CTSimU aRTist Module

This is a module for the radiographic simulator aRTist which reads and sets up the scenario from a CTSimU Scenario Description file. With the module, it is also possible to simulate the complete CT scan as described in the JSON scenario.

Requirements

• aRTist, version 2.12

Installation

1. Download the aRTist package file (CTSimU-<version>.artp) for the latest release.
2. Drag and drop the .artp file into your aRTist window to install the module.

Example scenarios

You can find example scenario files that you can load with this module on the Github repository for the CTSimU Scenario Descriptions.

Known limitations

• Pixel binning is not supported. Such an operation can be applied as a post-processing step using the CTSimU Toolbox.
• Continuous-motion scans are not supported, only stop&go mode. Drifts between the frames of an averaged projection are not handled. Therefore, motion blurring within one projection image cannot be simulated.
• Parallel beam geometries are not supported. As a workaround, a very high source-detector distance could be chosen in the JSON file.
• Detector: rear window and filters are not supported, only the front versions. If they are needed for a scatter simulation, they should be included in the scene as samples (fixed to the world coordinate system).
• If flat-field correction is turned on in the JSON file, aRTist's internal flat-field correction will be used. This means that the flat-field corrected image will be re-scaled to the mean intensity of the flat field image instead of the free-beam maximum, and an ideal (noise-free) image will be taken for aRTist's internal flat-field image. It is recommended to turn off flat-field correction in the JSON file and instead use the CTSimU Toolbox to run the correction after the simulation (using the simulated flat field and dark field images). A flat-field correction Python script will be generated by the module along with the simulation output files.

Graphical User Interface (GUI)

CT Setup

The module's GUI only shows a few simulation parameters that can be changed before a simulation run. The full parameters of the CT scan are loaded from a CTSimU scenario description JSON file and stored in the background, unseen by the user. To use the module, it is therefore always necessary to start by loading a CTSimU scenario file.

Loading a CTSimU scenario

1. Select a JSON file that contains a CTSimU scenario.
2. Press the button Load.

CT Projection settings

The terms projection and frame are used interchangeably by the module. Strictly speaking, a number of frames is averaged into a projection image, but since the module does not currently support motion blurring, they both mean the same here.

• Start angle: the rotation angle of the sample stage at the start of the CT scan. (This value is automatically imported from the scenario file.)
• Stop angle: the rotation angle of the sample stage at the end of the CT scan. (This value is automatically imported from the scenario file.)
• Projections: number of projection images to take for the CT scan simulation. In standard mode, the projection images are taken at equidistant angular positions between start angle and stop angle. However, in general, the JSON file can use drift parameters to alter the angular step size between projection images. Such drifts are not displayed in the GUI but would still be taken into account during the simulation. (This value is automatically imported from the scenario file.)
• Display frame nr.: The frame number, i.e. projection number, to be set up in the aRTist scene. Press the button Show frame if you wish to see this specific frame configuration in the aRTist scene.
• Final projection is taken at stop angle: if activated, the last projection image will be taken at the stop angle. If not activated, the last projection image will be taken at the last angular step before the stage reaches the stop angle. This means that this option slightly changes the angular step size, depending on the choice. (This parameter is automatically imported from the scenario file.)
• Show frame: The aRTist scene will be set up to show the projection number chosen under Display frame nr. The other parameters described above are taken into account as well.
• The buttons < (previous frame) and > (next frame) let you browse through the CT scan frame by frame.

Simulation

For a full CT scan simulation, the following options are provided.

• Output folder: The folder where the simulation output should be stored: projection images, metadata and configuration files. By default, the module will set a folder that carries the same name as the JSON file.

• Projection base name: the base file name that the output files should carry, with a sequential number that will be added to the projection images. In the shown example screenshot, the 21 generated projection images would be called:

  01_full_example_0000.tif, ..., 01_full_example_0020.tif

• Start at projection nr.: The projection number where the simulation should be picked up. The first projection image always has the number 0, but if you want to start the simulation at a later projection number, you can enter that here. This can be used if a simulation was interrupted and should be continued at a certain projection number.

