ANTs:
A retrospective
Brian (PENN) and
Nick (UVA)
This talk is online at http://stnava.github.io/ANTs2015/ with colored links meant to be clicked for more information.
Retrospective
“Contingency.
AFNI is one result of unexpected events.
I met a woman and fell in love.”
ANTs–begotten in SyN
Founding developers
\(+\) neurodebian, slicer, brainsfit, nipype, itk and more …
ANTs lineage
Image mapping & perception: 1877-8
Francis Galton: Can we see criminality in the face?
Speaking of criminality…
Can we say anything about the U.S. Congress?
Maybe he should have used ANTs?
Image mapping & biology: 1917
D’’Arcy Thompson: Comparison of related forms
Image mapping & biology: Current
Look where we are now!
ANTs family tree
Initial scope
Diffeomorphisms:plausible physical modeling
“differentiable map with differentiable inverse”
Fine-grained and flexible maps
Diffeomorphisms: image parameterization in a metric space
General purpose library for multivariate image registration, segmentation & statistical analysis tools
170,000+ lines of C++, 6\(+\) years of work, 15+ collaborators.
Generic mathematical methods that are tunable for application specific domains.
no free lunch: An algorithm must use prior knowledge about a problem to do well on that problem.
Still quite a bit to do
On documentation
Developers can be blind to doc deficiencies while users are blind to what we provide!
Competitions
Competitions
Klein 2009
- 14 algorithms
- publicly available data
- different labeling protocols
- developers tune their own algorithms
Results
“One of the most significant findings of this study is that the relative performances of the registration methods under comparison appear to be little affected by the choice of subject population, labeling protocol, and type of overlap measure…. ART, SyN, IRTK, and SPM’s DARTEL Toolbox gave the best results according to overlap and distance measures, with ART and SyN delivering the most consistently high accuracy across subjects and label sets.”
EMPIRE 2010
- Alignment of lung boundaries,
- major fissures,
- annotated landmark pairs, and
- topology of displacement field.
Results
Multi-Atlas 2012
Results
SATA 2013
BRATS 2013
Results
Optimal symmetric multimodal templates and concatenated random forests for supervised brain tumor segmentation (simplified) with ANTsR, Neuroinformatics.
STACOM 2014
Major papers
“Papers are just advertisements for the science.”
Symmetric Normalization: (SyN)
\(\int_{t=0}^{0.5} \left(\|\mathbf{v}_1(x,t)\|_L^2 + \|\mathbf{v}_2(x,t)\|_L^2\right)dt\)
\(+ \|I\left(\phi_1(x,0.5)\right) - J_i\left(\phi_2(x,0.5)\right)\|^2\)
Beyond original SyN
antsRegistration
$ antsRegistration --help
COMMAND:
antsRegistration
This program is a user-level registration application meant to utilize
ITKv4-only classes. The user can specify any number of "stages" where a stage
consists of a transform; an image metric; and iterations, shrink factors, and
smoothing sigmas for each level.Note that dimensionality, metric, transform,
output, convergence, shrink-factors and smoothing-sigmas parameters are
mandatory.
OPTIONS:
--version
Get Version Information.
-d, --dimensionality 2/3
This option forces the image to be treated as a specified-dimensional image. If
not specified, we try to infer the dimensionality from the input image.
-o, --output outputTransformPrefix
[outputTransformPrefix,<outputWarpedImage>,<outputInverseWarpedImage>]
Specify the output transform prefix (output format is .nii.gz ). Optionally, one
can choose to warp the moving image to the fixed space and, if the inverse
transform exists, one can also output the warped fixed image. Note that only the
images specified in the first metric call are warped. Use antsApplyTransforms to
warp other images using the resultant transform(s).
-j, --save-state saveSateAsTransform
Specify the output file for the current state of the registration. The state
file is written to an hdf5 composite file. It is specially usefull if we want to
save the current state of a SyN registration to the disk, so we can load and
restore that later to continue the next registration process directly started
from the last saved state. The output file of this flag is the same as the
write-composite-transform, unless the last transform is a SyN transform. In that
case, the inverse displacement field of the SyN transform is also added to the
output composite transform. Again notice that this file cannot be treated as a
transform, and restore-state option must be used to load the written file by
this flag.
