A new adaptive-weighted total variation sparse-view computed tomography image reconstruction with local improved...

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Imaging

Background:

• Compressed sensing (CS) theory enhances sparse-view computed tomography (CT) image quality by incorporating prior information.
• Total variation (TV) algorithms leverage image sparsity and edge direction for improved sparse-view reconstruction, preserving critical image edges.

Purpose of the Study:

• To introduce a novel adaptive-weighted total variation (NAWTV) algorithm for sparse-view CT image reconstruction.
• To enhance image quality by considering local gradient direction continuity and anisotropic edge properties.

Main Methods:

• The NAWTV algorithm consolidates image sparsity using anisotropic edge properties, with weights as a combination of exponential and cosine functions.
• Weights are adaptively adjusted based on local image intensity gradients.
• Numerical implementation utilized the gradient descent method with Shepp-Logan and FORBILD head phantoms.

Main Results:

• The NAWTV algorithm demonstrated effectiveness and feasibility in sparse-view CT image reconstruction.
• Comparative experiments against TV and AwTV algorithms showed superior performance.
• The NAWTV algorithm successfully suppressed artifacts and preserved edge structure details.

Conclusions:

• The proposed NAWTV algorithm offers a significant advancement in sparse-view CT image reconstruction.
• It provides a robust method for improving image quality, particularly in preserving fine details and reducing noise.
• NAWTV is a promising technique for clinical applications requiring high-quality CT images from limited data.