Search for a command to run...
Dual-energy CT is known to enable possible improvement of material separation over regular CT. However, in clinical implementation most of the dual-energy techniques show limited results mainly due to their sensitivity to noisy data. We simulate data acquisition by a dual-layer CT based on two scintillation layers one on top of the other with which the data is acquired simultaneously. We map the results of the reconstruction into a plane created from the Hounsfield units (HU) of the upper-layer image versus the HU of the lower-layer image. We find that different scanned materials end up in different definable regions in the HU-plane. Application of a special correction on the reconstructed images achieves stability on the HU-plane despite beam-hardening effects. In order to assess the practical material separation capabilities, part of the simulations were done with exact noise calculations. We analyze the material separation capabilities with such a configuration and conclude that the combination of the dual-layer CT with the classification analysis in the HU-plane is a practical and robust method that significantly improves clinical applications, in particular those involving iodine-calcium separation such as analysis and classification of coronary artery calcifications and soft plaques.