usage: __main__.py [-h] [--eddy] [--duty]
[--no_b0_start NO_B0_START | --b0_every B0_EVERY]
[--b0_end] [--b0_value B0_VALUE] [--b0_philips]
(--bvals bvals [bvals ...] | --b_lin_max B_LIN_MAX | --q_lin_max Q_LIN_MAX)
(--fsl | --mrtrix) [-v [{DEBUG,INFO,WARNING}]] [-f]
nb_samples_per_shell [nb_samples_per_shell ...]
out_basename
Generate multi-shell gradient sampling with various processing options. Helps
accelerate gradients, optimize duty cycle and avoid artefacts.
Multi-shell gradient sampling is generated as in [1]. The bvecs are then
flipped to maximize spread for eddy current correction, b0s are interleaved at
equal spacing and the non-b0 samples are finally shuffled to minimize the total
diffusion gradient amplitude over a few TR.
Formerly: scil_generate_gradient_sampling.py
positional arguments:
nb_samples_per_shell Number of samples on each non b0 shell.
If multishell, provide a number per shell.
out_basename Gradient sampling output basename (don't include extension).
Please add options --fsl and/or --mrtrix below.
options:
-h, --help show this help message and exit
--eddy If set, we apply eddy optimization.
B-vectors are flipped to be well spread without symmetry.
--duty If set, we apply duty cycle optimization.
B-vectors are shuffled to reduce consecutive colinearity in the samples. [False]
-v [{DEBUG,INFO,WARNING}]
Produces verbose output depending on the provided level.
Default level is warning, default when using -v is info.
-f Force overwriting of the output files.
b0 acquisitions:
Default if you add no option is to have a b0 at the start.
--no_b0_start NO_B0_START
If set, do not add a b0 at the beginning.
--b0_every B0_EVERY Interleave a b0 every n=b0_every values. Starts after the first b0
(cannot be used with --no_b0_start). Must be an integer >= 1.
--b0_end If set, adds a b0 as last sample.
--b0_value B0_VALUE b-value of the b0s. [0.0]
--b0_philips If set, replaces values of b0s bvecs by existing bvecs for Philips handling.
Non-b0 acquisitions:
--bvals bvals [bvals ...]
bval of each non-b0 shell.
--b_lin_max B_LIN_MAX
b-max for linear bval distribution in *b*.
--q_lin_max Q_LIN_MAX
b-max for linear bval distribution in *q*;
the square root of b-values will be linearly distributed..
Save as:
--fsl Save in FSL format (.bvec/.bval).
--mrtrix Save in MRtrix format (.b).
References: [1] Emmanuel Caruyer, Christophe Lenglet, Guillermo Sapiro,
Rachid Deriche. Design of multishell gradient sampling with uniform coverage
in diffusion MRI. Magnetic Resonance in Medicine, Wiley, 2013, 69 (6),
pp. 1534-1540. <http://dx.doi.org/10.1002/mrm.24736>