Spatial resolution characterization of a clinical Photon Counting Computed Tomography

Abstract Computed Tomography (CT) is a cornerstone of diagnostic imaging but conventional systems, equipped with energy-integrating detectors (EIDs), present limitations in spatial resolution, noise, and tissue contrast. Photon-Counting CT (PCCT) addresses these limitations using photon-counting detectors (PCDs) that directly convert incoming X-ray photons into electric signals, enabling energy discrimination, material quantification, and virtual monoenergetic image reconstruction. Clinical applications have already included cardiovascular, thoracic, and neuroimaging examinations. We performed a characterization of the spatial resolution of the first clinical PCCT scanner, the Naeotom Alpha® (Siemens Healthineers), equipped with CdTe PCDs. High-contrast presampled Modulation Transfer Function (MTF) was measured using a custom-built tungsten wire phantom tilted by 3 ∘ C. Spatial resolution in radial and tangential was evaluated by shifting the phantom within the field of view. Three acquisition protocols were tested (head, abdomen/thorax, inner ear), including standard and Ultra-High Resolution (UHR) modes. Images were reconstructed with multiple iterative algorithms and kernels, and MTF analysis was automated via Python. Results showed that spatial resolution increases with sharper reconstruction kernels across all anatomical protocols, with negligible differences between standard and vascular algorithms. UHR mode significantly improved resolution in the inner ear protocol (f 50% from 0.56 to 0.73 mm -1 ; f 10% from 0.87 to 1.09 mm -1 ). In conclusion, this work provides an initial assessment of PCCT spatial resolution for acquisition protocols and reconstructions usually adopted in clinical routine. Future investigations will employ standardized phantoms and comparisons with other CT systems to further validate performance and assess its potential to enhance diagnostic imaging quality.