Volume Scorers

Here are the available volume scorers:

Quantity Description
AmbientDoseEquivalent sum of fluence times fluence-to-effective dose conversion coefficients
DoseToMedium sum of energy deposits divided by mass
DoseToWater from energy-dependent stopping power conversion (see below)
DoseToMaterial from energy-dependent stopping power conversion (see below)
EnergyDeposit sum of energy deposits
Fluence sum of step lengths divided by volume
EnergyFluence sum of step lengths times energy divided by volume
StepCount counting number of Geant4 steps in the volume
OpticalPhotonCount fills an ntuple with information about optical photons seen in volume
Charge counting method described below
EffectiveCharge counting method described below
ProtonLET various methods described below

Volume Scorers must indicate the relevant Component:

s:Sc/MyScorer/Component = "Phantom"

When your scoring component is the Parent of other components, you have the option to set scoring to record particles not only in the parent component, but also in its children:

b:Sc/MyScorer/PropagateToChildren = "True"

This action is recursive to all levels of subcomponents.

For DoseToMaterial, you must also specify the Material:

s:Sc/MyScorer/Material = "SomeMaterial"

Note that in this case, the material name must exactly match the case defined in Geant4. To check what materials have been defined, add the parameter:

i:Ma/Verbosity = 1

For AmbientDoseEquivalent, the scoring is performed per single particle:

s:Sc/MyScorer/Quantity = "AmbientDoseEquivalent"
s:Sc/MyScorer/Component = "MyDetectorComponent"
s:Sc/MyScorer/GetAmbientDoseEquivalentForParticleNamed = 1 “neutron”

The scorer uses a track-length estimator, a variance reduction technique that consists of retrieving the absorbed dose at the scoring regions (voxels) located along the particle path until the point of interaction. That improves the computational efficiency substantially for neutral particles. To that end, the ambient dose equivalent is obtained by folding the incident particle fluence, defined as the particle’s track-length divided by the scorer’s volume [1], with linearly-interpolated fluence-to-effective dose conversion coefficients. Then, the user needs to provide the corresponding fluence-to-effective dose conversion coefficients for a range of incident energy values using a couple of dimensioned double vectors:

dv:Sc/MyScorer/FluenceToDoseConversionEnergies = 30 ... MeV
dv:Sc/MyScorer/FluenceToDoseConversionValues = 30 ... Sv*mm2

The example AmbientDoseEquivalent.txt provides a complete example for neutron particles. It uses the fluence-to-effective dose conversion coefficients from reference [2], downloaded from reference [3].

[1] Attix FH, Introduction to radiological physics and radiation dosimetry, 1986 Wyley-VCH, Chapter 1, Section III.D.

[2] Pelliccioni, M. Overview of Fluence-to-Effective Dose and Fluence-to-Ambient Dose Equivalent Conversion Coefficients for High Energy Radiation Calculated Using the FLUKA Code, Radiat. Prot. Dosim. 88(4), 279-297 (2000).

[3] http://www.fluka.org/fluka.php?id=examples&sub=example4 Accessed on March 10, 2021.

For DoseToWater and DoseToMaterial, we use energy-dependent stopping power conversion as in:

dose_to_new_material = dose_to_medium * ( density_of_medium / density_of_new_material ) * ( dEdX_in_new_material / dEdX_in_medium )

The dEdX comes from the Geant4 EmCalculator utility.

The DoseToWater and DoseToMaterial scorers are somewhat slow since, for every hit, they need to compute stopping power ratios based on the current energy of the particle. You can obtain better speed by adding the option:

b:Sc/MyScorer/PreCalculateStoppingPowerRatios = "True" # defaults to "False"
  • False gives the best accuracy, calculating stopping power on-the-fly for the exact energy.
  • True gives the best speed, looking up stopping power from a pre-calculated table binned by energy. It is about 50% faster than the default option for typical patient simulations. The difference in accuracy is not significant for most studies.

For PreCalculateStoppingPowerRatios, the table of stopping power ratios can be tuned by:

Sc/MyScorer/ProtonEnergyBinSize # default is 1 MeV
Sc/MyScorer/MinProtonEnergyForStoppingPowerRatio # default is 1 MeV
Sc/MyScorer/MaxProtonEnergyForStoppingPowerRatio # default is 500 MeV
Sc/MyScorer/ElectronEnergyBinSize # default is 1 keV
Sc/MyScorer/MinElectronEnergyForStoppingPowerRatio # default is 1 keV
Sc/MyScorer/MaxElectronEnergyForStoppingPowerRatio # default is 1 MeV

For Charge and EffectiveCharge:

  • If a particle reaches zero kinetic energy in the scoring volume, its charge is accumulated
  • If a particle is generated in the scoring volume, its charge is subtracted
  • For Charge, the charge comes from the particle definition’s PDGCharge
  • For EffectCharge, the charge comes from the PreStepPoint’s Charge

ProtonLET Scorer

The ProtonLET scorer gives the LET of primary and secondary protons, including the energy deposited by associated secondary electrons. It uses techniques discussed in two recent articles on best practices to score LET in Geant4:

  • Phys. Med. Biol. 60 (2015) 2645–2669 by MA Cortes-Giraldo and A Carabe
  • Phys. Med. Biol. 60 (2015) N283–N291 by DA Granville and GO Sawakuchi

In particular, we adopt the methods developed by Granville and Sawakuchi. We compute dose-averaged LET, but you may instead request track-averaged:

s:Sc/MyScorer/WeightBy = "Track" # defaults to "Dose"

By default, the LET is computed by dividing the energy deposited by the step length. Such distributions can feature spurious spikes, caused by events where the step length is severely constrained by a voxel boundary crossing. Three solutions to this issue are provided:

  • By default, a step-by-step upper cut-off is set, such that steps contributing greater than this value are not be scored:

    d:Sc/MyScorer/MaxScoredLET = 100 MeV/mm/(g/cm3) # default 100 MeV/mm/(g/cm3)

  • Alternatively, you can set the LET computation to look up the electronic stopping power for the pre-step proton energy:

    b:Sc/MyScorer/UsePreStepLookup = "True" # defaults to “False”

  • Or you can increase the electron production threshold:

    d:Ph/Default/CutForElectron = 1 mm # defaults to 0.05 mm

The ProtonLET Scorer can give values that are too high in air, where the mean path length between discrete processes can be larger than the voxel size. This can be avoided by neglecting secondary electrons, with:

d:Sc/MyScorer/NeglectSecondariesBelowDensity = 0.1 g/cm3

Even when you do this, rare events that produce very low energy protons (e.g. a recoiling hydrogen nucleus) will produce spikes in LET. This is also seen in the PreStepLookup version of the scorer. They are not seen in the fluence-averaged version of the scorer, since they are rare events. For this reason we introduce the parameter:

d:Sc/MyScorer/UseFluenceWeightedBelowDensity = 0. g/cm3

We set this to zero by default because it is strange to mix both types of LET in a single distribution, and could be significantly wrong at the end of range. We expect users to want to enable this when making a pretty plot of LET to overlay on a CT scan, without spikes in cavities and outside the patient.