-k, --restore-state restoreStateAsATransform
Specify the initial state of the registration which get immediately used to
directly initialize the registration process. The flag is mutually exclusive
with other intialization flags.If this flag is used, none of the
initial-moving-transform and initial-fixed-transform cannot be used.
-a, --write-composite-transform 1/(0)
Boolean specifying whether or not the composite transform (and its inverse, if
it exists) should be written to an hdf5 composite file. This is false by default
so that only the transform for each stage is written to file.
<VALUES>: 0
-p, --print-similarity-measure-interval <unsignedIntegerValue>
Prints out the CC similarity metric measure between the full-size input fixed
and the transformed moving images at each iteration a value of 0 (the default)
indicates that the full scale computation should not take placeany value greater
than 0 represents the interval of full scale metric computation.
<VALUES>: 0
-v, --write-interval-volumes <unsignedIntegerValue>
Writes out the output volume at each iteration. It helps to present the
registration process as a short movie a value of 0 (the default) indicates that
this option should not take placeany value greater than 0 represents the
interval between the iterations which outputs are written to the disk.
<VALUES>: 0
-z, --collapse-output-transforms (1)/0
Collapse output transforms. Specifically, enabling this option combines all
adjacent transforms wherepossible. All adjacent linear transforms are written to
disk in the forman itk affine transform (called xxxGenericAffine.mat).
Similarly, all adjacent displacement field transforms are combined when written
to disk (e.g. xxxWarp.nii.gz and xxxInverseWarp.nii.gz (if available)).Also, an
output composite transform including the collapsed transforms is written to the
disk (called outputCollapsed(Inverse)Composite).
<VALUES>: 1
-i, --initialize-transforms-per-stage (1)/0
Initialize linear transforms from the previous stage. By enabling this option,
the current linear stage transform is directly intialized from the previous
stage's linear transform; this allows multiple linear stages to be run where
each stage directly updates the estimated linear transform from the previous
stage. (e.g. Translation -> Rigid -> Affine).
<VALUES>: 0
-n, --interpolation Linear
NearestNeighbor
MultiLabel[<sigma=imageSpacing>,<alpha=4.0>]
Gaussian[<sigma=imageSpacing>,<alpha=1.0>]
BSpline[<order=3>]
CosineWindowedSinc
WelchWindowedSinc
HammingWindowedSinc
LanczosWindowedSinc
Several interpolation options are available in ITK. These have all been made
available. Currently the interpolator choice is only used to warp (and possibly
inverse warp) the final output image(s).
-g, --restrict-deformation PxQxR
This option allows the user to restrict the optimization of the displacement
field, translation, rigid or affine transform on a per-component basis. For
example, if one wants to limit the deformation or rotation of 3-D volume to the
first two dimensions, this is possible by specifying a weight vector of '1x1x0'
for a deformation field or '1x1x0x1x1x0' for a rigid
transformation.Low-dimensional restriction only works if there are no preceding
transformations.
-q, --initial-fixed-transform initialTransform
[initialTransform,<useInverse>]
[fixedImage,movingImage,initializationFeature]
Specify the initial fixed transform(s) which get immediately incorporated into
the composite transform. The order of the transforms is stack-esque in that the
last transform specified on the command line is the first to be applied. In
addition to initialization with ITK transforms, the user can perform an initial
translation alignment by specifying the fixed and moving images and selecting an
initialization feature. These features include using the geometric center of the
images (=0), the image intensities (=1), or the origin of the images (=2).
-r, --initial-moving-transform initialTransform
[initialTransform,<useInverse>]
[fixedImage,movingImage,initializationFeature]
Specify the initial moving transform(s) which get immediately incorporated into
the composite transform. The order of the transforms is stack-esque in that the
last transform specified on the command line is the first to be applied. In
addition to initialization with ITK transforms, the user can perform an initial
translation alignment by specifying the fixed and moving images and selecting an
initialization feature. These features include using the geometric center of the
images (=0), the image intensities (=1), or the origin of the images (=2).
