Development of virtual monochromatic imaging technique with spectral CT based on a photon-counting detector

Virtual monochromatic imaging (VMI) in spectral computed tomography (CT) is able to provide a single set of images for routine diagnosis and improve the image quality by enhancing CNR and reducing beam-hardening artifacts. However, the conventional VMI techniques are limited to utilize multi-energy information because those models are based on dualenergy measurements. Such a drawback causes the improper reflection of material characteristics and the degradation of the VMI image quality. In this study, the triple-energy-based VMI model was proposed, and the model was validated by using a simulated spectral CT system with a photon-counting detector (PCD). The image quality was also evaluated by analyzing the contrast, noise, and contrast-to-noise ratio (CNR). The results showed that the triple-energy-based VMI images were optimized at 69 and 59 keV for minimizing the image noise and maximizing the CNR, respectively. The triple-energy-based VMI images at the optimal monochromatic energies improved the noise property and CNR compared to a single-energy CT image. Therefore, the triple-energy-based VMI model can provide improved CT images for routine diagnosis and increase the availability of spectral CT for clinical applications.