Oct 22, 2018, 2:26:43 PM10/22/18
Hi ... processing some NODDI data acquired as back-to-back acquisitions, the first at b=2500 and the second at b=800, with several b=0 images at the beginning of each. I've done the usual steps of preprocessing w/ fsl 5 and eddy current correction etc. My question is what to do if the b=0 images from the first acquisition are different in magnitude from the b=0 images of the second acquisition. They will never be exactly the same, but due to differences in patient motion between acquisitions etc., sometimes they are within 1-2%, but other times as much as 15% different in signal levels. Is some kind of normalization recommended, or is this already accounted for in some way, or does it not really matter?