A CT reconstruction approach from sparse projection with adaptive-weighted diagonal total-variation in biomedical...

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Imaging

Background:

• Traditional Total Variation (TV) methods lack directivity, limiting edge detail preservation in sparse-representation-based image reconstruction.
• Few-view CT reconstruction poses challenges due to insufficient data, often resulting in artifacts and poor image quality.

Purpose of the Study:

• To introduce an Adaptive-weighted Diagonal Total Variation (AwDTV) method for enhanced few-view CT image reconstruction.
• To improve the preservation of edge details and overall image quality in reconstructed CT images.

Main Methods:

• Developed AwDTV incorporating diagonal gradients and adaptive weights based on local image intensity.
• Employed the steepest descent method to solve the optimization problem for image reconstruction.
• Validated the algorithm through numerical simulations using few-view projections.

Main Results:

• The proposed AwDTV algorithm successfully reconstructed high-quality CT images from limited projection data.
• AwDTV demonstrated superior performance over Algebraic Reconstruction Technique (ART) and standard TV methods.
• Quantitative analysis showed lower Root Mean Square Error (RMSE) and higher Universal Quality Index (UQI) for AwDTV reconstructions.

Conclusions:

• AwDTV effectively addresses the limitations of TV methods in few-view CT reconstruction.
• The adaptive weighting strategy in AwDTV significantly enhances the preservation of fine image details and edges.
• AwDTV offers a promising approach for improving the diagnostic accuracy of low-dose or limited-angle CT scans.