• Scatter image interval: if scatter simulation is activated in the JSON scenario, aRTist will use McRay to calculate scatter images for each projection image. If the scatter simulation should not run for every projection image, a scatter image interval can be specified here to save computation time. A calculated scatter image will remain valid for this number of projection images. In the example screenshot, the scatter image will be re-calculated for every fifth projection image, and then be used for the following five simulated images. For a value of 0 or 1, the scatter image is calculated for each projection. (This value is automatically imported from the scenario file if the corresponding aRTist-specific parameter has been set. See below for aRTist-specific parameters supported by the module.)

• File format: for the projection image output. Can be TIFF or RAW.

• Data type: for the projection image output. Can be 16-bit unsigned integer data (uint16) or 32-bit floating point data (float32).

Controlling the CT simulation:

• Run scenario: simulates the scenario as described in the JSON file, taking into account the simulation settings from the GUI. Projection images, metadata and configuration files will be stored in the given output folder. (Note that a scenario must always be loaded before it can be run.)
• Stop: cancel a running simulation.

Batch Manager

The module comes with a batch manager where JSON scenario files can be queued and simulated sequentially in a batch.

Controlling the batch manager

The batch manager can be controlled from a row of buttons at the bottom of the module window:

• + choose a JSON scenario file and add it to the job list.
• – delete the selected job from the job list.
• Run: run all pending jobs from the job list.
• Stop: stop the current simulation and halt the execution of the batch.
• Save List...: save the job list as a CSV file.
• Import...: import a job list from a CSV file.
• Clear: remove all jobs from the job list.

Editing batch jobs

Most fields in the job list can be edited. The fields have the following meanings.

• Status: current status of the job. The status can be changed manually by clicking on the current status entry.
  • Pending: the job will be run during the batch simulation.
  • Running: the job is currently being simulated. The number of the current run will be shown.
  • Inactive: the job will be skipped during the batch simulation.
  • Done: the simulation of the job is done.
  • ERROR: an error occurred during the simulation of the batch job. Load the scenario file manually and try to simulate it to find the error. If the scenario loads correctly, the error might be that you don't have write permission or enough memory to store the projection images.
• Runs: the number of simulation repetitions, i.e. how many times the job is simulated. Mostly useful for noisy scans with random elements. A separate folder will automatically be created for each simulation run if more than one run is being simulated by the batch manager.
• StartRun: the number of the run where the simulation should start. If a simulation has been interrupted, it can be picked up at another run number, given here as the StartRun.
• StartProjNr: the projection number where the simulation should start for the StartRun. If a simulation has been interrupted, it can be picked up at this projection number. The StartProjNr only applies to the StartRun; subsequent higher runs will all start at projection 0 again.
• JSON Scenario: file that contains the CTSimU scenario description.
• Output Format: file format and data type for the simulated projection images. By default, the settings from the CT Setup tab are used.
• Output Folder: where to store the simulated projection images and additional metadata and configuration files.
• Projection Base Name: the base file name that the output files should carry, with a sequential number that will be added to the projection images. (See CT Setup above.)

Settings

Parameters that can be set up in the Settings tab:

• Show stage coordinate system in scene: when a scenario is loaded, the sample stage will be added as an object to aRTist's part list and will be displayed in the scene as a coordinate system. The stage's material will be set to none, it will therefore not have any influence on the projection image. The stage model displays the current location and orientation of the sample stage in the selected frame. The size of the stage model will be scaled to match 2/3 of the scenario's detector height.
• Restart aRTist after each batch run: to save memory, the batch manager can restart aRTist after each completion of a batch run. The batch simulation will continue after the restart.
• Single runs: include run number in file names: The batch manager allows repeated simulation runs of a single scenario. For each run, a separate folder (with adapted file names) is created. For consistency, even single runs can be forced to include the run number (run001) in their file and folder names. This may allow for better automatic post-processing.
• Skip simulation, only create configs and metadata: This option can be activated if aRTist should not run a simulation of projection images, but instead only create the folder structure, metadata files and reconstruction configuration files.
• Metadata contact name: Metadata files include a field for the contact name, i.e., the person or institution that was in charge of the simulation. This name can be entered here.

The options in the scenario loading group let you exclude certain features from the simulation. This can speed up scenario loading and is meant for testing and development purposes. If any of these options is deactivated, a warning is show during the simulation that only a partially loaded scenario is being simulated. Additionally, the full_simulation property in the metadata file will be set to false.

