- 1 example_ARF
- 2 example_CT
- 3 example_DNA
- 4 example_dosimetry
Angular Response Function (ARF) Tables Data computation
In this step, the data needed to generate the ARF tables are computed from a rectangular source located at the center of the field of view. The SPECT head is duplicated twice and located at roughly 30 cm from the axial axis. The command to tell GATE to compute the ARF data is:
The ARF data are stored in the ROOT format from the GATE command output :
By default the maximum size of a ROOT file is 1.8 Gb. Once the file has reached this size, ROOT automatically closes it and open a new file name testARFdata_1.root. when this file reaches 1.8 Gb, it is closed and a new file testARFdata_2.root if needed is created and so on ...
A template macro file is provided in gate_v4.0.0/examples/example_ARF/generateARFdata.mac which summarizes the commands listed before.
Computation of the ARF tables from ARF data (previous step)
Now that the data have been stored to ROOT files, we may compute the tables and store them to a binary file :
# COMPUTE THE ARF TABLES FROM ARF DATA /gate/systems/SPECThead/arf/setARFStage computeTables
# DIGITIZER PART OF THE ARF SENSISTIVE DETECTOR /gate/systems/SPECThead/ARFTables/setEnergyDepositionThreshHold 328. keV /gate/systems/SPECThead/ARFTables/setEnergyDepositionUpHold 400. keV /gate/systems/SPECThead/ARFTables/setEnergyResolution 0.10 /gate/systems/SPECThead/ARFTables/setEnergyOfReference 140. keV
Here we shot 364.5 keV photons. We choose an energy resolution of 10% from which we obtain 328 keV and 400 keV as the energy window boundaries. The energy of reference is chosen to be 140 keV which allows to define the FWHM of the response of the detector as FWHM = 0.10 x sqrt(140. x Edep ), where Edep is the photon deposited energy.
If we do not want to consider photons which deposit less than 130 keV we may use:
/gate/systems/SPECThead/setEnergyDepositionThreshHold 130. keV
The ARF tables depend strongly on the distance from the detector to the source. We have to enter this parameter to get proper tables. The detector plane is set to be the half—middle plan of the detector part of the SPECT head. In the given example, we set the translation of the SPECT head to be 34.5 cm along the X direction (radial axis), the detector is 2 cm wide along X and its center is located at X = 1.5 cm with respect to the SPECThead frame. This is what we call the detector plane ( x = 1.5 cm ). So the distance from the source to the detector plane is 34.5 +1.5 = 36 cm :
# DISTANCE FROM SOURCE TO DETECTOR PLANE TAKEN TO BE THE PLANE AT HALF DIMENSION OF THE CRYSTAL RESPECTIVELY TO THE SPECTHEAD FRAME /gate/systems/SPECThead/ARFTables/setDistanceFromSourceToDetector 36 cm
Now we compute the ARF tables from a text file which contains informations regarding the incident photons called ARFData.txt (provided in gate/examples/example_ARF).
# NOW WE ARE READY TO COMPUTE THE ARF TABLES /gate/systems/SPECThead/ARFTables/ComputeTablesFromEnergyWindows ARFData.txt
The ARFData.txt file reads like this :
# This file contains all the energy windows computed during first step with their associated root files # It has to be formatted the following way : # [Emin,Emax] is the energy window of the incident photons # The Base FileName is the the name of the root files name.root, name_1.root name_2.root ... # The number of these files has to be given as the last parameter. # # Enter the datas for each energy window : # Inident Energy Window : Emin (keV) - Emax (keV) | Associated Root Base FileName | number of these files 0. 365. test1head 20
Here we have only one incident energy window for which we want to compute the corresponding ARF tables. The data for the first one are stored inside 20 files whose base file name is test1head which were generated in the previous stage. It means the ARF data are stored in test1head.root, test1head_1.root ... test1head_20.root.
Finally we store the computed tables to a binary file :
/gate/systems/SPECThead/ARFTables/list # SAVE ARF TABLES TO A BINARY FILE FOR PRODUCTION USE /gate/systems/SPECThead/ARFTables/saveARFTablesToBinaryFile ARFSPECTBench.bin
Use of the ARF Tables
The command to tell GATE to use the ARF tables is :
The ARF sensitive detector is attached to the SPECThead :
These tables are loaded from binary files with :
The CT example simulation (Figure 1) is made of an X-Rays conical source, with 6.8° angle emission, a CT scanner with 100x100 Silicon (0.5x0.5x1 mm3) pixels and a cylindrical phantom made of Water containing 4 balls (Aluminium, Glass, SpineBone, PVC). During the simulation 16,835,281 photons are detected.
There are 2 runs. Each run performs during 90 seconds. The phantom turns around its central axis at 1 deg per second.
