Run a registration of a series of 2d images.
mia-2dmyomilles -i <in-file> -o <out-file> [options]
mia-2dmyomilles This program is use to run a modified version of the ICA based registration approach described in Milles et al. 'Fully Automated Motion Correction in First-Pass Myocardial Perfusion MR Image Sequences', Trans. Med. Imaging., 27(11), 1611-1621, 2008. Changes include the extraction of the quasi-periodic movement in free breathingly acquired data sets and the option to run affine or rigid registration instead of the optimization of translations only.
input perfusion data set
output perfusion data set
file name base for registered files
save synthetic reference images to this file base
save cropped image set to this file
save the features images resulting from the ICA and some intermediate images used for the RV-LV segmentation with the given file name base to PNG files. Also save the coefficients of the initial best and the final IC mixing matrix.
verbosity of output, print messages of given level and higher priorities. Supported priorities starting at lowest level are:
info \(hy Low level messages
trace \(hy Function call trace
fail \(hy Report test failures
warning \(hy Warnings
error \(hy Report errors
debug \(hy Debug output
message \(hy Normal messages
fatal \(hy Report only fatal errors
print copyright information
print this help
print a short help
print the version number and exit
ICA components 0 = automatic estimation
normalized ICs
don't strip the mean from the mixing curves
use initial guess for myocardial perfusion
segment and scale the crop box around the LV (0=no segmentation)
skip images at the beginning of the series as they are of other modalities
maximum number of iterations in ICA
Segmentation method
delta-peak \(hy difference of the peak enhancement images
features \(hy feature images
delta-feature \(hy difference of the feature images
Maxiumum number of threads to use for processing,This number should be lower or equal to the number of logical processor cores in the machine. (-1: automatic estimation).
registration criterion
Optimizer used for minimization For supported plugins see PLUGINS:minimizer/singlecost
transformation type For supported plugins see PLUGINS:2dimage/transform
multi-resolution levels
Global reference all image should be aligned to. If set to a non-negative value, the images will be aligned to this references, and the cropped output image date will be injected into the original images. Leave at -1 if you don't care. In this case all images with be registered to a mean position of the movement
registration passes
mirror
Spline interpolation boundary conditions that mirror on the boundary
(no parameters)
repeat
Spline interpolation boundary conditions that repeats the value at the boundary
(no parameters)
zero
Spline interpolation boundary conditions that assumes zero for values outside
(no parameters)
bspline
B-spline kernel creation , supported parameters are:
d = 3 (int)
Spline degree. in [0, 5]
omoms
OMoms-spline kernel creation, supported parameters are:
d = 3 (int)
Spline degree. in [3, 3]
affine
Affine transformation (six degrees of freedom)., supported parameters are:
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
rigid
Rigid transformations (i.e. rotation and translation, three degrees of freedom)., supported parameters are:
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
rot-center = [[0,0]] (streamable)
Relative rotation center, i.e. <0.5,0.5> corresponds to the center of the support rectangle.
rotation
Rotation transformations (i.e. rotation about a given center, one degree of freedom)., supported parameters are:
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
rot-center = [[0,0]] (streamable)
Relative rotation center, i.e. <0.5,0.5> corresponds to the center of the support rectangle.
spline
Free-form transformation that can be described by a set of B-spline coefficients and an underlying B-spline kernel., supported parameters are:
anisorate = [[0,0]] (2dfvector)
anisotropic coefficient rate in pixels, nonpositive values will be overwritten by the 'rate' value..
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
kernel = [bspline:d=3] (factory)
transformation spline kernel.. For supported plug-ins see PLUGINS:1d/splinekernel
penalty = (factory)
Transformation penalty term. For supported plug-ins see PLUGINS:2dtransform/splinepenalty
rate = 10 (float)
isotropic coefficient rate in pixels. in [1, 3.40282e+38]
translate
Translation only (two degrees of freedom), supported parameters are:
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
vf
This plug-in implements a transformation that defines a translation for each point of the grid defining the domain of the transformation., supported parameters are:
imgboundary = mirror (factory)
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc
imgkernel = [bspline:d=3] (factory)
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel
divcurl
divcurl penalty on the transformation, supported parameters are:
curl = 1 (float)
penalty weight on curl. in [0, 3.40282e+38]
div = 1 (float)
penalty weight on divergence. in [0, 3.40282e+38]
norm = 0 (bool)
Set to 1 if the penalty should be normalized with respect to the image size.