• Compute full detector: If deactivated, a simple, ideal detector will be taken instead of computing the full detector characteristics.
• Compute full X-ray source: If deactivated, a simple monochromatic X-ray source will be simulated instead of computing a full spectrum. Note that this may not play well with a custom-made detector. When in doubt, also deactivate the full detector computation.
• Load samples: Import the samples from the scenario file?
• Set multisampling: Set multisampling for X-ray source and detector; either the CTSimU standard values or the values defined in the scenario. If deactivated, a very simple 1x1 multisampling will be used for X-ray source and detector to speed up the simulation.
• Set scattering if required: Activate scattering if required by the scenario. If deactivated, scatter simulation will always be turned off to save simulation time.

Reconstruction

The module lets you generate configuration files for third-party reconstruction software. Details about these config files can be set in the Reconstruction tab:

• Volume data type: data type of the reconstruction's output volume file.
• Use uncorrected projections: Typically, simulated projection images need to be flat-field corrected (and possibly dark-field corrected) after the simulation, before they are given to the reconstruction software. The CTSimU Toolbox provides means to do so, and the module will create a flat field correction Python script specifically as part of the simulation. When this option is activated, the reconstruction configuration will not refer to the corrected projection images created by the toolbox, but instead to the uncorrected images from the simulation. Some reconstruction software provide their own means for flat-field correction, and the module will specifically enable flat-field and dark-field correction in the OpenCT config files. This option should also be activated if flat-field correction is done in aRTist during the simulation.
• Create CERA config file: creates a CERA reconstruction configuration file as part of a scan simulation. The configuration file will come with a list of projection matrices for each frame that CERA can use for the reconstruction. However, the reconstruction configuration will also contain parameters for a circular trajectory reconstruction. If these parameters shall be used by CERA, the list of projection matrices should be de-referenced in the config file, i.e. the parameter ProjectionMatrixFilename should be commented or removed from the config file.

OpenCT options

• Create OpenCT config file: creates an OpenCT reconstruction configuration file, for example supported by VGSTUDIO.
• Use absolute file paths: absolute paths will be used to refer to projection images in the config file instead of relative paths.
• Enforce circular format instead of free trajectory: OpenCT comes in two variants: circular trajectory and free trajectory. By default, the module creates the free trajectory format, as this is the most general. However, depending on the reconstruction software and the available licenses, the circular trajectory might be useful. Keep in mind that the circular variant is only meant for ideal circular cone-beam CT trajectories, with the rotation axis and cone beam centered on the detector and no tilts.
• Create clFDK run script: Creates a batch file that allows to call BAM clFDK with the OpenCT config file.

Tcl API

The module is written in Tcl, in a way that it can also be used without aRTist (in a limited way). There are also functions to control the module within aRTist by passing a prepared Tcl script as an argument to an aRTist instance. The documentation of the code, API and classes can be found in the docs folder. You can find demo Tcl scripts in the examples folder of this repository.

Feature support for JSON scenarios

The following table lists the module's current support status for the JSON parameters defined by the CTSimU file format.

