At least some subset of HCP data, as well as others I'm sure, report the b=0 scans as truly being b=5 (normal b is often on the scale of 1000), but as much downstream processing depends on the b=0 locations this naturally breaks stuff since they can't be found. Will write a case to handle this in the initial parse b module and spit out a b values file that happies fsl/camino.
At least some subset of HCP data, as well as others I'm sure, report the b=0 scans as truly being b=5 (normal b is often on the scale of 1000), but as much downstream processing depends on the b=0 locations this naturally breaks stuff since they can't be found. Will write a case to handle this in the initial parse b module and spit out a b values file that happies fsl/camino.