Multi-limited-angle spectral CT image reconstruction based on average image induced relative total variation model

Area of Science:

• Medical Imaging
• Computed Tomography
• Image Reconstruction

Background:

• Spectral computed tomography (CT) is gaining attention for its advanced imaging capabilities.
• Traditional spectral CT requires full-angular scanning, which can be time-consuming.
• Developing faster and more cost-effective reconstruction algorithms is crucial.

Purpose of the Study:

• To develop a low-cost and fast energy spectral CT reconstruction algorithm.
• To implement multi-limited-angle scanning for accelerated data acquisition.
• To improve image quality and reduce artifacts in spectral CT.

Main Methods:

• Simulated multi-source spectral CT using a dual X-ray source/detector system.
• Employed multi-limited-angle scanning across energy channels to enhance speed.
• Proposed an average image induced relative total variation (Aii-RTV) model for reconstruction.
• Utilized an iterative algorithm incorporating weighted average projection data and windowing total variation.

Main Results:

• The Aii-RTV algorithm effectively suppresses limited-angle artifacts in spectral CT images.
• Quantitative analysis showed significant improvements in peak signal-to-noise ratio (PSNR).
• Reconstruction results demonstrated superior performance compared to prior image constrained compressed sensing (PICCS) and RTV methods.

Conclusions:

• The proposed Aii-RTV algorithm offers a promising solution for fast and high-quality spectral CT reconstruction.
• Multi-limited-angle scanning combined with the Aii-RTV model enhances efficiency and image fidelity.
• The study highlights the importance of parameter selection for optimal regularization and reconstruction outcomes.