Parameter	Support	Drift Support
environment material_id	yes	only in material definition (density, composition)
environment temperature	no	no
geometry detector center x/y/z	yes	yes
geometry detector vector_u x/y/z	yes	yes
geometry detector vector_w x/y/z	yes	yes
geometry detector deviations	yes	yes
geometry source type	cone	no
geometry source beam_divergence	no	no
geometry source center x/y/z	yes	yes
geometry source vector_u x/y/z	yes	yes
geometry source vector_w x/y/z	yes	yes
geometry source deviations	yes	yes
geometry stage center x/y/z	yes	yes
geometry stage vector_u x/y/z	yes	yes
geometry stage vector_w x/y/z	yes	yes
geometry stage deviations	yes	yes
detector model	yes	no
detector manufacturer	yes	no
detector type	yes	no
detector columns	yes	yes
detector rows	yes	yes
detector pixel_pitch u/v	yes	yes
detector bit_depth	yes	no
detector integration_time	yes	yes
detector dead_time	no	no
detector image_lag	no	no
detector gray_value imax	yes	yes
detector gray_value imin	yes	yes
detector gray_value factor	yes	yes
detector gray_value offset	yes	yes
detector gray_value intensity_characteristics_file	yes	yes
detector gray_value efficiency_characteristics_file	yes	yes
detector noise snr_at_imax	yes	yes
detector noise noise_characteristics_file	yes	yes
detector gain	no	no
detector unsharpness basic_spatial_resolution	yes	yes
detector unsharpness mtf	yes	yes
detector bad_pixel_map	no	no
detector scintillator material_id	yes	only in material definition (density, composition)
detector scintillator thickness	yes	yes
detector window front	yes	yes
detector window rear	no	no
detector filters front	yes	yes
detector filters rear	no	no
source model	yes	no
source manufacturer	yes	no
source voltage	yes	yes
source current	yes	yes
source target material_id	yes	only in material definition (density, composition)
source target type	yes	no
source target thickness	yes	yes
source target angle incidence	yes	yes
source target angle emission	yes	yes
source spot size u/v/w	yes	yes
source spot sigma u/v/w	yes	yes
source spot intensity_map file	yes	yes
source spot intensity_map type	yes	no
source spot intensity_map dim_x/y/z	yes	no
source spot intensity_map endian	yes	no
source spot intensity_map headersize	yes	no
source spectrum monochromatic	yes	no
source spectrum file	yes	yes
source window material_id	yes	only in material definition (density, composition)
source window thickness	yes	yes
source filters material_id	yes	only in material definition (density, composition)
source filters thickness	yes	yes
samples name	yes	no
samples file	yes	yes
samples unit	yes	no
samples scaling_factor r/s/t	yes	yes
samples material_id	yes	only in material definition (density, composition)
samples position center u/v/w/x/y/z	yes	yes
samples position vector_r u/v/w/x/y/z	yes	yes
samples position vector_t u/v/w/x/y/z	yes	yes
samples position deviations	yes	yes
acquisition start_angle	yes	no
acquisition stop_angle	yes	no
acquisition direction	yes	no
acquisition scan_mode	"stop+go"	no
acquisition scan_speed	no	no
acquisition number_of_projections	yes	no
acquisition include_final_angle	yes	no
acquisition frame_average	yes	no
acquisition dark_field number	yes	no
acquisition dark_field frame_average	-	no
acquisition dark_field ideal	only ideal	no
acquisition dark_field correction	no	no
acquisition flat_field number	yes	no
acquisition flat_field frame_average	yes	no
acquisition flat_field ideal	yes	no
acquisition flat_field correction	yes	no
acquisition pixel_binning u/v	no	no
acquisition scattering	yes (McRay)	no
materials id	yes	no
materials name	yes	no
materials density	yes	yes
materials composition formula	yes	yes
materials composition mass_fraction	yes	yes

Deviations

Geometry deviations as specified in the file format are fully supported.

Parameter	Support	Drift Support
type	yes	no
axis as name string	yes	no
axis x/y/z	yes	yes
axis u/v/w	yes	yes
axis r/s/t	yes	yes
pivot x/y/z	yes	yes
pivot u/v/w	yes	yes
pivot r/s/t	yes	yes
amount	yes	yes
known_to_reconstruction	yes	no

aRTist-specific parameters

The module supports additional parameters that are specific for aRTist. These can be set up in the "simulation" section of the JSON file:

{
  "simulation": {
    "aRTist": {
      "multisampling_detector":    {"value": "3x3"},
      "multisampling_spot":        {"value": "30"},
      "spectral_resolution":       {"value": 1, "unit": "keV"},
      "scattering_mcray_photons":  {"value": 5e7},
      "scattering_image_interval": {"value": 5},
      "long_range_unsharpness": {
        "extension": {"value":  2, "unit": "mm"},
        "ratio":     {"value": 10, "unit": "%"}
      },
      "primary_energies": false,
      "primary_intensities": false
    }
  }
}

Parameter	Support	Drift Support
simulation aRTist multisampling_detector	yes	yes
simulation aRTist multisampling_spot	yes	yes
simulation aRTist spectral_resolution	yes	no
simulation aRTist scattering_mcray_photons	yes	no
simulation aRTist scattering_image_interval	yes	no
simulation aRTist long_range_unsharpness extension	yes	yes
simulation aRTist long_range_unsharpness ratio	yes	yes
simulation aRTist primary_energies	yes	no
simulation aRTist primary_intensities	yes	no

