Energy-resolving x-ray detectors may enable producing iodine-specific images of the coronary arteries without the presence of motion artifacts. We refer to this approach as energy-resolved angiography (ERA), which uses basis material decomposition to produce iodine-specific images. We compared the theoretical iodine pixel signal-to- noise ratio (SNR) and the zero-frequency SNR of ERA with that of conventional digital subtraction angiography (DSA), the latter of which produces iodine-specific images by subtracting images acquired before and after iodine injection. For both ERA and DSA, we modeled iodine SNR with and without the response of realistic x-ray detectors. For ERA, we used a validated model of the energy response of a cadmium zinc telluride (CZT) spectroscopic x-ray detector to account for spectral degradation and spatio-energetic cross talk due to charge sharing. For DSA, we modeled the response of a cesium-iodine (CsI)-based detector and validated our model by comparison with published data. Incorporating a realistic energy response for spectroscopic x-ray detectors decreased the pixel SNR and zero-frequency SNR by greater than a factor of two. In the case of DSA, optical blur in the scintillator increased iodine SNR relative to ideal systems, a result attributable to reduced high-frequency noise in the presence of optical blur. Our results suggest that, for the same patient x-ray exposure, the pixel SNR and zero-frequency SNR of ERA will be ~1/6 and ~1/3 of that DSA, respectively.
|