Dear Liz,
> Hi. I am trying to use topup on some diffusion-weighted images of a rat spinal cord. The results are great except I am losing some of the slices! The original data has 12 slices (all including the spinal cord, no blank ones on either side), but after running topup only some of the slices in the middle have image in them, and a few on either end are blank. I want to compare different subjects and this is a problem because on each subject a different number of slices are preserved. I am trying to find a way to use topup on this data and preserve all 12 slices every time.
topup contains also a model for subject movement, so the reason some slices disappear is because topup detects/thinks that the subject has moved so that there are a slice/slices at the top of one scan that is not present in the other, and vice verse for the bottom. If you then combine the images using —method=lsr (for applytopup) it will only give you the slices that are common to both volmes (i.e. there will be slices missing at top and bottom).
One way around this is to use applytopup with —method=jac and run it separately for the two different blip directions. You will then end up with one set of volumes that have data missing only at the top and one set that have data missing only at the bottom. These could in principle then be combined such that you take the average of all voxels that are non-zero in both volumes and the only non-zero one in the others.
Just remember that for those slices with data only for one volume the distortion correction will be less accurate.
Good luck Jesper
>
> So far I have tried padding my images with "empty" slices on either end first (to give more space for error), but that doesn't work. I would like to see if there is a way to apply topup 2-dimensionally, on each slice separately. I used fslsplit but realized topup won't each slice individually because 1 is an odd number in the slice dimension. So I combined the slices into pairs of two adjacent and tried running topup on these, but all I get is an "Abort trap: 6"
>
> Is there any way that I haven't thought of to get topup to preserve all 12 slices? Thank you.
|