Simulation of a Compton-based detector for low-dose high-resolution time-of-flight positron emission tomography

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Detector Physics

Background:

• Time-of-flight positron emission tomography (TOF-PET) faces challenges in spatial resolution and patient radiation dose due to detection technology limitations.
• Existing TOF-PET systems are constrained by detector capabilities, hindering further advancements in image quality and safety.

Approach:

• A Monte Carlo simulation using TOPAS Geant4 was developed to evaluate a novel low-atomic number (low-Z) detector concept.
• The simulation focused on linear alkylbenzene (LAB) doped with a molecular recorder, assessing its performance for next-generation TOF-PET.

Key Points:

• The simulation quantified tradeoffs in energy, spatial, and timing resolution for the proposed detector design.
• A likelihood-based method identified 87.1% of first interaction pairs with minimal error and rejected 90% of in-patient scatters.
• Achieved TOF-PET sensitivity of ~66.7% and a 4.6 mm point spread function (PSF) width with clear contrast.

Conclusions:

• The proposed low-Z detector, with optimized specifications, significantly enhances TOF-PET performance.
• Simulations demonstrate the potential for clear brain phantom imaging at less than 1% of a standard radiotracer dose.
• This technology offers a promising path towards safer and more effective TOF-PET imaging.