Iterative Reconstruction for Low-Dose CT using Deep Gradient Priors of Generative Model

TL;DR

This paper introduces a novel iterative reconstruction method for low-dose CT that leverages deep generative models to learn data gradients, improving noise reduction and detail preservation without manual priors.

Contribution

It integrates data consistency into a generative model for low-dose CT, using learned gradients from normal-dose images and iterative updates with stochastic gradient descent.

Findings

Effective noise reduction in low-dose CT images.

Preservation of image details compared to traditional methods.

Demonstrated superior performance in experimental comparisons.

Abstract

Dose reduction in computed tomography (CT) is essential for decreasing radiation risk in clinical applications. Iterative reconstruction is one of the most promising ways to compensate for the increased noise due to reduction of photon flux. Rather than most existing prior-driven algorithms that benefit from manually designed prior functions or supervised learning schemes, in this work we integrate the data-consistency as a conditional term into the iterative generative model for low-dose CT. At the stage of prior learning, the gradient of data density is directly learned from normal-dose CT images as a prior. Then at the iterative reconstruction stage, the stochastic gradient descent is employed to update the trained prior with annealed and conditional schemes. The distance between the reconstructed image and the manifold is minimized along with data fidelity during reconstruction. Experimental comparisons demonstrated the noise reduction and detail preservation abilities of the proposed method.