Provably Convergent Learned Inexact Descent Algorithm for Low-Dose CT Reconstruction

TL;DR

This paper introduces ELDA, a provably convergent neural network-based algorithm for low-dose CT reconstruction, combining interpretability, convergence guarantees, and improved image quality through novel features.

Contribution

The paper presents ELDA, a new neural network architecture with convergence guarantees and enhanced reconstruction quality for LDCT, integrating non-local features and regularization.

Findings

ELDA outperforms state-of-the-art methods like RED-CNN and Learned Primal-Dual.

ELDA achieves better reconstruction quality with only 19 layers.

ELDA demonstrates promising solution accuracy and parameter efficiency.

Abstract

We propose a provably convergent method, called Efficient Learned Descent Algorithm (ELDA), for low-dose CT (LDCT) reconstruction. ELDA is a highly interpretable neural network architecture with learned parameters and meanwhile retains convergence guarantee as classical optimization algorithms. To improve reconstruction quality, the proposed ELDA also employs a new non-local feature mapping and an associated regularizer. We compare ELDA with several state-of-the-art deep image methods, such as RED-CNN and Learned Primal-Dual, on a set of LDCT reconstruction problems. Numerical experiments demonstrate improvement of reconstruction quality using ELDA with merely 19 layers, suggesting the promising performance of ELDA in solution accuracy and parameter efficiency.