Diffusion Prior Regularized Iterative Reconstruction for Low-dose CT

TL;DR

This paper introduces a novel iterative reconstruction algorithm for low-dose CT that combines a diffusion prior with data fidelity, achieving high-quality images with reduced radiation exposure.

Contribution

It presents a new unsupervised reconstruction method using a diffusion prior and Nesterov acceleration, improving image quality in low-dose CT scans.

Findings

High-quality reconstructions with fewer steps

Effective reduction of radiation dose

Unsupervised learning framework

Abstract

Computed tomography (CT) involves a patient's exposure to ionizing radiation. To reduce the radiation dose, we can either lower the X-ray photon count or down-sample projection views. However, either of the ways often compromises image quality. To address this challenge, here we introduce an iterative reconstruction algorithm regularized by a diffusion prior. Drawing on the exceptional imaging prowess of the denoising diffusion probabilistic model (DDPM), we merge it with a reconstruction procedure that prioritizes data fidelity. This fusion capitalizes on the merits of both techniques, delivering exceptional reconstruction results in an unsupervised framework. To further enhance the efficiency of the reconstruction process, we incorporate the Nesterov momentum acceleration technique. This enhancement facilitates superior diffusion sampling in fewer steps. As demonstrated in our experiments, our method offers a potential pathway to high-definition CT image reconstruction with minimized radiation.