-m, --metric CC[fixedImage,movingImage,metricWeight,radius,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
MI[fixedImage,movingImage,metricWeight,numberOfBins,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
Mattes[fixedImage,movingImage,metricWeight,numberOfBins,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
MeanSquares[fixedImage,movingImage,metricWeight,radius=NA,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
Demons[fixedImage,movingImage,metricWeight,radius=NA,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
GC[fixedImage,movingImage,metricWeight,radius=NA,<samplingStrategy={None,Regular,Random}>,<samplingPercentage=[0,1]>]
ICP[fixedPointSet,movingPointSet,metricWeight,<samplingPercentage=[0,1]>,<boundaryPointsOnly=0>]
PSE[fixedPointSet,movingPointSet,metricWeight,<samplingPercentage=[0,1]>,<boundaryPointsOnly=0>,<pointSetSigma=1>,<kNeighborhood=50>]
JHCT[fixedPointSet,movingPointSet,metricWeight,<samplingPercentage=[0,1]>,<boundaryPointsOnly=0>,<pointSetSigma=1>,<kNeighborhood=50>,<alpha=1.1>,<useAnisotropicCovariances=1>]
IGDM[fixedImage,movingImage,metricWeight,fixedMask,movingMask,<neighborhoodRadius=0x0>,<intensitySigma=0>,<distanceSigma=0>,<kNeighborhood=1>,<gradientSigma=1>]
These image metrics are available--- CC: ANTS neighborhood cross correlation,
MI: Mutual information, Demons: (Thirion), MeanSquares, and GC: Global
Correlation. The "metricWeight" variable is used to modulate the per stage
weighting of the metrics. The metrics can also employ a sampling strategy
defined by a sampling percentage. The sampling strategy defaults to 'None' (aka
a dense sampling of one sample per voxel), otherwise it defines a point set over
which to optimize the metric. The point set can be on a regular lattice or a
random lattice of points slightly perturbed to minimize aliasing artifacts.
samplingPercentage defines the fraction of points to select from the domain. In
addition, three point set metrics are available: Euclidean (ICP), Point-set
expectation (PSE), and Jensen-Havrda-Charvet-Tsallis (JHCT).
-t, --transform Rigid[gradientStep]
Affine[gradientStep]
CompositeAffine[gradientStep]
Similarity[gradientStep]
Translation[gradientStep]
BSpline[gradientStep,meshSizeAtBaseLevel]
GaussianDisplacementField[gradientStep,updateFieldVarianceInVoxelSpace,totalFieldVarianceInVoxelSpace]
BSplineDisplacementField[gradientStep,updateFieldMeshSizeAtBaseLevel,totalFieldMeshSizeAtBaseLevel,<splineOrder=3>]
TimeVaryingVelocityField[gradientStep,numberOfTimeIndices,updateFieldVarianceInVoxelSpace,updateFieldTimeVariance,totalFieldVarianceInVoxelSpace,totalFieldTimeVariance]
TimeVaryingBSplineVelocityField[gradientStep,velocityFieldMeshSize,<numberOfTimePointSamples=4>,<splineOrder=3>]
SyN[gradientStep,updateFieldVarianceInVoxelSpace,totalFieldVarianceInVoxelSpace]
BSplineSyN[gradientStep,updateFieldMeshSizeAtBaseLevel,totalFieldMeshSizeAtBaseLevel,<splineOrder=3>]
Exponential[gradientStep,updateFieldVarianceInVoxelSpace,velocityFieldVarianceInVoxelSpace,<numberOfIntegrationSteps>]
BSplineExponential[gradientStep,updateFieldMeshSizeAtBaseLevel,velocityFieldMeshSizeAtBaseLevel,<numberOfIntegrationSteps>,<splineOrder=3>]
Several transform options are available. The gradientStep or learningRate
characterizes the gradient descent optimization and is scaled appropriately for
each transform using the shift scales estimator. Subsequent parameters are
transform-specific and can be determined from the usage. For the B-spline
transforms one can also specify the smoothing in terms of spline distance (i.e.
knot spacing).
-c, --convergence MxNxO
[MxNxO,<convergenceThreshold=1e-6>,<convergenceWindowSize=10>]
Convergence is determined from the number of iterations per level and is
determined by fitting a line to the normalized energy profile of the last N
iterations (where N is specified by the window size) and determining the slope
which is then compared with the convergence threshold.
-s, --smoothing-sigmas MxNxO...
Specify the sigma of gaussian smoothing at each level. Units are given in terms
of voxels ('vox') or physical spacing ('mm'). Example usage is '4x2x1mm' and
'4x2x1vox' where no units implies voxel spacing.
-f, --shrink-factors MxNxO...
Specify the shrink factor for the virtual domain (typically the fixed image) at
each level.
-u, --use-histogram-matching
Histogram match the images before registration.
