KAHIBARO
47.5. Adding an F-18 Source
Table of Contents
Understanding the Role of the F-18 Source
In the PET scanner example, the F-18 source represents the radioactive tracer used in clinical PET imaging. Your PET geometry and digitizer chain only become meaningful when they are irradiated by a realistic positron-emitting source. In GATE, this means configuring a radioactive source that follows the physics of F-18 decay, produces positrons, and consequently generates 511 keV annihilation photons inside the field of view.
In this chapter you will learn how to add such a source to the PET ring you already created. You will focus on how to define the nuclide, how to place the activity in space, and how to connect the source definition to the acquisition time and activity settings used in a realistic PET simulation.
Basic F-18 Source Definition
You typically add an F-18 source after the PET ring geometry and before configuring the PET physics and digitizer. In OpenGATE, a source is usually created through the simulation object. Conceptually, you will always perform the same steps:
- Create a new source object and give it a descriptive name, for example
"F18_source". - Select a radioactive type source and specify the radionuclide, here F-18.
- Define the activity, which controls how many decays per unit time occur.
- Define the spatial distribution, which controls where in the scanner the decays start.
- Optionally, define the time profile if you want a constant or time varying activity over the acquisition.
A simple conceptual Python-style snippet, omitting framework details that are handled in other chapters, would look like this:
source = sim.add_source("F18_source")
source.particle = "ion"
source.is_radioactive = True
source.nuclide = "F18"
source.activity = 1.0 * gate.g4_units.MBqThe exact property names might differ slightly depending on the OpenGATE version, but the structure is the same. You identify the source as radioactive, select the nuclide and assign an activity with proper units.
Always specify the radionuclide explicitly (for example nuclide = "F18") and assign the activity with correct units, typically in Bq, kBq, or MBq. Forgetting to mark a source as radioactive or omitting units can result in a non-decaying, non-PET-like source.
F-18 Decay and PET Relevance
F-18 is a positron emitter. It undergoes $\beta^{+}$ decay to O-18, emitting a positron that quickly loses energy and annihilates with an electron. The annihilation produces two almost back-to-back 511 keV photons. You do not have to program this decay chain manually, because GATE relies on Geant4’s radioactive decay models when you define a radionuclide source.
The choice of F-18 has two main effects on your simulation:
First, it sets the positron energy spectrum and range in tissue, which influences the small blurring of the annihilation position relative to the initial decay position. Second, it fixes the time behavior according to the half-life of F-18. In a simple PET acquisition of a few minutes, you will often ignore decay during acquisition and use a constant activity, but for longer scans you may choose a time dependent activity.
As long as you use an F-18 radionuclide source, the annihilation photons and positron physics are handled by the physics configuration you will adjust in the PET physics configuration chapter, not here.
Point versus Distributed F-18 Activity
The next key choice is the geometry of the radioactive distribution. In a basic PET example you commonly start with either a point-like F-18 source at the center of the scanner or with a simple phantom volume filled with F-18.
A point source is useful for checking basic PET geometry and coincidence handling. You might, for example, place a source at the center of the field of view:
source.position.type = "point"
source.position.center = [0, 0, 0]Here the center coordinates must be given in the correct length units, such as millimeters or centimeters, consistent with the geometry already defined.
For a more realistic test, you may want a uniform distribution of activity inside a simple phantom volume, like a cylinder or a box that you already created in the scanner. In that case, you can define a volume type position and associate it with the phantom:
source.position.type = "volume"
source.position.volume = "phantom"
source.position.distribution = "uniform"GATE will then randomly sample decay positions inside that phantom volume. This is often the default approach when testing scanner sensitivity, scatter fraction, or reconstruction pipelines.
When using a volume-based F-18 distribution, ensure the named volume actually exists and is fully inside the PET field of view. Mis-typed volume names or phantoms placed partly outside the scanner can lead to sources that effectively emit outside the sensitive detector region.
Setting F-18 Activity and Acquisition Time
In PET, activity and acquisition time jointly determine the total number of detected events. In GATE you usually specify the activity in Bq units and separately control the simulation time in the global run settings.
For example, you might configure:
source.activity = 10.0 * gate.g4_units.MBq
sim.time_stop = 300.0 * gate.g4_units.sThis corresponds to a constant 10 MBq F-18 source simulated over 300 seconds. The Monte Carlo will then sample the actual decay times stochastically over this interval.
The relation between activity $A$ and number of decays $N$ during a time interval $T$ (assuming constant activity) is
$$
N \approx A \times T.
$$
This gives you a rough idea of how many decays you might simulate. Remember that only a fraction of decays will lead to coincidences in your detectors.
Use activity values and acquisition times that lead to a manageable number of simulated decays. Extremely high activity or very long acquisition times can create huge event counts, long runtimes, and very large output files.
If you want to avoid thinking in terms of real time and activity, some setups allow you to configure a fixed number of decays instead of an activity. However, for a PET example focused on realism, you should practice with activity and acquisition time, because these are directly related to clinical scenarios.
Spatial and Angular Characteristics of the F-18 Source
For many PET simulations, you can treat F-18 emission as isotropic at the decay point. You therefore do not need to impose any custom emission direction, since the radioactive decay engine will govern both the positron and subsequent photon emission. You typically leave any direction-specific settings at their defaults for radioactive sources.
What matters more is the spatial distribution of activity, which has already been discussed in terms of point and volume sources. For more advanced testing, you may create non-uniform spatial distributions by sampling from an image-based activity map, but this belongs to the more advanced internal dosimetry and voxelized activity chapters.
In the PET scanner example, you should restrict yourself to a simple yet well controlled spatial configuration, such as a central point F-18 source for basic validation, and a uniform cylindrical phantom source for scanner performance checks. These are easier to interpret when you later analyze coincidences, lines of response, and PET performance metrics.
Connecting the F-18 Source to the PET Workflow
Once your F-18 source is configured with nuclide, activity, and spatial distribution, it becomes the driving input for the complete PET workflow. The F-18 decays will generate positrons and 511 keV photons, which then interact in the detector crystals to produce hits. The digitizer chain will transform those hits into singles and coincidences, which you analyze in later chapters.
At this stage, you should verify only a few basic properties:
First, confirm that the source is created and recognized by the simulation, for example by printing the simulation configuration. Second, check that the source position is where you expect, particularly relative to the PET ring geometry. Third, run a short test with a small time interval and modest activity to ensure that detectors record hits and singles, which indicates that the F-18 configuration is physically active.
You do not need to interpret spectra or coincidence statistics yet; that will be handled when you record hits, create singles, and sort coincidences in the following chapters. Here, the focus is to establish a correct and realistic F-18 source, which serves as the foundation for all subsequent PET analyses.
Views: 9