Noise-resilient approach for deep tomographic imaging

TL;DR

This paper introduces a deep learning-based reconstruction method for X-ray tomography that is highly resistant to noise, enabling effective imaging even with low photon counts without needing noisy training data.

Contribution

It presents a novel noise-resilient deep reconstruction algorithm that does not require noisy training examples, advancing low-photon tomographic imaging capabilities.

Findings

Strong noise resilience demonstrated

Effective low-photon imaging enabled

No need for noisy training data

Abstract

We propose a noise-resilient deep reconstruction algorithm for X-ray tomography. Our approach shows strong noise resilience without obtaining noisy training examples. The advantages of our framework may further enable low-photon tomographic imaging.