Optimization-Based Image Reconstruction Regularized with Inter-Spectral Structural Similarity for Limited-Angle...

Area of Science:

• Medical Imaging
• Cone-Beam CT
• Dual-Energy Imaging

Background:

• Limited-angle (LA) dual-energy (DE) cone-beam CT (CBCT) offers fast, low-dose DE imaging potential on existing scanners.
• Clinical use is limited by challenges in reconstructing images from LA projections.
• Existing methods require spectral calibration or paired datasets, restricting their application.

Purpose of the Study:

• To develop a practical solution for image reconstruction in LA-DECT.
• To enable clinical applications of fast and low-dose DECT.

Main Methods:

• Integrated inter-spectral structural similarity regularization into iterative image reconstruction for LA-DECT.
• Enforced similarity between DE images to reduce LA artifacts.
• Evaluated the method using physical and digital phantoms.

Main Results:

• Successfully reduced limited-angle artifacts in reconstructed DECT images.
• Demonstrated efficacy in quantitative DECT imaging.
• Achieved accurate image reconstruction without spectral calibration or paired datasets.

Conclusions:

• The proposed method offers a practical solution for LA-DECT image reconstruction.
• It bypasses the need for X-ray spectra measurement or paired training data.
• Shows significant potential for clinical implementation of LA-DECT.