Sparse-view X-ray CT based on a box-constrained nonlinear weighted anisotropic TV regularization.

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Imaging

Background:

• Sparse-view computed tomography (CT) reduces radiation exposure but causes streaking artifacts due to undersampling.
• Reconstructing meaningful images from limited projections is an ill-posed inverse problem.
• Existing methods often rely on regularization to mitigate artifacts and improve image quality.

Purpose of the Study:

• To develop a novel regularization technique for sparse-view CT image reconstruction.
• To improve the speed and quality of CT image reconstruction from limited X-ray projections.
• To address the challenge of severe artifacts in sparse-view CT imaging.

Main Methods:

• Proposed a box-constrained nonlinear weighted anisotropic total variation (box-constrained NWATV) regularization method.
• Utilized an alternative direction method of multipliers (ADMM) to minimize the energy functional.
• Validated the method using phantom models, walnut X-ray projections, and human lung images.

Main Results:

• The proposed box-constrained NWATV method significantly accelerates image reconstruction compared to existing L1/L2 regularization.
• Central Processing Unit (CPU) time was reduced by more than 8 times.
• Demonstrated effective artifact reduction and image quality improvement in sparse-view CT.

Conclusions:

• The proposed box-constrained NWATV regularization offers a faster and effective solution for sparse-view CT image reconstruction.
• This method holds promise for reducing radiation dose in medical imaging while maintaining diagnostic image quality.
• The ADMM-based minimization approach enhances computational efficiency.