DuDo-RAC: Dual-domain optimization for ring artifact correction in photon counting CT

Area of Science:

• Medical Imaging
• Computed Tomography
• Detector Physics

Background:

• Photon counting detectors (PCDs) exhibit inconsistent pixel responses, leading to low-frequency ring artifacts in CT images.
• Existing CT ring artifact correction methods struggle with low-frequency artifacts and may leave residual errors.
• Polynomial correction methods can be limited by coefficient inaccuracies and modeling limitations.

Purpose of the Study:

• To propose a novel dual-domain optimization method (DuDo-RAC) for correcting both high and low-frequency ring artifacts in PCD-CT images.
• To develop a method that is independent of spectral information and training data, enhancing its applicability across various energy thresholds.

Main Methods:

• Modeling inconsistent detector response using pixel-wise polynomial functions with coefficients determined by dual-domain optimization.
• Employing polar transformations and smoothing to address ring artifacts, which appear as stripes.
• Utilizing a multi-resolution gradient loss function for iterative optimization of polynomial correction coefficients.

Main Results:

• The proposed DuDo-RAC method effectively corrects high and low-frequency ring artifacts in PCD-CT images.
• The technique successfully preserves essential image structures and details during artifact correction.

Conclusions:

• DuDo-RAC provides an effective solution for ring artifact correction in photon counting detector CT imaging.
• The method demonstrates robust performance in mitigating artifacts while maintaining image quality.