A sequential regularization based image reconstruction method for limited-angle spectral CT

Area of Science:

• Medical Imaging
• Computational Imaging
• Image Reconstruction

Background:

• Spectral computed tomography (CT) utilizes multiple x-ray spectra for enhanced material decomposition and image correction.
• Conventional spectral CT reconstruction methods suffer from artifacts and decomposition errors when using limited-angle projection data.
• Limited-angle data acquisition is a practical constraint in various spectral CT applications.

Purpose of the Study:

• To develop a novel reconstruction model for limited-angle spectral CT.
• To address artifacts and decomposition errors inherent in limited-angle spectral CT data.
• To improve the accuracy and quality of reconstructed images from limited angular data.

Main Methods:

• A sequential regularization-based reconstruction model was proposed for limited-angle spectral CT.
• A numerical solver was developed to implement the proposed reconstruction model.
• The method was validated using both simulated and real experimental data.

Main Results:

• The proposed method effectively suppresses limited-angle related artifacts in spectral CT images.
• Edge preservation is maintained, leading to sharper reconstructed images.
• Basis image decomposition errors are significantly reduced compared to conventional methods.

Conclusions:

• The sequential regularization-based model offers a robust solution for limited-angle spectral CT reconstruction.
• The developed numerical solver enables practical implementation and validation of the method.
• This approach enhances the diagnostic utility of spectral CT in scenarios with limited angular data acquisition.