-l, --use-estimate-learning-rate-once
turn on the option that lets you estimate the learning rate step size only at
the beginning of each level. * useful as a second stage of fine-scale
registration.
-w, --winsorize-image-intensities [lowerQuantile,upperQuantile]
Winsorize data based on specified quantiles.
-x, --masks [fixedImageMask,movingImageMask]
Image masks to limit voxels considered by the metric.
--float
Use 'float' instead of 'double' for computations.
<VALUES>: 0
-v, --verbose (0)/1
Verbose output.
-h
Print the help menu (short version).
--help
Print the help menu. Will also print values used on the current command line
call.Template building
antsMultivariateTemplateConstruction2.sh
$ antsMultivariateTemplateConstruction2.sh
Usage:
antsMultivariateTemplateConstruction2.sh -d ImageDimension -o OutputPrefix <other options> <images>
Compulsory arguments (minimal command line requires SGE/PBS cluster, otherwise use -c and
-j options):
-d: ImageDimension: 2 or 3 (for 2 or 3 dimensional registration of single volume)
ImageDimension: 4 (for template generation of time-series data)
-o: OutputPrefix; A prefix that is prepended to all output files.
<images> List of images in the current directory, eg *_t1.nii.gz. Should be at the end
of the command. Optionally, one can specify a .csv or .txt file where each
line is the location of the input image. One can also specify more than
one file for each image for multi-modal template construction (e.g. t1 and t2).
For the multi-modal case, the templates will be consecutively numbered (e.g.
template0.nii.gz, template1.nii.gz, ...).
NB: All images to be added to the template should be in the same directory, and this
script should be invoked from that directory.
Optional arguments:
-a image statistic used to summarize images (default 1)
0 = mean
1 = mean of normalized intensities
2 = median
-b: Backup images and results from all iterations (default = 0): Boolean to save
the transform files, bias corrected, and warped images for each iteration.
-c: Control for parallel computation (default 0):
0 = run serially
1 = SGE qsub
2 = use PEXEC (localhost)
3 = Apple XGrid
4 = PBS qsub
-e use single precision ( default 1 )
-g: Gradient step size (default 0.25): smaller in magnitude results in more
cautious steps.
-i: Iteration limit (default 4): iterations of the template construction
(Iteration limit)*NumImages registrations.
-j: Number of cpu cores to use locally for pexec option (default 2; requires "-c 2")
-k: Number of modalities used to construct the template (default 1): For example,
if one wanted to create a multimodal template consisting of T1,T2,and FA
components ("-k 3").
-w: Modality weights used in the similarity metric (default = 1): specified as
e.g. 1x0.5x0.75.
-q: Max iterations for each pairwise registration (default = 100x100x70x20):
specified in the form ...xJxKxL where
J = max iterations at coarsest resolution (here, reduced by power of 2^2)
K = middle resolution iterations (here, reduced by power of 2)
L = fine resolution iteratioxns (here, full resolution).
Finer resolutions take much more time per iteration than coarser resolutions.
-f: Shrink factors (default = 6x4x2x1): Also in the same form as -q max iterations.
Needs to have the same number of components.
-s: Smoothing factors (default = 3x2x1x0): Also in the same form as -q max
iterations. Needs to have the same number of components.
-n: N4BiasFieldCorrection of moving image: 0 == off, 1 == on (default 1).
-p: Commands to prepend to job scripts (e.g., change into appropriate directory, set
paths, etc)
-r: Do rigid-body registration of inputs before creating template (default 0):
0 == off 1 == on. Only useful when you do not have an initial template
-l: Use linear image registration stages during the pairwise (template/subject)
deformable registration. Otherwise, registration is limited to SyN or
B-spline SyN (see '-t' option). This is '1' by default.
-m: Type of similarity metric used for registration (default = CC): Options are
CC = cross-correlation
MI = mutual information
MSQ = mean square difference
DEMONS = demon's metric
A similarity metric per modality can be specified. If the CC metric is chosen,
one can also specify the radius in brackets, e.g. '-m CC[4]'.
-t: Type of transformation model used for registration (default = SyN): Options are
SyN = Greedy SyN
BSplineSyN = Greedy B-spline SyN
TimeVaryingVelocityField = Time-varying velocity field
TimeVaryingBSplineVelocityField = Time-varying B-spline velocity field
-u: Walltime (default = 20:00:00): Option for PBS qsub specifying requested time
per pairwise registration.
