Dear Xufei,
> Dear FSL experts,
> Could you help me with the use of topup? My concern is that, take b value=2000 s/mm2 for example, my raw data has two phase encoding directions and each phase encoding direction contains 66 diffusion directions. But, after dealing with topup for correction, the final data opened, the number of diffusion directions is 66, the half of 132. But i hope a better setting
> is to acquire only two b0 with two phase directions and apply the correction to 132 directions acquires at different orientation. So it will be no waste of the angular resolution.
the fact that applytopup has merged pairs of directions imply that you have two copies of 66 directions, i.e. your angular resolution is determined by 66 directions. So you haven’t actually lost any resolution. Instead you have potentially gained some spatial resolution in distorted areas because of the way that applytopup combines the pairs.
If you prefer to have your 132 volumes you can add --method=jac to your applytopup command.
Having said all that, applytopup is sort of superseded by eddy these days. I recommend taking a look at the eddy page and instead of running applytopup you feed the output file from topup into eddy along with your raw diffusion data. That way your data will also be corrected for eddy currents and subject movement (in addition to susceptibility).
Jesper
> As i am a beginner of topup, could you help me with modifying the commands?
> And my topup commands were like these:
> fslroi AP2K b0_AP2K 0 1
> fslroi PA2K b0_PA2K 0 1
> fslmerge -t both_b0 b0_AP2K b0_PA2K
> topup --imain=both_b0 --datain=my_acqparam.txt --config=b02b0.cnf --out=my_topup_results
> applytopup --imain=AP2K,PA2K --inindex=1,2 --datain=my_acqparam.txt --topup=my_topup_results --out=my_hifi_images
> Thank you very much!
|