Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review

<h3>Background</h3><p dir="ltr">Research on the diagnosis of COVID-19 using lung images is limited by the scarcity of imaging data. Generative adversarial networks (GANs) are popular for synthesis and data augmentation. GANs have been explored for data augmentation to enh...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Hazrat Ali (421019) (author)
مؤلفون آخرون: Zubair Shah (231886) (author)
منشور في: 2022
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513519116877824
author Hazrat Ali (421019)
author2 Zubair Shah (231886)
author2_role author
author_facet Hazrat Ali (421019)
Zubair Shah (231886)
author_role author
dc.creator.none.fl_str_mv Hazrat Ali (421019)
Zubair Shah (231886)
dc.date.none.fl_str_mv 2022-06-29T03:00:00Z
dc.identifier.none.fl_str_mv 10.2196/37365
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Combating_COVID-19_Using_Generative_Adversarial_Networks_and_Artificial_Intelligence_for_Medical_Images_Scoping_Review/25658880
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Health sciences
Health services and systems
augmentation
artificial intelligence
COVID-19
diagnosis
generative adversarial networks
diagnostic
lung image
imaging
data augmentation
X-ray
CT scan
data scarcity
image data
neural network
clinical informatics
dc.title.none.fl_str_mv Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <h3>Background</h3><p dir="ltr">Research on the diagnosis of COVID-19 using lung images is limited by the scarcity of imaging data. Generative adversarial networks (GANs) are popular for synthesis and data augmentation. GANs have been explored for data augmentation to enhance the performance of artificial intelligence (AI) methods for the diagnosis of COVID-19 within lung computed tomography (CT) and X-ray images. However, the role of GANs in overcoming data scarcity for COVID-19 is not well understood.</p><h3>Objective</h3><p dir="ltr">This review presents a comprehensive study on the role of GANs in addressing the challenges related to COVID-19 data scarcity and diagnosis. It is the first review that summarizes different GAN methods and lung imaging data sets for COVID-19. It attempts to answer the questions related to applications of GANs, popular GAN architectures, frequently used image modalities, and the availability of source code.</p><h3>Methods</h3><p dir="ltr">A search was conducted on 5 databases, namely PubMed, IEEEXplore, Association for Computing Machinery (ACM) Digital Library, Scopus, and Google Scholar. The search was conducted from October 11-13, 2021. The search was conducted using intervention keywords, such as “generative adversarial networks” and “GANs,” and application keywords, such as “COVID-19” and “coronavirus.” The review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines for systematic and scoping reviews. Only those studies were included that reported GAN-based methods for analyzing chest X-ray images, chest CT images, and chest ultrasound images. Any studies that used deep learning methods but did not use GANs were excluded. No restrictions were imposed on the country of publication, study design, or outcomes. Only those studies that were in English and were published from 2020 to 2022 were included. No studies before 2020 were included.</p><h3>Results</h3><p dir="ltr">This review included 57 full-text studies that reported the use of GANs for different applications in COVID-19 lung imaging data. Most of the studies (n=42, 74%) used GANs for data augmentation to enhance the performance of AI techniques for COVID-19 diagnosis. Other popular applications of GANs were segmentation of lungs and superresolution of lung images. The cycleGAN and the conditional GAN were the most commonly used architectures, used in 9 studies each. In addition, 29 (51%) studies used chest X-ray images, while 21 (37%) studies used CT images for the training of GANs. For the majority of the studies (n=47, 82%), the experiments were conducted and results were reported using publicly available data. A secondary evaluation of the results by radiologists/clinicians was reported by only 2 (4%) studies.</p><h3>Conclusions</h3><p dir="ltr">Studies have shown that GANs have great potential to address the data scarcity challenge for lung images in COVID-19. Data synthesized with GANs have been helpful to improve the training of the convolutional neural network (CNN) models trained for the diagnosis of COVID-19. In addition, GANs have also contributed to enhancing the CNNs’ performance through the superresolution of the images and segmentation. This review also identified key limitations of the potential transformation of GAN-based methods in clinical applications.</p><h2>Other Information</h2><p dir="ltr">Published in: JMIR Medical Informatics<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.2196/37365" target="_blank">https://dx.doi.org/10.2196/37365</a></p>
eu_rights_str_mv openAccess
id Manara2_fa2a8f00cf97385af6ee6f30afe87395
identifier_str_mv 10.2196/37365
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/25658880
publishDate 2022
