Abstract
BACKGROUND: This study aimed to analyse the quantitative capabilities of cadmium-zinc-telluride (CZT)-based SPECT-CT cameras using (99m)Tc, comparable to the analysis performed a decade ago for the sodium iodide (NaI) SPECT-CT systems available on the market at that time. This survey assessed one dual-head (GE Discovery NM870 CZT) and two ring (GE Starguide, Spectrum Dynamics Veriton 200) CZT cameras, as well as a state-of-the-art NaI dual-head system (Siemens Intevo Bold) that served as reference. Attenuation and scatter correction accuracy was explored, contrast recovery for small cold and hot rods measured, as well as the quantification in a large uniform area using user-determined conversion factors. Tomography reconstruction was performed with the manufacturers' iterative algorithms that allowed for attenuation correction, scatter correction and resolution recovery. RESULTS: Using the NEMA NU-2 1994 dedicated phantom, attenuation and scatter corrections seemed to perform very well. Equally, the contrast recovery of cold rods seemed to be superior for the CZT systems. However, the contrast recovery for the hot rods was inferior to the NaI camera, whereas it was comparable without the scatter correction. Finally, a quantification error of less than 5% was shown to be reachable when using adequate user-determined conversion factors. For the NaI camera, all results were similar to those of the past study. CONCLUSIONS: Without scatter correction, the CZT SPECT systems showed contrast performance similar to the NaI camera. With scatter correction, this held true for cold objects but the contrast of hot objects was not significantly improved or was degraded depending on the system considered and the object size. Quantification errors of less than 5% were achievable. It is hoped that on-going developments at both manufacturers will improve the scatter correction accuracy.