Speaker
Description
This study presents the GATE-based Monte Carlo model and performance evaluation of a non-time-of-flight, collimator-free tomographic medical imaging system based on non-collinear cascade gamma-ray Coincidence (CGC) imaging. A CGC imaging model was developed to reconstruct three-dimensional decay vertices from valid coincidence events. A custom geometric back-projection reconstruction algorithm was implemented to generate tomographic images in transverse, coronal, and sagittal views, enabling quantitative assessment of spatial resolution, sensitivity, and coincidence detection efficiency (CDE). The results show, with 111In-ion point source at the center of field of view (FoV), the modeled imaging system achieved sub-millimeter isotropic spatial resolution of approximately 0.477 mm (FWHM) along all axes and demonstrated resolving capability between 1.5-2.0 mm for closely spaced point sources. The CGC imager achieved a coincidence efficiency of 1.50588 × 10⁻2 and sensitivity of 15,058.8 cps/MBq for a 111In source at the center of the FoV in air, their corresponding values for source in PMMA phantom were 1.25279 × 10-2 and 12,527.9 cps/MBq, respectively. These values were significantly higher than those reported for conventional parallel-hole, focused, and hybrid collimator-based systems, representing improvements of several orders of magnitude. For positional reconstruction, the analysis confirmed the reconstructed source position within ±16 mm transaxially and ±17 mm axially resembles to simulated position. To evaluate the impact of radionuclide decay characteristics on imaging performance, four cascade gamma emitters (43K, 73Se, 111In and 177Lu) were simulated under identical conditions. 111In and 73Se exhibit the highest CDEs, followed by 177Lu, while 43K shows the lowest due to its extremely short intermediate-state half-life of 46 ps, which is below the detector timing resolution (~0.549 ns), leading to missed valid coincidence events. The higher CDE for 111In is partly from random coincidences caused by its longer half-life and wider timing window. These results emphasize that detector timing resolution, isotope selection, and coincidence window design are critical for optimizing cascade gamma imaging performance. On the other hand, spatial resolution remained the same (~0.477 mm FWHM) across radionuclides. This finding confirms that system resolution is governed by detector geometry and reconstruction parameters rather than decay properties. The findings demonstrate that non-collinear CGC imaging enables the simultaneous achievement of both high sensitivity and sub-millimeter spatial resolution. However, experimental testing of the system should be conducted to assess its practical performance.