Building and Running AnalyzeCT.cpp
Compile the C++ file (in the classic directory) with the following command:
g++ -O3 `root-config -cflags -glibs` AnalyzeCT.cpp -o AnalyzeCT
Figure 2 shows the Energy spectrum for the first and second projection (first and second run). For the first Run, from top-left to bottom-right, the plot shows the energy spectrum outside the phantom, in the phantom, behind the 2 Top balls and behind the 2 bottom balls. For the second Run, from top-left to bottom-right, the plot shows the energy spectrum behind the PVC ball, behind the Aluminium ball, behind the Glass ball and behind the Spine Bone ball.
This example shows how to mix Standard and DNA physics list in the same simulation. The geometry consists in a world volume made of liquid water (!! Be careful you must use the so-called "G4_WATER" material while using Geant4 DNA processes and models !!) including three boxes Target1, Target2, Target3 made of liquid water (G4_WATER) of thickness 50 micrometers aligned along the Z axis. Monoenergetic protons of 8 MeV are shot along the Z axis at 10 micrometers of the face of the first box Target1. Geant4 Standard option 3 physics list is assigned to the world volume with a cut of 4 micrometers (2.9 keV electrons in liquid water). Geant4 DNA physics list is assigned to Target1 and Target3 (note that cut is not active using Geant4 DNA processes and models). A KillActor is used to kill electrons with kinetic energy below 8 eV. It prevents electrons to be trapped into single scattering.
The production of secondary particles is scored in each Target via EnergySpectrumActor. To analyse the spectrum with ROOT, use the plot.C macro file. One can check the correct production of secondaries: from 2.9 keV in Target2, whereas the Geant4 DNA models in Target1 and Target3 are able to produce secondaries below 2.9 keV.
This example illustrates the simulation of a brachytherapy treatment using a titanium capsule and a source of Iodine 125.
This example describes the simulation of a Clinac 2100C linear accelerator operating in 6 MeV electron mode for a thoracic irradiation of a RANDO phantom.
!!Warning!! This example should only be used for illustration purpose. Due to NDA with manufacturer, the geometry of the devices provided in the macros have been modified and are not the real ones.
The linac head consists in :
+ a vacuum window : exit window in beryllium + a 6 MeV scattering foil : in tantalum + an ionization chamber : kapton-air-copper slabs + X and Y Jaws : defining a 20x20cm2 aperture adapted to the 10x10cm2 applicator + a 10x10cm2 applicator + a 10x10cm2 cerrobend cut-out
The source is described as a circular surface source. The electron spectrum is assumed to be Gaussian with a mean energy of 7.2 MeV and a standard deviation of 3%.
external beam therapy photon
This example illustrates a photon beam therapy in a patient CT image. The output is a 3D dose distribution map (with associated uncertainty). The phantom being a patient CT image, one needs to generate materials from the image Hounsfield units :
/gate/HounsfieldMaterialGenerator/SetMaterialTable data/Schneider2000MaterialsTable.txt /gate/HounsfieldMaterialGenerator/SetDensityTable data/Schneider2000DensitiesTable.txt /gate/HounsfieldMaterialGenerator/SetDensityTolerance 0.1 g/cm3 /gate/HounsfieldMaterialGenerator/SetOutputMaterialDatabaseFilename data/patient-HUmaterials.db /gate/HounsfieldMaterialGenerator/SetOutputHUMaterialFilename data/patient-HU2mat.txt /gate/HounsfieldMaterialGenerator/Generate
After the conversion from HU to materilas, the patient CT is inserted as the simulation phantom :
/gate/world/daughters/name patient /gate/world/daughters/insert ImageRegionalizedVolume /gate/geometry/setMaterialDatabase data/patient-HUmaterials.db /gate/patient/geometry/setHUToMaterialFile data/patient-HU2mat.txt /gate/patient/geometry/setImage data/patient-2mm.mhd <==== real patient CT scan /gate/patient/placement/setTranslation 0 0 0 mm /gate/patient/geometry/TranslateTheImageAtThisIsoCenter 109.7 99.3 146.2 mm
A DoseActor is used to store 3D distributions of dose/edep/uncertainty/nbHit into files (mhd image file format) :
/gate/actor/addActor DoseActor doseDistribution /gate/actor/doseDistribution/attachTo patient /gate/actor/doseDistribution/stepHitType random /gate/actor/doseDistribution/setPosition 0 0 0 mm /gate/actor/doseDistribution/setVoxelSize 2 2 2 mm /gate/actor/doseDistribution/enableEdep true /gate/actor/doseDistribution/enableUncertaintyEdep true /gate/actor/doseDistribution/enableDose true /gate/actor/doseDistribution/enableNumberOfHits false /gate/actor/doseDistribution/save output/photon.mhd