oa GATE Simulation of Philips TF PET Scanner

Positron Emission Tomography (PET) is a noninvasive imaging technique used for the diagnosis and assessment of many diseases, particularly cancer. It relies on positron emitting radioisotopes to analyze the tissues and organs functions. A PET scanner consists of a set of detectors surrounding the patient that will detect coincident gamma annihilation photons originating from the β^+ decay of the radiopharmaceutical injected into the patient and thus creating tomographic images.

Geant4 Application for Tomography Emission (GATE) is a Monte Carlo based simulation platform developed by the OpenGATE collaboration and used in the field of medical imaging. Monte Carlo methods are useful in the field of radiation medicine because of the stochastic nature of the processes involving radiation. GATE enables the modeling of scanners based on emission tomography, in particular the PET scanner. The GATE software consists of defining the geometry of the scanner, the shielding layers, the characteristics of the crystal detector and the radioactive source, the phantom where this latter is encapsulated as well as the physics processes taking place. Then, the simulation is carried out to identify the main performance parameters of the scanner and compare them to the experimental values.

The scanner modeled in this work is the Gemini TF PET/CT (Philips Medical Systems). It is being simulated using GATE and the results of sensitivity (S), scatter fraction (SF) and spatial resolution (SR) are being studied and compared to the published measurements.

The sensitivity of the scanner represents its ability to detect coincident photons emitted from inside the Field Of View (FOV) of the scanner. It is defined as the number of counts per unit time of true coincident event for a given source strength.

The scatter fraction is a measure of the system's sensitivity to scatter. The scattering of one or both of the gamma rays due to the interaction with the surrounding tissue results in falsely located coincidence events. Therefore, it is important to determine the ratio of scattered counts to the total of scattered and true counts. This ratio is called the Scatter Fraction.

The Spatial resolution of a PET scanner represents its ability to reproduce the image of an object while clearly showing the variations in the distribution of radioactivity. It is defined as the minimum distance between two points in an image that the scanner can detect.

In order to study each one of these performance parameters, dedicated phantoms that are proposed by the National Electrical Manufacturers Association (NEMA) were used in our simulation. The NEMA protocol is widely accepted as methodology for the assessment of the performance of PET systems. The radionuclide used for each type of simulation as well as the source distribution were chosen according to NEMA standards as well.

The simulated data were analyzed using ROOT, a data statistical analysis framework written in C??. Image reconstruction follows the data processing, and tomographic images are created through traditional filtered back-projection, or through an iterative series of back and forward-projection steps.

All simulations and analyses presented in this work were carried out using the Texas A&M at Qatar High Performance Cluster (HPC). We will report on the results of the sensitivity, scatter fraction and spatial resolution of the scanner described.