Influence of Radionuclide Cascade Decay Characteristics on Non-Collinear Coincidence Imaging Performance: A...

This study investigated the influence of radionuclide cascade decay characteristics on the performance of a non-collinear cascade coincidence imaging modality using Monte Carlo simulations implemented in the GATE platform. To evaluate their imaging performance, four cascade gamma-emitting radionuclides, namely 43K, 73Se, 111In and 177Lu were simulated at the center of the field of view of a small-animal PET scanner. The scanner is equipped with dual-layer Cerium-doped Gadolinium Oxyorthosilicate (GSO:Ce) detectors. The effects of cascade gamma-ray energies, emission probabilities, and intermediate-state half-lives on cascade coincidence detection efficiency (CCDE), sensitivity, random coincidence contribution, noise performance, scatter fraction and spatial resolution were evaluated. Simulated intermediate-state half-lives were compared with reference data to validate the simulation framework. The results showed that the simulated intermediate-state half-lives were in good agreement with the reference values, confirming the validity of the simulation model. All radionuclides achieved identical isotropic spatial resolution of approximately 0.477 mm FWHM, indicating that spatial resolution is primarily determined by detector geometry and the reconstruction algorithm rather than radionuclide decay characteristics. In contrast, the remaining imaging performance metrics strongly depended on cascade decay properties and coincidence timing conditions. 111In exhibited the highest sensitivity (15,058.8 cps/MBq) and CCDE, but also produced the highest random-to-true coincidence ratio (24.6%) due to its relatively long intermediate-state half-life and wider coincidence timing window. 73Se achieved the highest noise-equivalent cascade count rate, while scatter fractions remained negligible for most radionuclides. Therefore, under idealized Monte Carlo simulation conditions, radionuclide cascade decay characteristics, energy window and coincidence timing window optimization play critical roles in determining the performance of non-collinear cascade coincidence imaging systems. The reported performance represents an ideal detector model and should be validated under realistic detector and experimental conditions.