A Model-based Deep Learning Reconstruction for X-ray CT

TL;DR

This paper introduces a model-based deep learning approach for low dose X-ray CT reconstruction that trains without ground-truth images, enabling fast, noise-reduced reconstructions in practical dental CT applications.

Contribution

It proposes a novel training method for deep learning-based CT reconstruction that does not require ground-truth images, simplifying the process and improving efficiency.

Findings

Significant noise reduction in dental CT images.

No iterative reconstruction needed after training.

Effective application of PWLS cost function in deep learning.

Abstract

Low dose CT is of great interest in these days. Dose reduction raises noise level in projections and decrease image quality in reconstructions. Model based image reconstruction can combine statistical noise model together with prior knowledge into an Bayesian optimization problem so that significantly reduce noise and artefacts. In this work, we propose a model-base deep learning for CT reconstruction so that a reconstruction network can be trained with no ground-truth images needed. Instead of minimizing cost function for each image, the network learns to minimize an ensemble cost function for the whole training set. No iteration will be needed for real data reconstruction using such a trained network. We experimented with a penalized weighted least-squares (PWLS) cost function for low dose CT reconstruction and tested on data from a practical dental CT. Very encouraging results with great noise reductions are obtained.