Low-dose spectral CT reconstruction using L0 image gradient and tensor dictionary

TL;DR

This paper introduces L0TDL, a novel spectral CT reconstruction method combining tensor dictionary learning with L0-norm gradient constraints, significantly improving edge preservation and image quality in low-dose scenarios.

Contribution

The paper proposes an innovative tensor dictionary learning approach with L0-norm gradient constraint for enhanced low-dose spectral CT reconstruction, addressing edge preservation issues.

Findings

L0TDL outperforms TV, TV+LR, and TDL methods in simulations.

The method improves edge preservation and image quality.

Numerical and real mouse studies validate effectiveness.

Abstract

Spectral computed tomography (CT) has a great superiority in lesion detection, tissue characterization and material decomposition. To further extend its potential clinical applications, in this work, we propose an improved tensor dictionary learning method for low-dose spectral CT reconstruction with a constraint of image gradient L0-norm, which is named as L0TDL. The L0TDL method inherits the advantages of tensor dictionary learning (TDL) by employing the similarity of spectral CT images. On the other hand, by introducing the L0-norm constraint in gradient image domain, the proposed method emphasizes the spatial sparsity to overcome the weakness of TDL on preserving edge information. The alternative direction minimization method (ADMM) is employed to solve the proposed method. Both numerical simulations and real mouse studies are perform to evaluate the proposed method. The results show that the proposed L0TDL method outperforms other competing methods, such as total variation (TV) minimization, TV with low rank (TV+LR), and TDL methods.