The devil in the details

• If X-ray spectra are loaded from an external file instead of being generated in aRTist, the JSON file should still specify the tube's correct acceleration voltage. If the maximum energy in the spectrum drifts significantly (by more than 20 keV), the source's voltage parameter should also drift, reflecting the maximum energy in the spectrum. Alternatively, instead of a drift, the voltage parameter could simply represent the maximum acceleration voltage during the scan. The reason is that the maximum photon energy is needed for the computation of the detector's sensitivity characteristics, in order to save computation time and only run the computations for the necessary energy range. This maximum photon energy is determined from the source's voltage parameter and not from an externally loaded spectrum.
• X-ray spectra are always expected to be already filtered by any defined source window. If the window thickness or its material drifts, the externally loaded spectrum will not be re-filtered to account for these changes. Instead, the externally loaded spectrum should drift as well (to account for the changing window). If aRTist calculates the spectrum itself, any window drifts are taken into account.
• When using frame averaging, make sure to understand how aRTist's frame averaging works. aRTist's approach is to shift the SNR curve to the intensity of the integration time of the accumulated number of averaged frames, and then take one projection image. It does not calculate several frames and mathematically average them. This can give an undesired result if you have a non-standard SNR curve, especially when aRTist has to interpolate the curve to higher intensities if they are not explicitly given.

Deploying a new version

The deploy.sh script can be used to create an .artp file for aRTist for a certain version number. The new version number is passed as an argument:

./deploy.sh "<version>"

For example:

./deploy.sh "1.2.0"

Note: the aRTist package file (.artp) should not be part of the git repository. Instead, it can be uploaded to Github as a file attachment to a new release.

Version History

1.2.3

• Option to create post-processing Python scripts for combined flat-field correction and creation of reconstruction configuration files using the CTSimU toolbox.
• Updated to metadata file format version 1.2.
• Projection matrices: code simplification, minor speedup.

1.2.2

• New options:
  • Option to skip the simulation and only create metadata and reconstruction configurations.
  • Quick scenario loading: certain time-consuming computations can be deactivated when a scenario is loaded. A warning is displayed during the simulation of a partly loaded scenario and the metadata file includes information about which features have been excluded from the simulation.
  • Option to always include the run number in file names, even for single runs.
  • Contact name for metadata files can be specified in settings.
  • Reconstruction config files can be generated for uncorrected projection images (instead of corrected ones that are assumed to be corrected by the CTSimU toolbox). This is especially useful in combination with VGSTUDIO and OpenCT, where the flat-field correction is done by the reconstruction software in this case.
• OpenCT config files:
  • Option to use absolute paths in config files instead of relative paths.
  • Correction images (bright, dark) are included in OpenCT config if JSON loader is instructed to create a config file for uncorrected projections.
  • Circular trajectory variant of the file format can be enforced. Only meant for ideal circular trajectories (see README in the main git repository).
  • Updated to full OpenCT file format version 1.0.0, but still includes legacy keys for compatibility.
• Extended metadata files:
  • Two metadata files are created: one for projections, one for reconstruction.
  • Reconstruction metadata includes information about the tomogram volume to be reconstructed.
  • All metadata files include some aRTist-specific simulation information and the complete scenario JSON that has been simulated.
• Bug fixes:
  • Run number is now included in flat-field correction filenames (for multiple runs, or when this behavior is enforced in the settings).
  • Flat-field correction is turned off while taking the dark image. (Only applicable for scenarios where flat-field correction is done by aRTist during the simulation.)
  • Projection matrices: if no image/volume coordinate system is given, a default one is created automatically.
  • Drifts imply their parent parameter's unit if they don't have their own unit specified in the scenario (as required by the CTSimU file format).
• Minor changes (only relevant for developers):
  • Prepared native unit for angular velocity (deg/s) to be on the same level as the toolbox. Angular velocities still do not play any role in a CTSimU simulation in aRTist.

1.2.1

• Bug fix: projection counter error for more than one ideal flat field.

1.2.0

• General support for file format version 1.0, 1.1 and 1.2
• Parameter drifts are supported
• Scatter images can be kept for a user-defined number of frames, instead of calculating a new scatter image for each frame.
• Option to restart aRTist after each batch run. (To free memory.)
• New supported JSON features:
  • new geometrical deviations (translations and rotations along arbitrary axes or pivot points)
  • source: spot images
  • detector: MTF
  • detector: quantum efficiency
  • detector: external noise characteristics files
  • known_to_reconstruction parameter for geometrical deviations and drifts is obeyed when calculating projection matrices.

0.8.18

• Fix in CERA reconstruction config files: the Datatype parameter is now inactive (made into a comment), only OutputDatatype left as valid parameter for the volume data type.

