Dear Cristina, > Siemens 3T Prisma, multiband acquisition, 64-channel head coil > > 12 directions, 2 B0s (A>>P), b= 500 s/mm2 > 12 directions, 2 B0s (P>>A), b= 500 s/mm2 > > 29 directions, 3 B0s (A>>P), b=1500 s/mm2 > 29 directions, 3 B0s (P>>A), b=1500 s/mm2 > > 64 directions, 4 B0s (A>>P), b=3000 s/mm2 > 64 directions, 4 B0s (A>>P), b=3000 s/mm2 > > I hope to run eddy first and then popup and applytopup. > you should run topup first, followed by eddy. There is no need to run applytopup if you run eddy. You run eddy on the raw data, feeding only information about the field and subject movement from topup into eddy. > For eddy, my initial question is whether I should combine all runs of my diffusion data (including the B0s) and then run eddy, or if eddy should be run on each run separately or each pair of runs. If the subject has remained in the scanner throughout, and if there have been no re-shimmings, I would concatenate all data and run eddy once. > > Similarly, should topup be run on concatenated B0s for all runs or should it be conducted on each pair of runs? Under the same assumptions as above I would pick a (or possibly two) pair of A->P and P->A volumes from the first scans (I assume that the above reflects the order in which they were acquired) and run topup on that. > > In addition, to for the my_acq_param.txt file, I would like to confirm that the number of rows should correspond to the number of B0s and tat for any Siemens scanner, the first three values should be 0 +/-1 0 as noted in the popup faq. The number of rows in the my_acq_param.txt that you pass to topup should be the same number as the number of volumes you feed into topup. And yes, the first three should be 0 +/-1 0, and the fourth you can either jump through the hoops to calculate, or you can set it to 0.05. Good luck Jesper > > Thank you in advance for your guidance! > Cristina