Removing Ring Artefacts for Photon-Counting Detectors Using Neural Networks in Different Domains

Wei Fang,Liang Li,Zhiqiang Chen

doi:10.1109/access.2020.2977096

Wei Fang, Liang Li + Show 1 more

Open Access

https://doi.org/10.1109/access.2020.2977096

Copy DOI

Abstract

The development of energy-resolving photon-counting detectors provides a new approach for obtaining spectral information in computed tomography. However, the responses of different photon counting detector pixels can be inconsistent, which will always cause stripe artefacts in projection domain and concentric ring artefacts in image domain. Traditional ring artifacts processing methods are mostly based on averaging and filtering. In this paper, we propose to use deep learning methods for ring artifacts removal respectively in image domain, projection domain and the polar coordinate system. Besides, by incorporating reconstruction process into neural networks, we unite the information from image domain and projection domain for ring artifacts removal under the framework of deep learning for the first time. A traditional ring artifacts removal method, which is based on wavelet and Fourier transform, is implemented for comparison. Quantitative analysis is performed on simulation and experimental results and it shows that deep learning based methods are promising in solving the problem of non-uniformity correction for photon-counting detectors.

Highlights

Photon counting detectors (PCDs) are attracting more and more attention in the generation design of computed tomography systems [1]–[3]
We can see that the processed images of traditional WF methods remain a lot of low-frequency ring artefacts whether it is performed in projection domain or polar coordinate system
By judging from the error image, we can tell the slight difference between the deep learning methods that work respectively in image domain, projection domain, polar coordinates and the comprehensive model

Summary

Introduction

Photon counting detectors (PCDs) are attracting more and more attention in the generation design of computed tomography systems [1]–[3]. The associate editor coordinating the review of this manuscript and approving it for publication was Hengyong Yu. by applying optimal energy weighting functions, such as assigning a higher weight to low-energy photons for their carrying more attenuation information [4]; and (c) with the ability of acquiring the spectral information, images in multiple energy windows can be generated with one-shot spectral imaging using photon counting detectors (PCDs), enabling more efficient quantitative material identification [9]–[11]. Photon counting detectors (PCDs) have a promising potential to offer significant improvements to the existing CT imaging techniques and make completely new CT applications possible.

Objectives

Methods

Results

Discussion

Conclusion