weight = 1 (float)
weight of penalty energy. in [0, 3.40282e+38]
gdas
Gradient descent with automatic step size correction., supported parameters are:
ftolr = 0 (double)
Stop if the relative change of the criterion is below.. in [0, INF]
max-step = 2 (double)
Minimal absolute step size. in [1, INF]
maxiter = 200 (uint)
Stopping criterion: the maximum number of iterations. in [1, 2147483647]
min-step = 0.1 (double)
Maximal absolute step size. in [1e-10, INF]
xtola = 0.01 (double)
Stop if the inf-norm of the change applied to x is below this value.. in [0, INF]
gdsq
Gradient descent with quadratic step estimation, supported parameters are:
ftolr = 0 (double)
Stop if the relative change of the criterion is below.. in [0, INF]
gtola = 0 (double)
Stop if the inf-norm of the gradient is below this value.. in [0, INF]
maxiter = 100 (uint)
Stopping criterion: the maximum number of iterations. in [1, 2147483647]
scale = 2 (double)
Fallback fixed step size scaling. in [1, INF]
step = 0.1 (double)
Initial step size. in [0, INF]
xtola = 0 (double)
Stop if the inf-norm of x-update is below this value.. in [0, INF]
gsl
optimizer plugin based on the multimin optimizers ofthe GNU Scientific Library (GSL) https://www.gnu.org/software/gsl/, supported parameters are:
eps = 0.01 (double)
gradient based optimizers: stop when |grad| < eps, simplex: stop when simplex size < eps.. in [1e-10, 10]
iter = 100 (int)
maximum number of iterations. in [1, 2147483647]
opt = gd (dict)
Specific optimizer to be used.. Supported values are:
bfgs \(hy Broyden-Fletcher-Goldfarb-Shann
bfgs2 \(hy Broyden-Fletcher-Goldfarb-Shann (most efficient version)
cg-fr \(hy Flecher-Reeves conjugate gradient algorithm
gd \(hy Gradient descent.
simplex \(hy Simplex algorithm of Nelder and Mead
cg-pr \(hy Polak-Ribiere conjugate gradient algorithm
step = 0.001 (double)
initial step size. in [0, 10]
tol = 0.1 (double)
some tolerance parameter. in [0.001, 10]
nlopt
Minimizer algorithms using the NLOPT library, for a description of the optimizers please see 'http://ab-initio.mit.edu/wiki/index.php/NLopt_Algorithms', supported parameters are:
ftola = 0 (double)
Stopping criterion: the absolute change of the objective value is below this value. in [0, INF]
ftolr = 0 (double)
Stopping criterion: the relative change of the objective value is below this value. in [0, INF]
higher = inf (double)
Higher boundary (equal for all parameters). in [INF, INF]
local-opt = none (dict)
local minimization algorithm that may be required for the main minimization algorithm.. Supported values are:
gn-orig-direct-l \(hy Dividing Rectangles (original implementation, locally biased)
gn-direct-l-noscal \(hy Dividing Rectangles (unscaled, locally biased)
gn-isres \(hy Improved Stochastic Ranking Evolution Strategy
ld-tnewton \(hy Truncated Newton
gn-direct-l-rand \(hy Dividing Rectangles (locally biased, randomized)
ln-newuoa \(hy Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximation
gn-direct-l-rand-noscale \(hy Dividing Rectangles (unscaled, locally biased, randomized)
gn-orig-direct \(hy Dividing Rectangles (original implementation)
ld-tnewton-precond \(hy Preconditioned Truncated Newton
ld-tnewton-restart \(hy Truncated Newton with steepest-descent restarting
gn-direct \(hy Dividing Rectangles
ln-neldermead \(hy Nelder-Mead simplex algorithm
ln-cobyla \(hy Constrained Optimization BY Linear Approximation
gn-crs2-lm \(hy Controlled Random Search with Local Mutation
ld-var2 \(hy Shifted Limited-Memory Variable-Metric, Rank 2
ld-var1 \(hy Shifted Limited-Memory Variable-Metric, Rank 1
ld-mma \(hy Method of Moving Asymptotes
ld-lbfgs-nocedal \(hy None
ld-lbfgs \(hy Low-storage BFGS
