A limited-angle CT reconstruction method based on anisotropic TV minimization

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Imaging

Background:

• Compressed sensing (CS)-inspired methods are crucial for limited-angle computed tomography (CT) reconstruction.
• Current CS-based CT reconstruction relies on minimizing total variation (TV) subject to data consistency.
• Optimizing the balance between TV smoothing and data fidelity is key for high-quality CT images.

Purpose of the Study:

• To address the limitations of isotropic TV minimization in limited-angle CT.
• To introduce an anisotropic TV minimization method tailored for limited-angle CT challenges.
• To enhance image reconstruction quality in limited-angle CT scenarios.

Main Methods:

• Developed a compressed sensing (CS)-inspired reconstruction technique.
• Introduced an anisotropic total variation (TV) minimization approach.
• Evaluated the method using numerical simulations with phantom and real CT images.

Main Results:

• The anisotropic TV minimization method demonstrates superior performance compared to standard TV-based reconstruction.
• The proposed method effectively handles the angularly varying data consistency in limited-angle CT.
• Improved image quality was observed in simulations using both phantom and real CT data.

Conclusions:

• Anisotropic TV minimization is better suited for limited-angle CT than isotropic TV minimization.
• The developed method offers a significant advantage for reconstructing high-quality images from limited-angle CT data.
• This approach advances the field of medical image reconstruction for challenging CT acquisition geometries.