Open cookpa opened 6 months ago
The signal falls off pretty sharply with b-value so if we are interested in computing ADC, it might be wise to get some measurements at b=500, I'd probably replace the b=1000 shell with b=500
Actually, the b=1000 shell might be OK - it's faint but still distinguishable from background.
If there's interest, could look at other QC stats:
ref distortion (make sure the first two b=0 have no in-plane distortion), currently we only output (ref distortion) / (eddy distortion)
b=0 signal drift - might be tricky because the test protocol will be quite short, but could plot number anyway for tracking
Phantom ADC
Others?