-v: Memory limit (default = 8gb): Option for PBS qsub specifying requested memory
per pairwise registration.
-x: XGrid arguments (e.g., -x "-p password -h controlhost")
-y: Update the template with the full affine transform (default 1). If 0, the rigid
component of the affine transform will not be used to update the template. If your
template drifts in translation or orientation try -y 0.
-z: Use this this volume as the target of all inputs. When not used, the script
will create an unbiased starting point by averaging all inputs. Use the full
path.
For multiple modalities, specify -z modality1.nii.gz -z modality2.nii.gz ...
in the same modality order as the input images.
Example:
antsMultivariateTemplateConstruction2.sh -d 3 -i 3 -k 1 -f 4x2x1 -s 2x1x0vox -q 30x20x4 -t SyN -m CC -c 0 -o MY sub*avg.nii.gz
Multimodal example:
antsMultivariateTemplateConstruction2.sh -d 3 -i 3 -k 2 -f 4x2x1 -s 2x1x0vox -q 30x20x4 -t SyN -z t1.nii.gz -z t2.nii.gz -m CC -c 0 -o MY templateInput.csv
where templateInput.csv contains
subjectA_t1.nii.gz,subjectA_t2.nii.gz
subjectB_t1.nii.gz,subjectB_t2.nii.gz
...
--------------------------------------------------------------------------------------
ANTS was created by:
--------------------------------------------------------------------------------------
Brian B. Avants, Nick Tustison and Gang Song
Penn Image Computing And Science Laboratory
University of Pennsylvania
Please reference http://www.ncbi.nlm.nih.gov/pubmed/20851191 when employing this script
in your studies. A reproducible evaluation of ANTs similarity metric performance in
brain image registration:
* Avants BB, Tustison NJ, Song G, Cook PA, Klein A, Gee JC. Neuroimage, 2011.
Also see http://www.ncbi.nlm.nih.gov/pubmed/19818860 for more details.
The script has been updated and improved since this publication.
--------------------------------------------------------------------------------------
Script by Nick Tustison
--------------------------------------------------------------------------------------
Apple XGrid support by Craig Stark
--------------------------------------------------------------------------------------Nonparametric nonuniform intensity normalization (N3)
Developed at the Montreal Neurological Institute (John Sled, 1998)
Part of the standard preprocessing protocol in large scale projects such as ADNI
- The traditional de facto standard in MRI bias correction
- good performance
- public availability
Public availability — set of perl scripts coordinating various C++ programs
“Let’s incorporate N3 into ANTs!”
N3 adoption issues
N4
comparative evaluation
smaller spline distances (useful for higher magnet strengths)
multiresolution
weighted regional mask (used in
antsAtroposN4.sh)faster execution times
publicly available
N4BiasFieldCorrection
$ N4BiasFieldCorrection --help
COMMAND:
N4BiasFieldCorrection
N4 is a variant of the popular N3 (nonparameteric nonuniform normalization)
retrospective bias correction algorithm. Based on the assumption that the
corruption of the low frequency bias field can be modeled as a convolution of
the intensity histogram by a Gaussian, the basic algorithmic protocol is to
iterate between deconvolving the intensity histogram by a Gaussian, remapping
the intensities, and then spatially smoothing this result by a B-spline modeling
of the bias field itself. The modifications from and improvements obtained over
the original N3 algorithm are described in the following paper: N. Tustison et
al., N4ITK: Improved N3 Bias Correction, IEEE Transactions on Medical Imaging,
29(6):1310-1320, June 2010.
OPTIONS:
-d, --image-dimensionality 2/3/4
This option forces the image to be treated as a specified-dimensional image. If
not specified, N4 tries to infer the dimensionality from the input image.
-i, --input-image inputImageFilename
A scalar image is expected as input for bias correction. Since N4 log transforms
the intensities, negative values or values close to zero should be processed
prior to correction.
-x, --mask-image maskImageFilename
If a mask image is specified, the final bias correction is only performed in the
mask region. If a weight image is not specified, only intensity values inside
the masked region are used during the execution of the algorithm. If a weight
image is specified, only the non-zero weights are used in the execution of the
algorithm although the mask region defines where bias correction is performed in
the final output. Otherwise bias correction occurs over the entire image domain.
See also the option description for the weight image.
