Make sure scaling is the same with/without nav to compute metrics

[jcohenadad]

Notably for ghosting calculation #7

[po09i]

I found this great ressource on dcm2niix. We want the images to be proportional to the MR signal. I believe we want to use the FP values for that to be the case.

[po09i]

Dcm2niix mentions that the data is scaled to FP by default. The BIDS tag UsePhilipsFloatNotDisplayScaling will be 1 if the is the case. I checked all acquired data up until now (subject 23) and they are all using FP.