Low Dose CT Image Reconstruction With Learned Sparsifying Transform

TL;DR

This paper introduces a novel CT image reconstruction method that leverages learned sparsifying transforms to improve image quality at ultra-low X-ray doses, outperforming traditional regularization techniques.

Contribution

The paper presents a new PWLS-based CT reconstruction approach using learned sparsifying transforms, optimized with an accelerated algorithm for better low-dose imaging.

Findings

PWLS-ST significantly improves image quality at low doses.

The method outperforms PWLS-EP in numerical experiments.

Accelerated optimization reduces computational time.

Abstract

A major challenge in computed tomography (CT) is to reduce X-ray dose to a low or even ultra-low level while maintaining the high quality of reconstructed images. We propose a new method for CT reconstruction that combines penalized weighted-least squares reconstruction (PWLS) with regularization based on a sparsifying transform (PWLS-ST) learned from a dataset of numerous CT images. We adopt an alternating algorithm to optimize the PWLS-ST cost function that alternates between a CT image update step and a sparse coding step. We adopt a relaxed linearized augmented Lagrangian method with ordered-subsets (relaxed OS-LALM) to accelerate the CT image update step by reducing the number of forward and backward projections. Numerical experiments on the XCAT phantom show that for low dose levels, the proposed PWLS-ST method dramatically improves the quality of reconstructed images compared to PWLS reconstruction with a nonadaptive edge-preserving regularizer (PWLS-EP).