-r, --rescale-intensities 0/(1)
At each iteration, a new intensity mapping is calculated and applied but there
is nothing which constrains the new intensity range to be within certain values.
The result is that the range can "drift" from the original at each iteration.
This option rescales to the [min,max] range of the original image intensities
within the user-specified mask.
-w, --weight-image weightImageFilename
The weight image allows the user to perform a relative weighting of specific
voxels during the B-spline fitting. For example, some studies have shown that N3
performed on white matter segmentations improves performance. If one has a
spatial probability map of the white matter, one can use this map to weight the
b-spline fitting towards those voxels which are more probabilistically
classified as white matter. See also the option description for the mask image.
-s, --shrink-factor 1/2/3/4/...
Running N4 on large images can be time consuming. To lessen computation time,
the input image can be resampled. The shrink factor, specified as a single
integer, describes this resampling. Shrink factors <= 4 are commonly used.
-c, --convergence [<numberOfIterations=50x50x50x50>,<convergenceThreshold=0.0>]
Convergence is determined by calculating the coefficient of variation between
subsequent iterations. When this value is less than the specified threshold from
the previous iteration or the maximum number of iterations is exceeded the
program terminates. Multiple resolutions can be specified by using 'x' between
the number of iterations at each resolution, e.g. 100x50x50.
-b, --bspline-fitting [splineDistance,<splineOrder=3>]
[initialMeshResolution,<splineOrder=3>]
These options describe the b-spline fitting parameters. The initial b-spline
mesh at the coarsest resolution is specified either as the number of elements in
each dimension, e.g. 2x2x3 for 3-D images, or it can be specified as a single
scalar parameter which describes the isotropic sizing of the mesh elements. The
latter option is typically preferred. For each subsequent level, the spline
distance decreases in half, or equivalently, the number of mesh elements doubles
Cubic splines (order = 3) are typically used.
-t, --histogram-sharpening [<FWHM=0.15>,<wienerNoise=0.01>,<numberOfHistogramBins=200>]
These options describe the histogram sharpening parameters, i.e. the
deconvolution step parameters described in the original N3 algorithm. The
default values have been shown to work fairly well.
-o, --output correctedImage
[correctedImage,<biasField>]
The output consists of the bias corrected version of the input image.
Optionally, one can also output the estimated bias field.
--version
Get Version Information.
-v, --verbose (0)/1
Verbose output.
-h
Print the help menu (short version).
--help
Print the help menu.
<VALUES>: 1
“20+ years of development.
Show me the code!”
Brian, MICCAI workshop
Atropos
Algorithmic workflow
(Atropos + N4 \(\rightarrow\) antsAtroposN4.sh)
Atropos
$ Atropos --help
COMMAND:
Atropos
A finite mixture modeling (FMM) segmentation approach with possibilities for
specifying prior constraints. These prior constraints include the specification
of a prior label image, prior probability images (one for each class), and/or an
MRF prior to enforce spatial smoothing of the labels. Similar algorithms include
FAST and SPM. Reference: Avants BB, Tustison NJ, Wu J, Cook PA, Gee JC. An open
source multivariate framework for n-tissue segmentation with evaluation on
public data. Neuroinformatics. 2011 Dec;9(4):381-400.
OPTIONS:
-d, --image-dimensionality 2/3/4
This option forces the image to be treated as a specified-dimensional image. If
not specified, Atropos tries to infer the dimensionality from the first input
image.
-a, --intensity-image [intensityImage,<adaptiveSmoothingWeight>]
One or more scalar images is specified for segmentation using the
-a/--intensity-image option. For segmentation scenarios with no prior
information, the first scalar image encountered on the command line is used to
order labelings such that the class with the smallest intensity signature is
class '1' through class 'N' which represents the voxels with the largest
intensity values. The optional adaptive smoothing weight parameter is applicable
only when using prior label or probability images. This scalar parameter is to
be specified between [0,1] which smooths each labeled region separately and
modulates the intensity measurement at each voxel in each intensity image
between the original intensity and its smoothed counterpart. The smoothness
parameters are governed by the -b/--bspline option.
-b, --bspline [<numberOfLevels=6>,<initialMeshResolution=1x1x...>,<splineOrder=3>]
If the adaptive smoothing weights are > 0, the intensity images are smoothed in
calculating the likelihood values. This is to account for subtle intensity
differences across the same tissue regions.