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping ReviewHazrat Ali (421019)Zubair Shah (231886)Health sciencesHealth services and systemsaugmentationartificial intelligenceCOVID-19diagnosisgenerative adversarial networksdiagnosticlung imageimagingdata augmentationX-rayCT scandata scarcityimage dataneural networkclinical informatics<h3>Background</h3><p dir="ltr">Research on the diagnosis of COVID-19 using lung images is limited by the scarcity of imaging data. Generative adversarial networks (GANs) are popular for synthesis and data augmentation. GANs have been explored for data augmentation to enhance the performance of artificial intelligence (AI) methods for the diagnosis of COVID-19 within lung computed tomography (CT) and X-ray images. However, the role of GANs in overcoming data scarcity for COVID-19 is not well understood.</p><h3>Objective</h3><p dir="ltr">This review presents a comprehensive study on the role of GANs in addressing the challenges related to COVID-19 data scarcity and diagnosis. It is the first review that summarizes different GAN methods and lung imaging data sets for COVID-19. It attempts to answer the questions related to applications of GANs, popular GAN architectures, frequently used image modalities, and the availability of source code.</p><h3>Methods</h3><p dir="ltr">A search was conducted on 5 databases, namely PubMed, IEEEXplore, Association for Computing Machinery (ACM) Digital Library, Scopus, and Google Scholar. The search was conducted from October 11-13, 2021. The search was conducted using intervention keywords, such as “generative adversarial networks” and “GANs,” and application keywords, such as “COVID-19” and “coronavirus.” The review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines for systematic and scoping reviews. Only those studies were included that reported GAN-based methods for analyzing chest X-ray images, chest CT images, and chest ultrasound images. Any studies that used deep learning methods but did not use GANs were excluded. No restrictions were imposed on the country of publication, study design, or outcomes. Only those studies that were in English and were published from 2020 to 2022 were included. No studies before 2020 were included.</p><h3>Results</h3><p dir="ltr">This review included 57 full-text studies that reported the use of GANs for different applications in COVID-19 lung imaging data. Most of the studies (n=42, 74%) used GANs for data augmentation to enhance the performance of AI techniques for COVID-19 diagnosis. Other popular applications of GANs were segmentation of lungs and superresolution of lung images. The cycleGAN and the conditional GAN were the most commonly used architectures, used in 9 studies each. In addition, 29 (51%) studies used chest X-ray images, while 21 (37%) studies used CT images for the training of GANs. For the majority of the studies (n=47, 82%), the experiments were conducted and results were reported using publicly available data. A secondary evaluation of the results by radiologists/clinicians was reported by only 2 (4%) studies.</p><h3>Conclusions</h3><p dir="ltr">Studies have shown that GANs have great potential to address the data scarcity challenge for lung images in COVID-19. Data synthesized with GANs have been helpful to improve the training of the convolutional neural network (CNN) models trained for the diagnosis of COVID-19. In addition, GANs have also contributed to enhancing the CNNs’ performance through the superresolution of the images and segmentation. This review also identified key limitations of the potential transformation of GAN-based methods in clinical applications.</p><h2>Other Information</h2><p dir="ltr">Published in: JMIR Medical Informatics<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.2196/37365" target="_blank">https://dx.doi.org/10.2196/37365</a></p>2022-06-29T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.2196/37365https://figshare.com/articles/journal_contribution/Combating_COVID-19_Using_Generative_Adversarial_Networks_and_Artificial_Intelligence_for_Medical_Images_Scoping_Review/25658880CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/256588802022-06-29T03:00:00Z
spellingShingle Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
Hazrat Ali (421019)
Health sciences
Health services and systems
augmentation
artificial intelligence
COVID-19
diagnosis
generative adversarial networks
diagnostic
lung image
imaging
data augmentation
X-ray
CT scan
data scarcity
image data
neural network
clinical informatics
status_str publishedVersion
title Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
title_full Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
title_fullStr Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
title_full_unstemmed Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
title_short Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
title_sort Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review
topic Health sciences
Health services and systems
augmentation
artificial intelligence
COVID-19
diagnosis
generative adversarial networks
diagnostic
lung image
imaging
data augmentation
X-ray
CT scan
data scarcity
image data
neural network
clinical informatics