Monte Carlo simulation of digital photon counting PET

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Computational Physics

Background:

• Digital photon counting (DPC) PET systems offer advancements in sensitivity and resolution.
• Accurate simulation models are crucial for optimizing DPC PET system performance and data analysis.
• The Philips Vereos system utilizes silicon photomultiplier (SiPM) detectors, requiring specific modeling considerations.

Purpose of the Study:

• To develop and validate a GATE Monte Carlo model for the Philips Vereos DPC PET system.
• To assess the model's accuracy against experimental data using NEMA guidelines.
• To introduce a novel method for estimating intrinsic spatial resolution without image reconstruction.

Main Methods:

• Development of a GATE Monte Carlo model incorporating SiPM detectors for the Philips Vereos PET system.
• Experimental validation using NEMA-compliant measurements for count rates, scatter fraction, energy/timing resolution, sensitivity, and spatial resolution.
• Comparison of simulated and experimental data using listmode data to ensure independence from reconstruction algorithms.
• Implementation of a line of response-based method for intrinsic spatial resolution estimation.

Main Results:

• The validated Monte Carlo model showed good agreement with experimental data, with differences generally below 10% for clinically relevant activity concentrations.
• The proposed line of response-based method provided an accurate estimation of intrinsic spatial resolution.
• The model's performance was confirmed across various metrics including count rates, scatter fraction, energy/timing resolutions, and sensitivity.

Conclusions:

• The developed GATE Monte Carlo model for the Philips Vereos DPC PET system is validated and reliable.
• The model can be effectively utilized for various research applications, including the optimization of acquisition parameters and image reconstruction algorithms.
• The novel intrinsic spatial resolution estimation method offers a valuable tool for characterizing PET system performance.