Design and Initial Evaluation of a Prototype Preclinical PET Scanner with Submillimeter Spatial Resolution for Highprecision Imaging

Le résumé fourni par la source

This study presents the design and initial evaluation of a prototype preclinical PET scanner optimized for sub-millimeter spatial resolution. The system features a dual-ring coaxial architecture with a ring diameter of 6.5 cm, where each ring consists of eight identical detector modules, each with an overall dimension of$26.8 \times 26.8 ~\text{mm}^{2}$. Each module integrates a$44 \times 44$array of lutetium-yttrium oxyorthosilicate (LYSO) scintillator crystals, with individual crystal dimensions of$0.52 \times 0.52 \times 5 ~\text{mm}^{3}$. Adjacent crystals are optically isolated using a 0.08 mm layer of Enhanced Specular Reflector (ESR) to minimize optical crosstalk and enhance spatial resolution. The LYSO crystal arrays were coupled with a novel epitaxial-quenching-resistor silicon photomultiplier array and interfaced with the PETsys TOFPET2 ASIC front-end electronics. Flood image assessments demonstrated clear identification of all$44 \times 44$crystal pixels, indicating excellent detector pixel discrimination. The peak absolute sensitivity was measured at 18.6$\text{kcps} / \text{MBq}$for an on-axis point source, and decreased slightly to$17.4 \text{kcps} / \text{MBq}$at a 10 mm off-axis position. Monte Carlo simulations conducted with the Geant4 Application for Tomographic Emission (GATE) further validated the system's imaging capabilities. Image reconstruction was carried out using a multi-GPU-accelerated list-mode fully corrected ordered subsets expectation maximization (OSEM) algorithm. Reconstructions of the NEMA NU-4 image quality phantom yielded high-fidelity images, while scans of the micro-Derenzo phantom demonstrated the system's ability to distinctly resolve 0.5 mm rods with an average peak-to-valley of 1.275. These findings validate the sub-millimeter spatial resolution capability of the proposed PET system and highlight its potential for high-precision preclinical imaging applications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.