Image reconstruction method for incomplete CT projection based on self-guided image filtering

Area of Science:

• Medical Imaging
• Computational Imaging
• Image Reconstruction

Background:

• Incomplete computed tomography (CT) data acquisition is common in medical diagnosis and industrial testing due to radiation dose limits or other constraints.
• Reconstructing images from limited or incomplete projection data remains a significant challenge in CT imaging.

Purpose of the Study:

• To propose a novel image reconstruction model for few-view and segmental limited-angle (SLA) CT.
• To address the challenge of reconstructing CT images from incomplete projection data.

Main Methods:

• A new image reconstruction model incorporating a self-guided image filtering (SGIF) term was developed.
• The alternating direction method (ADM) was employed to solve the proposed model, termed the ADM-SGIF method.
• The core principle involves using the reconstructed image's structural features to guide the reconstruction process.

Main Results:

• The ADM-SGIF method demonstrated superior performance in preserving image structures and mitigating shading artifacts compared to existing methods.
• Experiments using digital phantoms and real CT data validated the effectiveness of the ADM-SGIF approach.
• Objective and subjective evaluations confirmed that ADM-SGIF outperformed Total Variation (TV), Relative Total Variation (RTV), and ADM-L0 methods.

Conclusions:

• The proposed ADM-SGIF method offers an effective solution for CT image reconstruction from incomplete projection data, particularly in few-view and SLA scenarios.
• The self-guided image filtering approach enhances structural preservation and artifact reduction in CT reconstruction.
• ADM-SGIF represents a significant advancement over conventional reconstruction techniques for limited-data CT imaging.