Post-processing method for the removal of mixed ring artifacts in CT images

Area of Science:

• Medical Imaging
• Image Processing
• Computational Science

Background:

• Ring artifacts degrade Computed Tomography (CT) image quality.
• Existing methods often fail to address both intensity-dependent and time-dependent artifacts.

Purpose of the Study:

• To propose a general post-processing method for effectively removing both intensity-dependent and time-dependent ring artifacts.
• To enhance the overall quality of CT images.

Main Methods:

• Transforming raw CT images to polar coordinates to visualize artifacts as stripes.
• Separating structure and texture components to isolate artifactual stripes.
• Extracting stripe artifacts using mean extraction and texture classification.
• Iteratively refining CT images by subtracting extracted artifacts.

Main Results:

• Demonstrated effective removal of both intensity-dependent and time-dependent ring artifacts.
• Preserved crucial image details and spatial resolution during artifact correction.
• Validated effectiveness on both simulated and real CT data.

Conclusions:

• The proposed method offers a robust solution for ring artifact reduction in CT imaging.
• The technique is adaptable and effective for advanced CT systems, including photon counting CT.