-i, --initialization Random[numberOfClasses]
Otsu[numberOfTissueClasses]
KMeans[numberOfTissueClasses,<clusterCenters(in ascending order and for first intensity image only)>]
PriorProbabilityImages[numberOfTissueClasses,fileSeriesFormat(index=1 to numberOfClasses) or vectorImage,priorWeighting,<priorProbabilityThreshold>]
PriorLabelImage[numberOfTissueClasses,labelImage,priorWeighting]
To initialize the FMM parameters, one of the following options must be
specified. If one does not have prior label or probability images we recommend
using kmeans as it is typically faster than otsu and can be used with
multivariate initialization. However, since a Euclidean distance on the inter
cluster distances is used, one might have to appropriately scale the additional
input images. Random initialization is meant purely for intellectual curiosity.
The prior weighting (specified in the range [0,1]) is used to modulate the
calculation of the posterior probabilities between the likelihood*mrfprior and
the likelihood*mrfprior*prior. For specifying many prior probability images for
a multi-label segmentation, we offer a minimize usage option (see -m). With that
option one can specify a prior probability threshold in which only those pixels
exceeding that threshold are stored in memory.
-s, --partial-volume-label-set label1xlabel2xlabel3
The partial volume estimation option allows one to modelmixtures of classes
within single voxels. Atropos currently allows the user to model two class
mixtures per partial volume class. The user specifies a set of class labels per
partial volume class requested. For example, suppose the user was performing a
classic 3-tissue segmentation (csf, gm, wm) using kmeans initialization. Suppose
the user also wanted to model the partial voluming effects between csf/gm and
gm/wm. The user would specify it using -i kmeans[3] and -s 1x2 -s 2x3. So, for
this example, there would be 3 tissue classes and 2 partial volume classes.
Optionally,the user can limit partial volume handling to mrf considerations only
whereby the output would only be the three tissues.
--use-partial-volume-likelihoods 1/(0)
true/(false)
The user can specify whether or not to use the partial volume likelihoods, in
which case the partial volume class is considered separate from the tissue
classes. Alternatively, one can use the MRF only to handle partial volume in
which case, partial volume voxels are not considered as separate classes.
-p, --posterior-formulation Socrates[<useMixtureModelProportions=1>,<initialAnnealingTemperature=1>,<annealingRate=1>,<minimumTemperature=0.1>]
Plato[<useMixtureModelProportions=1>,<initialAnnealingTemperature=1>,<annealingRate=1>,<minimumTemperature=0.1>]
Aristotle[<useMixtureModelProportions=1>,<initialAnnealingTemperature=1>,<annealingRate=1>,<minimumTemperature=0.1>]
Sigmoid[<useMixtureModelProportions=1>,<initialAnnealingTemperature=1>,<annealingRate=1>,<minimumTemperature=0.1>]]
Different posterior probability formulations are possible as are different
update options. To guarantee theoretical convergence properties, a proper
formulation of the well-known iterated conditional modes (ICM) uses an
asynchronous update step modulated by a specified annealing temperature. If one
sets the AnnealingTemperature > 1 in the posterior formulation a traditional
code set for a proper ICM update will be created. Otherwise, a synchronous
update step will take place at each iteration. The annealing temperature, T,
converts the posteriorProbability to posteriorProbability^(1/T) over the course
of optimization.
-x, --mask-image maskImageFilename
The image mask (which is required) defines the region which is to be labeled by
the Atropos algorithm.
-c, --convergence numberOfIterations
[<numberOfIterations=5>,<convergenceThreshold=0.001>]
Convergence is determined by calculating the mean maximum posterior probability
over the region of interest at each iteration. When this value decreases or
increases less than the specified threshold from the previous iteration or the
maximum number of iterations is exceeded the program terminates.
-k, --likelihood-model Gaussian
HistogramParzenWindows[<sigma=1.0>,<numberOfBins=32>]
ManifoldParzenWindows[<pointSetSigma=1.0>,<evaluationKNeighborhood=50>,<CovarianceKNeighborhood=0>,<kernelSigma=0>]
JointShapeAndOrientationProbability[<shapeSigma=1.0>,<numberOfShapeBins=64>, <orientationSigma=1.0>, <numberOfOrientationBins=32>]
LogEuclideanGaussian
Both parametric and non-parametric options exist in Atropos. The Gaussian
parametric option is commonly used (e.g. SPM & FAST) where the mean and standard
deviation for the Gaussian of each class is calculated at each iteration. Other
groups use non-parametric approaches exemplified by option 2. We recommend using
options 1 or 2 as they are fairly standard and the default parameters work
adequately.