0.8.17

• Small updates and fixes in reconstruction config file outputs.

0.8.16

• Added aRTist-specific JSON option to create primary intensity images (as an alternative to primary energy images).

0.8.15

• Minor bug fix: Parts not loaded centered at (0,0,0) since version 0.7.10 due to a spelling mistake. This bug had no impact on part placement: centering parts after loading is not necessary and has therefore been fully removed in this version.

0.8.14

• Removed iSRb correction scaling factor of 0.6830822016 (for aRTist 2.12+).
• Added aRTist-specific JSON option to create primary energy images.

0.8.13

• Added options for recon volume data type (uint16, float32).
• Cleaned up VGI output.
• Physical pixel size is used in factor/offset detector grey value method, instead of iSRb-scaled pixel area from the DetectorCalc module.

0.8.12

• Added new function: insertBatchJob, to allow better control of batch job parameters.

0.8.11

• Added spectral resolution for spectrum generator as aRTist-specific parameter.

0.8.10

• Finite spot size scenarios: spot sampling will be set to 20 and detector sampling to 2x2 (instead of 30 and 'source dependent' as before).

0.8.9

• Bug fix: relative paths in recon configs now correct for batches with multiple runs.
• Bug fix: CERA format specifier for float32 raw projection images.

0.8.8

• Bug fix: tubes with no window no longer wrongfully show 'Al 4 mm' in their name.

0.8.7

• Fixed aRTist unsharpness (moved from Gaussian sigma correction factor 1/sqrt(2) to 0.683).

0.8.6

• Added support for running the flat field correction during a scan.

0.8.5

• Added support for long range unsharpness.

0.8.4

• Scans can start at user-defined projection numbers so simulations can be continued after crashes.
• Index numbers in projection filenames only take 4 digits (or more, only if necessary).
• Scattering is automatically deactivated for dark images.

0.8.2 - 0.8.3

• No release; used for development tests.

0.8.1

• Added ffRescaleFactor to generated flat field correction scripts.

0.8.0

• Bug fix: memory leak when running a simulation
• Batch Manager: 'StartRun' specifies the run id where the batch should start. Enables picking up simulations after a crash.
• Support for simulation software-specific parameters: 'multisampling_detector', 'multisampling_spot', 'scattering_mcray_photons'

0.7.15

• Bug fix: division by zero when SOD is 0.

0.7.14

• Multiple JSON scenarios can be added at once to the batch queue.

0.7.13

• Added fileutil as a package requirement. Some systems showed problems without it.

0.7.12

• Added the object bounding box to clFDK/VG reconstruction configuration files.

0.7.11

• Added support for CERA's RDabcu0v0 coordinate system to provide a reconstruction configuration alternative to the projection matrix approach.

0.7.10

• For CT scans, reconstruction configurations for SIEMENS CERA, BAM clFDK, and also VG Studio (untested) are created using projection matrices.
• Bug fixes concerning unusual geometries.
• Last used settings for output file types and reconstruction configurations are saved when aRTist closes.

0.7.9

• Removed option to (not) import all samples from JSON, to avoid confusion. Loading a scenario will now always load complete scenario.

0.7.8

• 'rl_json' is now used for JSON handling, instead of tcllib's json module.

0.7.7

• Added a batch manager to handle job queues.
• Environment material is considered as an additional spectrum filter when calculating the detector characteristics. Therefore, the specified grey values at maximum intensity are reached for any environment material, not only in vacuum.

0.7.6

• Fixed memory leak when running a CT scan.
• JSON files are now closed after being read, and no longer blocked from deletion/renaming.
• Applied Florian Wohlgemuth's fix of CanClose() for using the module in batch simulations.

0.7.5

• Fixed detector saturation when using a filter in front of an ideal detector.

0.7.4

• CTSimU metadata files are created along CT scan projection image stacks.
• Fixed a bug concerning orientation vectors pointing in opposite directions (e.g. sample flipped in stage coordinate system).

0.7.2

• Bug fixes concerning coordinate system positioning.

0.7.1

• Implemented sample scaling (according to definition in JSON's sample section).

0.7.0

• Support for CTSimU JSON scenario definition version 0.7

0.6.0

• Support for CTSimU JSON scenario definition version 0.6

0.5.0

• Support for CTSimU JSON scenario definition version 0.5

0.4.0

• Support for CTSimU JSON scenario definition version 0.4