KAHIBARO
35.2. Proton Beam
Table of Contents
Beam energy
In proton therapy simulations, the beam energy is the main control parameter that sets the penetration depth of protons in the patient or in a phantom. In GATE, you define the proton beam energy in the source configuration, usually as a monoenergetic or slightly spread distribution around a nominal energy.
For a simple monoenergetic beam, you typically set the energy to a fixed value, for example 150 MeV, using the units provided by GATE. In Python-based GATE, this often looks like assigning a value such as $150 \,\text{MeV}$ to the source energy parameter, combined with the appropriate unit from the units module. Monoenergetic beams are useful when you want to study basic depth dose characteristics, compare with reference data, or clearly identify the Bragg peak.
Clinical proton beams are rarely perfectly monoenergetic. They usually have an energy spread described by a Gaussian or other distribution. In GATE, you can describe such realistic beams by selecting an energy distribution instead of a single value, and by specifying the mean energy and the full width at half maximum (FWHM) or standard deviation of the distribution. This is particularly relevant when you later build a spread out Bragg peak with multiple energies, but even for a single pencil beam a small spread will make your simulation closer to measurements.
The choice of beam energy should always be consistent with the water equivalent depth you want to reach. For example, a proton beam around 70 MeV stops in a few centimeters of water, while a 230 MeV beam can reach close to 30 cm. These ranges depend on the physics models and material composition, so part of the proton therapy workflow is to verify that the simulated range for a given energy is realistic.
For a given proton beam, the range in water is closely related to the beam energy. When you change the energy, you change the depth of the Bragg peak. Always verify that the simulated range for a selected energy agrees with reference data or treatment planning system predictions.
Another important consideration is the distinction between nominal beam energy and energy at the phantom surface. If your geometry includes beamline elements such as vacuum windows, air gaps, or range shifters, protons will lose energy before entering the phantom. In that case, when you set the source energy in GATE you are specifying the energy upstream, and you should be aware that the effective energy at the phantom surface will be lower. For accurate simulations, it can be useful to place an actor just before the phantom to record the energy spectrum there.
Beam size
The beam size describes the lateral spatial distribution of the proton fluence at the point where it enters the phantom or patient geometry. In GATE, you control this through the source position distribution. For a pencil beam, you often start with a narrow Gaussian distribution in the transverse directions, characterized by a standard deviation or FWHM in $x$ and $y$.
A typical configuration uses a point-like source in $z$ (the beam axis) and a Gaussian or circular distribution in the transverse plane. For instance, you may define a Gaussian profile with a sigma of a few millimeters, which yields a realistic clinical pencil beam spot size on the phantom surface. In some cases you will use a uniform distribution over a small circle, which is useful when modeling a top-hat beam profile.
In proton therapy simulations, the beam size has a strong impact on lateral dose distributions and penumbra. A smaller beam size gives a sharper lateral falloff of dose, while a larger beam size broadens the high dose region. This is especially important when you later combine multiple pencil beams in a treatment plan. Even in a basic water phantom study, you will see the influence of beam size when you examine lateral profiles at different depths.
You should define the beam size at a meaningful reference plane, usually at the entrance of the phantom. If you place the source upstream in air, the beam will diverge due to its initial angular spread and multiple scattering. In GATE, this means that the geometric source width that you specify will not equal the beam width at the phantom surface unless you consider the distance and divergence. For a simple beginner setup, it is common to place the source just at or very close to the phantom surface so that the defined size directly corresponds to the spot size on the phantom.
Table: qualitative effect of beam size on dose profiles in water
| Beam size (sigma) | Lateral penumbra | Central peak dose (for same total protons) |
|---|---|---|
| Very small | Very sharp, narrow field | Higher central dose |
| Moderate | Clinically realistic | Moderate central dose |
| Large | Broad field, wide penumbra | Lower central dose, more spread out |
Set the beam size to match the physical scenario you want to study. For a pencil beam in a water phantom, define a narrow spatial distribution at the phantom surface. For broader clinical fields, either use a larger distribution or later combine multiple pencil beams.
Direction
The beam direction specifies the orientation of the proton beam in your geometry. In proton therapy simulations with GATE, you control this by defining a direction vector for the primary particles in the source configuration. For a simple vertical or horizontal beam, you typically use a fixed unit vector, for example $(0, 0, 1)$ if your beam travels along the positive $z$ axis toward the phantom.
Direction is particularly important in proton therapy, because dose distributions and Bragg peak locations are sensitive to the path the beam takes through heterogeneous materials such as bone, lung, and soft tissue. In a simple water phantom example, you usually align the beam normal to the entrance surface, but as soon as you simulate more complex patient geometries you may need beams entering from different angles. In GATE, this is handled by adjusting the direction vector and, if necessary, the source position so that the beam intersects the target volume along the desired trajectory.
For advanced setups, you can also introduce an angular spread around the main direction, often modeled as a Gaussian distribution in polar angles. This reflects beam divergence from upstream optics and will influence both beam size growth with depth and lateral scattering. In the context of a basic proton beam chapter, the key idea is that a perfectly parallel beam uses a single fixed direction vector, while a more realistic beam can have a small angular spread superimposed.
When combining beam direction with other motion, for example gantry rotations or patient couch shifts, you often keep the source direction fixed relative to a beamline coordinate system while rotating either the geometry or the source itself. In GATE, you can achieve this either through transformations applied to volumes or by updating the source direction for each simulated beam angle. For simple educational simulations, you usually start with a static phantom and a single fixed beam direction, then later extend to multiple directions to represent clinical fields.
Always check that the beam direction points from the source toward the intended target. A small sign error in the direction vector can send protons away from the phantom, leading to almost no dose in the region of interest and misleading results.
In summary, for the proton beam in GATE you define energy to control depth, beam size to control lateral spread, and direction to control the path through the geometry. By adjusting these three elements systematically, you can reproduce basic proton therapy scenarios and prepare for more complex configurations in later chapters.
Views: 9