Cargando…

Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications

The reconstruction of computed tomography (CT) images is an active area of research. Following the rise of deep learning methods, many data-driven models have been proposed in recent years. In this work, we present the results of a data challenge that we organized, bringing together algorithm expert...

Descripción completa

Detalles Bibliográficos
Autores principales: Leuschner, Johannes, Schmidt, Maximilian, Ganguly, Poulami Somanya, Andriiashen, Vladyslav, Coban, Sophia Bethany, Denker, Alexander, Bauer, Dominik, Hadjifaradji, Amir, Batenburg, Kees Joost, Maass, Peter, van Eijnatten, Maureen
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8321320/
https://www.ncbi.nlm.nih.gov/pubmed/34460700
http://dx.doi.org/10.3390/jimaging7030044
_version_ 1783730823652966400
author Leuschner, Johannes
Schmidt, Maximilian
Ganguly, Poulami Somanya
Andriiashen, Vladyslav
Coban, Sophia Bethany
Denker, Alexander
Bauer, Dominik
Hadjifaradji, Amir
Batenburg, Kees Joost
Maass, Peter
van Eijnatten, Maureen
author_facet Leuschner, Johannes
Schmidt, Maximilian
Ganguly, Poulami Somanya
Andriiashen, Vladyslav
Coban, Sophia Bethany
Denker, Alexander
Bauer, Dominik
Hadjifaradji, Amir
Batenburg, Kees Joost
Maass, Peter
van Eijnatten, Maureen
author_sort Leuschner, Johannes
collection PubMed
description The reconstruction of computed tomography (CT) images is an active area of research. Following the rise of deep learning methods, many data-driven models have been proposed in recent years. In this work, we present the results of a data challenge that we organized, bringing together algorithm experts from different institutes to jointly work on quantitative evaluation of several data-driven methods on two large, public datasets during a ten day sprint. We focus on two applications of CT, namely, low-dose CT and sparse-angle CT. This enables us to fairly compare different methods using standardized settings. As a general result, we observe that the deep learning-based methods are able to improve the reconstruction quality metrics in both CT applications while the top performing methods show only minor differences in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). We further discuss a number of other important criteria that should be taken into account when selecting a method, such as the availability of training data, the knowledge of the physical measurement model and the reconstruction speed.
format Online
Article
Text
id pubmed-8321320
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-83213202021-08-26 Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications Leuschner, Johannes Schmidt, Maximilian Ganguly, Poulami Somanya Andriiashen, Vladyslav Coban, Sophia Bethany Denker, Alexander Bauer, Dominik Hadjifaradji, Amir Batenburg, Kees Joost Maass, Peter van Eijnatten, Maureen J Imaging Article The reconstruction of computed tomography (CT) images is an active area of research. Following the rise of deep learning methods, many data-driven models have been proposed in recent years. In this work, we present the results of a data challenge that we organized, bringing together algorithm experts from different institutes to jointly work on quantitative evaluation of several data-driven methods on two large, public datasets during a ten day sprint. We focus on two applications of CT, namely, low-dose CT and sparse-angle CT. This enables us to fairly compare different methods using standardized settings. As a general result, we observe that the deep learning-based methods are able to improve the reconstruction quality metrics in both CT applications while the top performing methods show only minor differences in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). We further discuss a number of other important criteria that should be taken into account when selecting a method, such as the availability of training data, the knowledge of the physical measurement model and the reconstruction speed. MDPI 2021-03-02 /pmc/articles/PMC8321320/ /pubmed/34460700 http://dx.doi.org/10.3390/jimaging7030044 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) ).
spellingShingle Article
Leuschner, Johannes
Schmidt, Maximilian
Ganguly, Poulami Somanya
Andriiashen, Vladyslav
Coban, Sophia Bethany
Denker, Alexander
Bauer, Dominik
Hadjifaradji, Amir
Batenburg, Kees Joost
Maass, Peter
van Eijnatten, Maureen
Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title_full Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title_fullStr Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title_full_unstemmed Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title_short Quantitative Comparison of Deep Learning-Based Image Reconstruction Methods for Low-Dose and Sparse-Angle CT Applications
title_sort quantitative comparison of deep learning-based image reconstruction methods for low-dose and sparse-angle ct applications
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8321320/
https://www.ncbi.nlm.nih.gov/pubmed/34460700
http://dx.doi.org/10.3390/jimaging7030044
work_keys_str_mv AT leuschnerjohannes quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT schmidtmaximilian quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT gangulypoulamisomanya quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT andriiashenvladyslav quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT cobansophiabethany quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT denkeralexander quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT bauerdominik quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT hadjifaradjiamir quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT batenburgkeesjoost quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT maasspeter quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications
AT vaneijnattenmaureen quantitativecomparisonofdeeplearningbasedimagereconstructionmethodsforlowdoseandsparseanglectapplications