Automatic image reconstruction for ODIN sequences

For image reconstruction of ODIN raw data, execute the program 'odinreco' within the directory of the file 'recoInfo' (the scan directory), or use the option '-r' to specify the location of 'recoInfo'. For example, the commands

odinreco -r /path/to/raw/data/recoInfo -o  /path/to/raw/data/image

and

cd /path/to/raw/data/recoInfo
odinreco

will do the same, i.e. create an image using the raw data in '/path/to/raw/data/'.

For other formats, e.g. the ISMRM Raw Data Format (ISMRMRD) or Siemens raw data, specify the raw data file using the option '-r':

odinreco -r simple_gre.h5

Full description:

odinreco: Automatic reconstruction for ODIN sequences

odinreco   provides   a   generic  reconstruction  for  imaging  sequences
(spin-warp,  EPI,  RARE,  ...).  It  retrieves  the  information about the
acquired  signal from the file 'recoInfo' which is stored on disk together
with  the  raw  data.  The  recoInfo  file  contains information about the
k-space  position  of  each acquisition windows (ADC). As output, odinreco
generates  a bunch of files with different image formats (depending on the
option  -f  and  third-party libraries available to ODIN). The filename of
each file starts with the given output prefix. 
Usage: odinreco [ options ] -r <input-file-to-be-read>
Options:
	-bcw: Correlation weighting of all blades against the mean of all blades (in image space)
	-bf: Filter function for blade phase correction (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=NoFilter)
	-bfw: Width of the filter function used for blade phase correction.[0.0 (0%) ... 1.0 (100%) of blade width; -1.0 (100%) ... 0.0 (0%) of center width] (default=1.0)
	-bkf: Blade-keyhole filter function (and its arguments) (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=CosSq)
	-bkt: Blade-keyhole transition width (fraction of oversampled k-space center) (default=0.20)
	-bwf: Window/Filter function for kspace weighting of each blade (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=Triangle)
	-bww: Window/Filter width for kspace weighting of each blade (default=1.0)
	-cp: Use direct Fourier transform with conjugate-phase correction instead of gridding
	-dce: TE used for field drift calculations, if non-zero [ms] (default=0.0ms)
	-dcs: Index selection string in recipe used for drift correction (space separated list of 'dim=selection', e.g. te=2)
	-dcw: Store field drift to file with this prefix
	-dd: Disable field-drift correction
	-df: Output format (file extension) of dump files (default=nii)
	-dg: Disable GRAPPA interpolation
	-dh: Disable homodyne reconstruction
	-dm: Disable fieldmap-based multi-frequency reconstruction
	-dp: Prefix of dump files (default=./dump)
	-ds: Disable slice-time correction
	-dt: Disable template-based phase correction
	-el: Level relative to mean value of first echo for setting mask of exponential decay fit (default=0.50)
	-f: Space sepapared list of output formats (file extensions), possible formats are '3db analyze asc coi dat dcm double float gz hdr idx ima interfile jdx mag mhd nii
ph png pos pro reg s16bit s32bit s8bit smp u16bit u32bit u8bit vtk xml xpro' (default=nii)
	-ff: k-space filter function (and its arguments) (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=NoFilter)
	-fm: Load fieldmap data from this file
	-fp: Filter plateau width (0=no plateau, 1=full plateau) (default=0.0)
	-gd: Fraction of k-psace points to discard in auto-calibration lines because of high residuals (default=0.0)
	-gp: Number of neigbouring measured k-space lines used for GRAPPA interpolation (default=2)
	-gr: Number of neigbours in read direction used for GRAPPA interpolation (default=5)
	-gs: Scale Factor for the grid used for gridding (default=1.0)
	-gt: Store GRAPPA template images
	-ip: Command line args for extra processing of final images (default=none)
	-j: Number of concurrent jobs(threads) in reco (default=1)
	-kg1d: Kernel function for 1d k-space gridding (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=Gauss(0.36169))
	-kg2d: Kernel function for 2d k-space gridding (options=Gauss NoFilter Triangle Hann Hamming CosSq Blackman BlackmanNuttall Exp , default=Gauss(0.36169))
	-kh: Keyhole reconstruction, e.g. for PROPELLER
	-ks: Store k-space data
	-kw1d: Kernel width for 1d k-space gridding [Pixel] (default=4.24264Pixel)
	-kw2d: Kernel width for 2d k-space gridding [Pixel] (default=3.0Pixel)
	-mf: Maximum number of frequencies for multi-frequency reconstruction (default=256)
	-mn: In sliding-window maximum-intensity projection, use this number of neighbors in each direction (default=4)
	-mr: Maximum recursion of reconstruction when waiting for blackboard data, zero means no recursion (default=100)
	-o: Prefix of output files (default=./image)
	-pc: Combine phase of channels using a phase correction
	-pe: Exponent for magnitude weighting when calculating combined phase of all channels (default=0.0)
	-pf: File name of FIFO (named pipe) to obtain the raw data from
	-pop: Recipe for post processing of 3D volume after channel combining (default=null)
	-prp: Recipe for post processing of 3D volume before channel combining (default=null)
	-pt: Combine and store phase timecourse from all channels
	-pz: Pad missing raw data (of interrupted measurement) with zeroes
	-qs: Quality control: Check for spikes in data
	-rd: Dump signal data and its protocol to files with this prefix, do not reconstruct images
	-rf: Read reco recipe (pipeline layout) from this file
	-rp: Reco recipe (pipeline layout)
	-s: Select certain dimensions ranges (space separated list of 'dim=selection', e.g. slice=3)
	-sc: Store images from channels separately instead of combining them
	-sp: Store phase in addition to magnitude
	-sw: Sliding window reconstruction, e.g. for Spiral/PROPELLER
	-t1f: Use this type of T1 fit, can be one of 'simple', 'magnitude' (default=simple)
	-t1l: Level relative to mean value of first echo for setting mask of T1 fit (default=0.50)
	-t1m: Upper limit for T1 fit [ms] (default=3000.0ms)
	-to: Timeout for reco in seconds, 0=inifinite (default=0)
	-type: Image representation type (options=automatic float double s32bit u32bit s16bit u16bit s8bit u8bit , default=automatic)
	-wdialect: Write data using given dialect of the format. (default is no dialect)
	-wp: Store the protocol separately to this file.
	-zp: Minimum size of final image in phase direction obtained zero filling (default=0)
	-zr: Minimum size of final image in read direction obtained zero filling (default=0)
	-zs: Minimum size of final image in slice direction obtained zero filling (default=0)
	-v <loglevel> or <component:loglevel> for debugging/tracing all components or a single component, respectively. Possible values for loglevel are: 0(noLog), 1(errorLog), 2(warningLog), 3(infoLog).
	-h, --help, -help, --version : Print help text or version information