It occurs to be now that the FSL internal left-handed coordinate system does not apply exclusively to bvec data or to diffusion models fit from those data. I think it may be used for spatial transforms also. So referring to that "reference frame" as "bvec" might not be optimal. It might need to be simply referred to as "the FSL reference frame".
(Have not had the chance to look to see if this has any existing handling elsewhere in the BIDS ecosystem)
It occurs to be now that the FSL internal left-handed coordinate system does not apply exclusively to
bvecdata or to diffusion models fit from those data. I think it may be used for spatial transforms also. So referring to that "reference frame" as "bvec" might not be optimal. It might need to be simply referred to as "the FSL reference frame".(Have not had the chance to look to see if this has any existing handling elsewhere in the BIDS ecosystem)