Adaptive-weighted total variation minimization for sparse data toward low-dose x-ray computed tomography image...

Area of Science:

• Medical Imaging
• Image Reconstruction
• Computational Imaging

Background:

• Piecewise-smooth image reconstruction from sparse-view CT data often uses total variation (TV) minimization.
• Conventional TV minimization can lead to over-smoothing of image edges due to its piecewise constant assumption.

Purpose of the Study:

• To develop an adaptive-weighted total variation (AwTV) minimization algorithm for improved sparse-view low-dose CT image reconstruction.
• To enhance edge detail preservation and mitigate over-smoothing artifacts in reconstructed CT images.

Main Methods:

• An adaptive-weighted total variation (AwTV) model was developed, incorporating anisotropic edge properties and adaptive weights based on local image intensity gradients.
• An AwTV-POCS (projection onto convex sets) algorithm was implemented for minimizing AwTV under data and constraint conditions.
• Qualitative and quantitative evaluations were conducted using computer simulations and phantom experiments.

Main Results:

• The AwTV-POCS algorithm demonstrated notable gains in noise-resolution tradeoff plots compared to conventional TV-POCS.
• Improved full-width at half-maximum (FWHM) values were observed, indicating better edge preservation.
• The proposed method effectively reconstructs images from sparse-view low-dose CT data with enhanced details.

Conclusions:

• The adaptive-weighted total variation (AwTV) minimization algorithm offers significant improvements for sparse-view low-dose CT reconstruction.
• AwTV-POCS effectively preserves edge details and enhances image quality over traditional TV minimization techniques.
• This approach represents a promising advancement in medical image reconstruction for challenging acquisition scenarios.