gn-direct-l \(hy Dividing Rectangles (locally biased)
none \(hy don't specify algorithm
ln-bobyqa \(hy Derivative-free Bound-constrained Optimization
ln-sbplx \(hy Subplex variant of Nelder-Mead
ln-newuoa-bound \(hy Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximation
ln-praxis \(hy Gradient-free Local Optimization via the Principal-Axis Method
gn-direct-noscal \(hy Dividing Rectangles (unscaled)
ld-tnewton-precond-restart \(hy Preconditioned Truncated Newton with steepest-descent restarting
lower = -inf (double)
Lower boundary (equal for all parameters). in [INF, INF]
maxiter = 100 (int)
Stopping criterion: the maximum number of iterations. in [1, 2147483647]
opt = ld-lbfgs (dict)
main minimization algorithm. Supported values are:
gn-orig-direct-l \(hy Dividing Rectangles (original implementation, locally biased)
g-mlsl-lds \(hy Multi-Level Single-Linkage (low-discrepancy-sequence, require local gradient based optimization and bounds)
gn-direct-l-noscal \(hy Dividing Rectangles (unscaled, locally biased)
gn-isres \(hy Improved Stochastic Ranking Evolution Strategy
ld-tnewton \(hy Truncated Newton
gn-direct-l-rand \(hy Dividing Rectangles (locally biased, randomized)
ln-newuoa \(hy Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximation
gn-direct-l-rand-noscale \(hy Dividing Rectangles (unscaled, locally biased, randomized)
gn-orig-direct \(hy Dividing Rectangles (original implementation)
ld-tnewton-precond \(hy Preconditioned Truncated Newton
ld-tnewton-restart \(hy Truncated Newton with steepest-descent restarting
gn-direct \(hy Dividing Rectangles
auglag-eq \(hy Augmented Lagrangian algorithm with equality constraints only
ln-neldermead \(hy Nelder-Mead simplex algorithm
ln-cobyla \(hy Constrained Optimization BY Linear Approximation
gn-crs2-lm \(hy Controlled Random Search with Local Mutation
ld-var2 \(hy Shifted Limited-Memory Variable-Metric, Rank 2
ld-var1 \(hy Shifted Limited-Memory Variable-Metric, Rank 1
ld-mma \(hy Method of Moving Asymptotes
ld-lbfgs-nocedal \(hy None
g-mlsl \(hy Multi-Level Single-Linkage (require local optimization and bounds)
ld-lbfgs \(hy Low-storage BFGS
gn-direct-l \(hy Dividing Rectangles (locally biased)
ln-bobyqa \(hy Derivative-free Bound-constrained Optimization
ln-sbplx \(hy Subplex variant of Nelder-Mead
ln-newuoa-bound \(hy Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximation
auglag \(hy Augmented Lagrangian algorithm
ln-praxis \(hy Gradient-free Local Optimization via the Principal-Axis Method
gn-direct-noscal \(hy Dividing Rectangles (unscaled)
ld-tnewton-precond-restart \(hy Preconditioned Truncated Newton with steepest-descent restarting
ld-slsqp \(hy Sequential Least-Squares Quadratic Programming
step = 0 (double)
Initial step size for gradient free methods. in [0, INF]
stop = -inf (double)
Stopping criterion: function value falls below this value. in [INF, INF]
xtola = 0 (double)
Stopping criterion: the absolute change of all x-values is below this value. in [0, INF]
xtolr = 0 (double)
Stopping criterion: the relative change of all x-values is below this value. in [0, INF]
Register the perfusion series given in 'segment.set' by using automatic ICA estimation. Skip two images at the beginning and otherwiese use the default parameters. Store the result in 'registered.set'.
mia-2dmyomilles -i segment.set -o registered.set -k 2
Gert Wollny
This software is Copyright (c) 1999\(hy2013 Leipzig, Germany and Madrid, Spain. It comes with ABSOLUTELY NO WARRANTY and you may redistribute it under the terms of the GNU GENERAL PUBLIC LICENSE Version 3 (or later). For more information run the program with the option '--copyright'.