-m, --mrf [<smoothingFactor=0.3>,<radius=1x1x...>]
[<mrfCoefficientImage>,<radius=1x1x...>]
Markov random field (MRF) theory provides a general framework for enforcing
spatially contextual constraints on the segmentation solution. The default
smoothing factor of 0.3 provides a moderate amount of smoothing. Increasing this
number causes more smoothing whereas decreasing the number lessens the
smoothing. The radius parameter specifies the mrf neighborhood. Different update
schemes are possible but only the asynchronous updating has theoretical
convergence properties.
-g, --icm [<useAsynchronousUpdate=1>,<maximumNumberOfICMIterations=1>,<icmCodeImage=''>]
Asynchronous updating requires the construction of an ICM code image which is a
label image (with labels in the range {1,..,MaximumICMCode}) constructed such
that no MRF neighborhood has duplicate ICM code labels. Thus, to update the
voxel class labels we iterate through the code labels and, for each code label,
we iterate through the image and update the voxel class label that has the
corresponding ICM code label. One can print out the ICM code image by specifying
an ITK-compatible image filename.
-r, --use-random-seed 0/(1)
Initialize internal random number generator with a random seed. Otherwise,
initialize with a constant seed number.
-o, --output [classifiedImage,<posteriorProbabilityImageFileNameFormat>]
The output consists of a labeled image where each voxel in the masked region is
assigned a label from 1, 2, ..., N. Optionally, one can also output the
posterior probability images specified in the same format as the prior
probability images, e.g. posterior%02d.nii.gz (C-style file name formatting).
-u, --minimize-memory-usage (0)/1
By default, memory usage is not minimized, however, if this is needed, the
various probability and distance images are calculated on the fly instead of
being stored in memory at each iteration. Also, if prior probability images are
used, only the non-negligible pixel values are stored in memory.
<VALUES>: 0
-w, --winsorize-outliers BoxPlot[<lowerPercentile=0.25>,<upperPercentile=0.75>,<whiskerLength=1.5>]
GrubbsRosner[<significanceLevel=0.05>,<winsorizingLevel=0.10>]
To remove the effects of outliers in calculating the weighted mean and weighted
covariance, the user can opt to remove the outliers through the options
specified below.
-e, --use-euclidean-distance (0)/1
Given prior label or probability images, the labels are propagated throughout
the masked region so that every voxel in the mask is labeled. Propagation is
done by using a signed distance transform of the label. Alternatively,
propagation of the labels with the fast marching filter respects the distance
along the shape of the mask (e.g. the sinuous sulci and gyri of the cortex.
<VALUES>: 0
-l, --label-propagation whichLabel[lambda=0.0,<boundaryProbability=1.0>]
The propagation of each prior label can be controlled by the lambda and boundary
probability parameters. The latter parameter is the probability (in the range
[0,1]) of the label on the boundary which increases linearly to a maximum value
of 1.0 in the interior of the labeled region. The former parameter dictates the
exponential decay of probability propagation outside the labeled region from the
boundary probability, i.e. boundaryProbability*exp( -lambda * distance ).
-v, --verbose (0)/1
Verbose output.
-h
Print the help menu (short version).
--help
Print the help menu.
<VALUES>: 1ANTs and the perils of circularity
$ CreateDTICohort -h
COMMAND:
CreateDTICohort
OPTIONS:
-d, --image-dimensionality 2/3
-a, --dti-atlas inputDTIAtlasFileName
-x, --label-mask-image maskImageFileName
lowerThresholdFunction
-n, --noise-sigma <noiseSigma=18>
-p, --pathology label[<percentageChangeEig1=-0.05>,<percentageChangeAvgEig2andEig3=0.05>,<numberOfVoxels=all or percentageOfvoxels>]
-w, --dwi-parameters [B0Image,directionFile,bvalue]
[B0Image,schemeFile]
-r, --registered-population textFileWithFileNames.txt
-o, --output [outputDirectory,fileNameSeriesRootName,<numberOfControls=10>,<numberOfExperimentals=10>]
-h
--helpWhich one is not warped?
Answer
