ANTs Atlas Registration

This performs the registration of predefined atlases to a target image (fixed image) using the antsRegistration algorithm from the ANTs package. The base template from the atlas dataset will be registered to the fixed image provided by the user and the computed warping will be applied to transform all atlas derivative images (e.g. masks, priors) into the fixed image space.

Usage

Inputs

• fixed_image: NIfTI file to be used as the fixed image in antsRegistration.
• fixed_image_mask: Optional NIfTI mask for the fixed_image

Configuration

Gear config

• debug (boolean, default False): Include debug statements in output.
• atlas: (string, default mindboggle): Atlas to register to fixed_image. Currently only supporting mindboggle

ANTs config

Single valued options

• dimension: (int, default 3): Image dimension, 2 or 3.
• interpolation: (string, default Linear) One of Linear, NearestNeighbor, CosineWindowedSinc, WelchWindowedSinc. HammingWindowedSinc, LanczosWindowedSinc, BSpline, MultiLabel, Gaussian
• interpolation_parameters: (JSON string) Optional parameters for interpolation method. Ex. if Interpolation is Gaussian, you may set sigma and alpha with "[.1,1]" which can be parsed by a JSON parser.
• collapse_output_transforms: (boolean, default True) Collapse output transforms. Specifically, enabling this option combines all adjacent linear transforms and composes all adjacent displacement field transforms before writing the results to disk.
• initialize_transforms_per_stage: (boolean, default False) Initialize linear transforms from the previous stage. By enabling this option, the current linear stage transform is directly intialized from the previous stages 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)
• float: (boolean, default False) Use float instead of double for computations.
• output_transform_prefix: (string, default "transform") Output transform prefix, only if output_warped_image is True
• output_warped_image: (boolean, default True) Output warped moving image to the fixed space and
• output_inverse_warped_image: (boolean, default True) Output warped fixed image to the fixed space and
• winsorize_upper_quantile: (number, default 0.995) Winsorize data based on specified upper quantile
• winsorize_lower_quantile: (number, default 0.005) Winsorize data based on specified lower quantile
• num_threads: (integer, default 1) Number of ITK threads to use.
• args: (string, default None) Additional arguments

Stage specific options

Each of the stage specific options must be a JSON string that parses into a list of the same length as the number of transforms. Each entry in this list is the value for a single stage.

• metric: (string, default ["MI","MI","CC"]) List of items which are CC or MeanSquares or Demons or GC or MI or Mattes. The metric(s) to use for each stage. Enclose a list of the previous values in square brackets, to use multiple metrics for a single stage i.e. ["MI","CC",["MI,"CC"]]
• metric_weight: (string, default [1,1,1]) The metric weight(s) for each stage (float, weights must sum to 1 per stage). Shape must match the config value metric
• radius_or_number_of_bins: (string, default [32,32,32]) The number of bins in each stage for the MI and Mattes metric, the radius for other metrics (integer). Shape must match the config value metric
• sampling_strategy: (string, default ["Regular","Regular","Regular"]) The metric sample strategy for each stage (one of '', 'Regular', or 'Random'). Shape must match the config value metric
• sampling_percentage: (string, default [0.25,0.25,0.25]) The metric sampling percentage(s) to use for each stage (float, 0 <= val <= 1). Shape must match the config value metric
• use_estimate_learning_rate_once: (string, default [false,false,false]) (JSON string) Estimate the learning rate step size only at the beginning of each level (boolean). Must match number of stages.
• use_histogram_matching: (string, default [false,false,false]) Histogram match the images before registration (boolean). Must match number of stages.
• interpolation: (string, default Linear) One of Linear, NearestNeighbor, CosineWindowedSinc, WelchWindowedSinc. HammingWindowedSinc, LanczosWindowedSinc, BSpline, MultiLabel, Gaussian
• interpolation_parameters: (string, default None) Parameters for interpolation method. (json string)
• transforms: (string, default ["Rigid","Affine","SyN"]) A list of items which are one of Rigid, Affine, CompositeAffine, Similarity, Translation, BSpline, GaussianDisplacementField, TimeVaryingVelocityField, TimeVaryingBSplineVelocityField, SyN, BSplineSyN, Exponential, or BSplineExponential. Must be the same size as number of Stages.
• transform_parameters: (string, default [0.1,0.1,[0.1,3.0,0.0]]) Parameters for transform at each stage. Must be same size as number of stages.
• restrict_deformation: (string, default None) A list of lists, one for each stage. At each stage items are 0.0 <= a floating point number <= 1.0. 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 1,1,0 for a deformation field or 1,1,0,1,1,0 for a rigid transformation. Low-dimensional restriction only works if there are no preceding transformations.
• number_of_iterations: (string, default [[1000,500,250,100],[1000,500,250,100],[100,100,70,50,20]]) List of lists of integers for each stage, corresponds to MxNxO in the antsRegistration documentation: 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.
• smoothing_sigmas: (string, default [[4,3,2,1],[4,3,2,1],[5,3,2,1,0]]) (JSON string) List of lists of floats for each stage. Specify the sigma of gaussian smoothing at each level.
• sigma_units: (string, default ["vox","vox","vox"]) Corresponding units of the smoothing sigmas, either mm or vox
• shrink_factors: (string, default [[12,8,4,2],[12,8,4,2],[10,6,4,2,1]]) (JSON string) List of list of integers for each stage. Specify the shrink factor for the virtual domain (typically the fixed image) at each level.
• convergence_threshold: (string, default ["1e-6","1e-6","1e-6"]) List of string valued convergence thresholds, one for each stage.
• convergence_window_size: (string, default [10,10,10]) List of integer window sizes, one for each stage.

Contributing

For more information about how to get started contributing to this gear, checkout CONTRIBUTING.md.