v0.6.0

MRIReco v0.6.0

Diff since v0.5.0

Closed issues:

• Offsets Bruker acquisition is not handle by the current reconstruction pipeline (#58)
• Strip of IO Part in Dedicated Package (#66)
• data from multiple dicom series into one (#85)
• Bruker pulse structure is not read correctly (#86)
• recoData(f::BrukerFile) only read Int16 format and procno 1 (#87)
• reading any bruker parameters takes 0.5 ms (#94)
• Complex Float32 data are not well handle (#103)
• KooshballTrajectory does not work for Float32 (#106)
• sdc on NFFTOp failed due to p.params.precompute = TENSOR (#107)
• Extremely slow multiCoil reconstruction for 32 coils (#108)
• Error when using CGNR (#110)

Merged pull requests:

• Reco Bruker 3DUTE initial implementation (#79) (@aTrotier)
• Change Function Scope and dims() -> ndims() (#80) (@alexjaffray)
• Multiple improvements for BrukerFile (#88) (@aTrotier)
• add offsets for bruker sequence (#90) (@aTrotier)
• issue with NFFTTools 0.2.4 -> force compat to 0.2.3 (#95) (@aTrotier)
• Read Bruker method file only the first time (#96) (@aTrotier)
• Ci use the subpackages version in the repository (#98) (@aTrotier)
• don't run setupIterativeReco when direct reco (#101) (@aTrotier)
• reco of repetitions along dims=6 (#105) (@aTrotier)
• WIP: espirit speedup (#111) (@tknopp)
• Update functions to handle 6-dims + add tests (#113) (@aTrotier)
• Add optional argument to ESPIRiT to output sensitivity maps of arbitrary size (#115) (@alexjaffray)
• OffsetBruker for 2D acquisition (#116) (@aTrotier)
• reshape hack so that ESPIRIT is compatible with the MKL backend of FFTW (#118) (@JakobAsslaender)