That makes sense, thanks for your help.
On 07/10/16 13:54, Jesper Andersson wrote:
> Dear Ian,
>
>> I'm trying to use eddy without topup on some diffusion data with a single PE direction. I've noticed that for FSL 5.0.9 and the later patched eddy, the top and bottom slices of the output image are blank (=0, not even noise). This doesn't happen with older eddy. I've tried both a single shell and a different two shell dataset. For the single shell dataset the bottom-most slice did have a small polygon of signal in one corner, it also happens when --repol is given.
>>
>> Is this a bug or could it be cause by an issue like not all volumes having full coverage of the FOV?
> yes, it will be caused by not all subjects having information for those voxels. Any slice that has moved out of the FOV and that is subsequently re-sampled back into the FOV is treated as “unknown” inside eddy. At the very end any voxel that is “unknown” in one or more volumes will be set to zero. For “regular” subjects it is typical that the top and bottom slices are “missing” after eddy. For subjects that move a lot there may be several slices missing at both ends.
>
> Jesper
>
>> Thanks for your time,
>> Ian
|