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

Area of Science:

• Medical Imaging
• Photon-Counting Computed Tomography (PCCT)

Background:

• Photon-counting CT (PCCT) offers simultaneous spatial and spectral imaging for enhanced material discrimination.
• Geometric calibration inaccuracies in PCCT can lead to reconstruction distortions, compromising image quality, spectral fidelity, and quantitative material analysis.

Purpose of the Study:

• Investigate the impact of scanner-induced geometric distortions on material quantification using a Medipix All Resolution System scanner.
• Examine the relationship between spatial image degradation and spectral signal integrity in PCCT.

Main Methods:

• Systematically perturbed geometric reconstruction parameters to create intentionally distorted PCCT reconstructions.
• Analyzed distortion patterns (edge blurring, shading, ring artifacts) and their effects on spectral signatures and material reconstruction.
• Developed and applied a comprehensive geometric calibration framework to mitigate distortions and enhance spectral consistency.

Main Results:

• Geometric perturbations caused significant spatial distortions and measurable spectral response degradation.
• Distorted reconstructions exhibited reduced material discrimination accuracy and altered spectral signatures.
• The proposed calibration approach effectively minimized reconstruction artifacts, improving spatial image quality and spectral reliability.

Conclusions:

• Spatial and spectral degradations in PCCT are intrinsically linked.
• Accurate geometric calibration is essential for precise material quantification in PCCT.
• Optimal spectral reconstruction performance in PCCT relies on robust geometric calibration.