Dear Jeff,
> hi jesper and saad, i'm developing a new dti protocol and it occurred to me that one strategy to minimize scan time and maximize dti angular resolution
> is instead of averaging the results of the same scan repeated once topup and once topdown but to split the number directions in half
> and do half of the directions topup and the other half topdown. have you done this? and does it work?
from the perspective of eddy being able to estimate the eddy currents and movements it should work fine. Part of the point with how eddy does the estimation is precisely that one shouldn't be forced to collect all directions blip-up and blip-down if one is strapped for time/directions.
There will obviously be a point where you have "sufficient" angular sampling and subsequent directions pretty much just improves your SNR. And in those cases you might as well collect them blip-up and blip-down.
Jesper
|