Simulation study of a brain PET scanner using TOF-DOI detectors equipped with first interaction position detection

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Detector Physics

Background:

• Positron Emission Tomography (PET) is crucial for brain imaging.
• Improving spatial resolution and image quality in PET is an ongoing challenge.
• Time-of-Flight (TOF) and Depth-of-Interaction (DOI) detectors offer potential advancements.

Purpose of the Study:

• To evaluate a novel brain PET scanner utilizing four-layer independent read-out TOF-DOI detectors.
• To assess performance characteristics including spatial resolution, sensitivity, and count rate capability.
• To compare image reconstruction methods for optimizing performance.

Main Methods:

• Simulated a PET detector with LYSO:Ce scintillator arrays and multi-pixel photon counters using GATE (Geant4).
• Implemented and compared true-FIP, WTA, and FL-FIP interaction position detection methods.
• Evaluated performance using modified NEMA NU 2-2018 procedures and phantom studies.

Main Results:

• Achieved intrinsic spatial resolutions of 0.52 mm (0°) and 0.76 mm (30°) FWHM.
• Reconstructed images showed sub-millimeter FWHM spatial resolution across the field-of-view.
• Demonstrated a maximum true count rate of 207.6 kcps and sensitivity of 45.2 cps/kBq.
• TOF and DOI reconstructions significantly improved image quality; FL-FIP outperformed WTA.

Conclusions:

• The proposed brain PET scanner achieves sub-millimeter resolution and high image quality with TOF and DOI reconstruction.
• The FL-FIP method offers improved spatial resolution and image quality compared to WTA.
• This technology holds promise for clinical oncology research and advanced brain imaging.