Simulation of ultra low-dose scans in quantum counting clinical CT

Contrary to conventional energy integrating detectors, electronics noise in quantum counting detectors (also frequently referred to as photon counting detectors) mainly affects the spectral resolution of the detector. There is almost no impact on the counting signal itself. This promises improved image quality due to image noise reduction in scans obtained from clinical computed tomography (CT) examinations with lowest X-ray tube currents or strongly attenuating obese patients. In most of these examinations, noise from the electronics dominates the image noise when using conventional detectors. Applying quantum counting detectors instead can improve image quality of ultra low-dose scans. This improvement may as well be used to reduce X-ray dose while maintaining image noise on the level of conventional detectors. To quantify these benefits, we have simulated sinograms of various slice scans of the human body, using the parametric 3D XCAT phantom (abdomen, shoulders) and a geometric DRASIM phantom (cranium). The simulation chain includes modeling the X-ray source, beam attenuation in the patient, and calculation of the detector response followed by data corrections and image reconstruction.We quantify the image noise in selected Region Of Interest (ROI) in the difference image of two scans that differ only in their image noise realization. Furthermore, we provide a direct comparison of image noise in energy integrating and quantum counting CT system concepts.