Investigating Image Distortion and Its Impact on Material Quantification in Photon-counting Computed Tomography

Photon-counting computed tomography (PCCT) enables simultaneous spatial and spectral imaging for improved material discrimination. However, inaccuracies in scanner geometric calibration can introduce reconstruction distortions that degrade both image quality and spectral fidelity, thereby affecting quantitative material analysis.

This study investigated the impact of scanner-induced geometric distortions on material quantification in the Medipix All Resolution System scanner and examined the relationship between spatial image degradation and spectral signal integrity in photon-counting CT.

Multiple phantoms were scanned using a geometrically calibrated system. Geometric reconstruction parameters were then systematically perturbed to generate intentionally distorted reconstructions and create a spectral image library for visual assessment. Distortion patterns, including edge blurring, shading artifacts, and ring artifacts, were analyzed alongside changes in spectral signatures and material reconstruction performance. A comprehensive geometric calibration framework was subsequently developed to mitigate spatial distortions and improve spectral consistency.

Geometric perturbations produced significant spatial distortions accompanied by measurable degradation in spectral response. Distorted reconstructions showed reduced material discrimination accuracy and altered spectral signatures. The proposed calibration approach effectively reduced reconstruction artifacts and improved both spatial image quality and spectral reliability.

Spatial and spectral degradations in photon-counting CT are intrinsically interconnected. Accurate geometric calibration is therefore essential for reliable material quantification and optimal spectral reconstruction performance in photon-counting CT systems.