM > To: "hcp-users@humanconnectome.org" <hcp-users@humanconnectome.org> > Subject: [HCP-Users] SBref Distortion Correction > > > Let us clarify what we are doing exactly here. We are using spin-echo images > with alternating phase-encoding directions in alternating runs (Topup PA and > Topup AP, in our case). We also decided to use the corresponding Topup (PA > or AP) as SBRef instead of an actual single-band reference image because we > had issues obtaining the single-band image and because the tissue contrast in > the spin-echo/Topup is pretty good (we tried using as SBRef the first frame > of the multiband EPI but the spin-echo/Topup definitely works better, > probably because of its improved tissue contrast). Even though using the > Topups instead of a B0 map for distortion correction and using the Topup as > SBRef instead of the first frame of the multiband EPI result in clear > improvements, the resulting undistorted SBRef is not perfectly aligned to the > anatomical images. For this reason, we think there's a problem with the > procedures creating or applying the WarpField (the topup procedure or the > applyward procedure that uses the WarpField to undistort the SBRef). The > undistorted SBRef in an example subject from the HCP dataset looks noticeably > better aligned to the anatomical image than ours. Furthermore, the > WarpField.nii in that example subject looks very similar to the corresponding > TopupField.nii (like a scaled version of it). In our data, in contrast, the > TopupField.nii looks probably like it's supposed to look like (with some > anatomical information) but the WarpField.nii is totally empty in some cases! > After realizing this is the case, we think that our (main) problem may be in > the topup procedure (or maybe the input images it uses or some later > modification to the WarpField.nii). Could you give us any suggestions as to > how to fix this issue? Maybe we need to adjust some of the configuration > files that the topup procedure uses? Did you have to tweak any parameters to > optimize this for your scanner? > > We're attaching one screencapture with the TopupField.nii and the > WarpField.nii, another image showing the T2 anatomical scan, and a third > image showing the SBRef (Topup in our case). > _______________________________________________ > HCP-Users mailing list > HCP-Users@humanconnectome.org > http://lists.humanconnectome.org/mailman/listinfo/hcp-users > > > > The materials in this message are private and may contain Protected > Healthcare Information or other information of a sensitive nature. If you are > not the intended recipient, be advised that any unauthorized use, disclosure, > copying or the taking of any action in reliance on the contents of this > information is strictly prohibited. If you have received this email in error, > please immediately notify the sender via telephone or return mail. > > _______________________________________________ > HCP-Users mailing list > HCP-Users@humanconnectome.org > http://lists.humanconnectome.org/mailman/listinfo/hcp-users _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
