KAHIBARO
Hits vs Detector Signals
Table of Contents
Physical interactions
Inside GATE and Geant4, a hit represents information closely tied to the underlying physics of particle transport. A hit is created when a particle interacts in a sensitive detector volume and deposits energy there. Each interaction step that deposits energy can lead to a hit entry, depending on how the detector and scorers are configured. Hits are therefore very near to the Monte Carlo description of what happens in matter.
From a physical point of view, a hit carries attributes that describe a single interaction or a small sequence of interactions within a volume. Typical attributes include the deposited energy, the position of the interaction, the time, the particle type, and identifiers for the volume and the event. These quantities are expressed in the same units and coordinate systems as the rest of the simulation and refer directly to what the particle did inside the sensitive volume.
The timing stored in hits is usually the global simulation time at which the interaction occurred. This time already includes any transport delays due to the particle flight path and interactions, but it does not include delays caused by detector electronics or light transport unless those are explicitly modeled at the physics level. Similarly, the position in a hit corresponds to where the energy was deposited, not to where an electronic signal was read out. This distinction becomes important when the sensitive volume is larger than the electronic channel or when multiple channels share the same piece of scintillator.
Hits are also typically more detailed than what the detector ultimately provides. A single gamma photon can create many hits as it Compton scatters and eventually photoabsorbs within a crystal. Likewise, an electron can produce a dense cascade of interactions, each represented as one or more hits. These low level interactions are useful for studying detector physics, validating models, or creating very fine grained scorers such as detailed energy deposition maps. They are not, however, what a real scanner or detector outputs in routine operation.
In GATE simulations for medical imaging or dosimetry, hits are the natural bridge between pure particle transport and any model of how a device responds. For example, in a PET detector, hits in the crystal record how much energy was deposited where and when, before any modeling of scintillation light, photodetectors, or electronics. In a dosimetry calculation, an actor may use hit information to sum deposited energy in voxels and convert it into absorbed dose. In all these cases, hits remain tied to the physical interaction picture and are not yet interpreted as detector measurements.
Because hits are typically produced in large numbers, saving all hit information can lead to very large output files and slow analysis. For many applications, only processed information such as singles or coincidences is needed, so hits are used internally by digitizers and actors but not always written to disk. When you are learning GATE, it is useful to think of hits as a detailed, intermediate description of the interaction history, which can be transformed into more compact, measurement-like quantities later.
Hits represent physical interactions inside sensitive volumes: energy deposition, position, and time of particle interactions. They are not the same as what a real detector channel measures.
Electronic detector response
In contrast to hits, detector signals describe what a real detector system reports after converting energy deposition into electrical measurements and after applying its readout and processing chain. In GATE, this transformation from hits to detector signals is handled by the digitizer. The digitizer takes raw interaction information and produces higher level objects such as singles and coincidences that mimic the output of real acquisition systems.
The first conceptual step from hits to signals is energy and time integration. A real detector channel does not see each microscopic interaction separately. Instead, energy deposits that occur close in space and time are combined into a single detection event. In a scintillation detector, for instance, all energy deposited in one crystal within a short time window will produce a single burst of scintillation light, which leads to one electronic pulse. The digitizer reproduces this behavior by summing the energy of hits associated with the same detector element and within a defined time window. The resulting object is a detector event or single, which has a total energy, an effective position, and a detection time.
This stage also introduces detector imperfections that are absent from raw hits. Real detectors have finite energy and timing resolution, limited spatial resolution, and non zero detection thresholds. The digitizer models these effects using blurring and cuts. Energy blurring, often implemented as a Gaussian smearing of the summed energy, simulates the detector energy resolution. Time blurring simulates timing resolution. Spatial blurring can be used to reflect uncertainties in the reconstructed interaction position inside a detector element. Energy thresholds and windows suppress low energy deposits and select relevant photopeak events. All these modifications mean that a detector signal no longer directly equals the sum of physical energy deposits, but instead reflects what a realistic measurement would look like.
Another key difference is how detector identity is represented. While hits may only know about the volume in which they occurred, detector signals typically carry identifiers corresponding to actual readout channels, such as crystal ID, module ID, or ring ID. This mapping from geometry and volume hierarchy to detector channels is an integral part of the digitizer configuration. It allows output data to be organized in ways compatible with reconstruction software, scanner specifications, or experimental setups.
Because electronic detector response involves time windows and channel mapping, signals can also be combined across detectors. In PET, for example, singles from different crystals are paired in time to form coincidence events, which represent two detectors firing within a coincidence window. These coincidences are even further from the raw physical interactions and closer to the data used in image reconstruction. They contain detector IDs, energies after blurring and energy windowing, and detection times after timing resolution has been applied.
The distinction between hits and detector signals is crucial when you design simulations. If you want to study fundamental interaction processes or detailed energy deposition patterns inside a detector or phantom, you focus on hits and actors that record deposited energy or dose. If you aim to model what an imaging scanner or radiation detector would actually output, you configure digitizers and analyze singles and coincidences. In practice, both levels are needed at different stages. For example, you might validate detector design at the hit level, then check imaging performance at the singles and coincidence level.
In summary, physical hits describe what happens inside materials at the Monte Carlo level, while detector signals describe what a readout system measures after integration, blurring, thresholds, and channel mapping. Understanding and controlling this transformation in GATE helps you align your simulations with real experimental data and with the performance metrics that matter in medical imaging and radiation detection.
Detector signals (such as singles and coincidences) are processed data: they result from summing hits, applying blurring and thresholds, and mapping interactions to detector channels. They represent what the detector actually measures, not the raw physical interactions.
Views: 9