MRIReco converts the raw into acq which currently handles :
- the slice dimension but assumed that the trajectory is the same for all slices
- Contrast
- Repetition
- Channel
- x/y/z (multi-slab are not possible)
It could be great to rethink all the system in order to handles all the standard dimension of MRD (At first, maybe not user::NTuple{8,UInt16} = ntuple(i->UInt16(0),8))
average::UInt16 = 0
slice::UInt16 = 0
contrast::UInt16 = 0
phase::UInt16 = 0
repetition::UInt16 = 0
set::UInt16 = 0
segment::UInt16 = 0
All dimension with an independent trajectory. It will really be helpful for pulseq / KomaMRI simulation where nothing forces the trajectory to be the same along the slices.
In the end, not having to worry about the type of dimension should also simplify the package. Each dimension will be associated to the corresponding kdate, traj and subsampling
Does one of you have an idea how to do that efficiently in term of code ? https://github.com/MagneticResonanceImaging/MRIReco.jl/blob/c12696b2f86c2787e1c34fb5ec4d6e32f8b9e64b/MRIBase/src/Datatypes/RawAcqData.jl#L303C1-L304C118
MRIReco converts the raw into acq which currently handles :
It could be great to rethink all the system in order to handles all the standard dimension of MRD (At first, maybe not
user::NTuple{8,UInt16} = ntuple(i->UInt16(0),8))All dimension with an independent trajectory. It will really be helpful for pulseq / KomaMRI simulation where nothing forces the trajectory to be the same along the slices.
In the end, not having to worry about the type of dimension should also simplify the package. Each dimension will be associated to the corresponding kdate, traj and subsampling
Does one of you have an idea how to do that efficiently in term of code ? https://github.com/MagneticResonanceImaging/MRIReco.jl/blob/c12696b2f86c2787e1c34fb5ec4d6e32f8b9e64b/MRIBase/src/Datatypes/RawAcqData.jl#L303C1-L304C118