Issue 4, 2023

A scalable neural network architecture for self-supervised tomographic image reconstruction

Abstract

We present a lightweight and scalable artificial neural network architecture which is used to reconstruct a tomographic image from a given sinogram. A self-supervised learning approach is used where the network iteratively generates an image that is then converted into a sinogram using the Radon transform; this new sinogram is then compared with the sinogram from the experimental dataset using a combined mean absolute error and structural similarity index measure loss function to update the weights of the network accordingly. We demonstrate that the network is able to reconstruct images that are larger than 1024 × 1024. Furthermore, it is shown that the new network is able to reconstruct images of higher quality than conventional reconstruction algorithms, such as the filtered back projection and iterative algorithms (SART, SIRT, CGLS), when sinograms with angular undersampling are used. The network is tested with simulated data as well as experimental synchrotron X-ray micro-tomography and X-ray diffraction computed tomography data.

Graphical abstract: A scalable neural network architecture for self-supervised tomographic image reconstruction

Supplementary files

Article information

Article type
Paper
Submitted
01 Oct 2022
Accepted
30 May 2023
First published
02 Jun 2023
This article is Open Access
Creative Commons BY license

Digital Discovery, 2023,2, 967-980

A scalable neural network architecture for self-supervised tomographic image reconstruction

H. Dong, S. D. M. Jacques, W. Kockelmann, S. W. T. Price, R. Emberson, D. Matras, Y. Odarchenko, V. Middelkoop, A. Giokaris, O. Gutowski, A. Dippel, M. von Zimmermann, A. M. Beale, K. T. Butler and A. Vamvakeros, Digital Discovery, 2023, 2, 967 DOI: 10.1039/D2DD00105E

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. You can use material from this article in other publications without requesting further permissions from the RSC, provided that the correct acknowledgement is given.

Read more about how to correctly acknowledge RSC content.

Social activity

Spotlight

Advertisements