File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: NIA-Network: Towards improving lung CT infection detection for COVID-19 diagnosis

TitleNIA-Network: Towards improving lung CT infection detection for COVID-19 diagnosis
Authors
KeywordsCOVID-19 diagnosis
Semi-supervised learning
Adversarial learning
Network-in-Network
Instance normalization
Issue Date2021
PublisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/artmed
Citation
Artificial Intelligence in Medicine, 2021, v. 117, p. article no. 102082 How to Cite?
AbstractDuring pandemics (e.g., COVID-19) physicians have to focus on diagnosing and treating patients, which often results in that only a limited amount of labeled CT images is available. Although recent semi-supervised learning algorithms may alleviate the problem of annotation scarcity, limited real-world CT images still cause those algorithms producing inaccurate detection results, especially in real-world COVID-19 cases. Existing models often cannot detect the small infected regions in COVID-19 CT images, such a challenge implicitly causes that many patients with minor symptoms are misdiagnosed and develop more severe symptoms, causing a higher mortality. In this paper, we propose a new method to address this challenge. Not only can we detect severe cases, but also detect minor symptoms using real-world COVID-19 CT images in which the source domain only includes limited labeled CT images but the target domain has a lot of unlabeled CT images. Specifically, we adopt Network-in-Network and Instance Normalization to build a new module (we term it NI module) and extract discriminative representations from CT images from both source and target domains. A domain classifier is utilized to implement infected region adaptation from source domain to target domain in an Adversarial Learning manner, and learns domain-invariant region proposal network (RPN) in the Faster R-CNN model. We call our model NIA-Network (Network-in-Network, Instance Normalization and Adversarial Learning), and conduct extensive experiments on two COVID-19 datasets to validate our approach. The experimental results show that our model can effectively detect infected regions with different sizes and achieve the highest diagnostic accuracy compared with existing SOTA methods.
Persistent Identifierhttp://hdl.handle.net/10722/304008
ISSN
2023 Impact Factor: 6.1
2023 SCImago Journal Rankings: 1.723
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLi, W-
dc.contributor.authorChen, J-
dc.contributor.authorChen, P-
dc.contributor.authorYu, L-
dc.contributor.authorCui, X-
dc.contributor.authorLi, Y-
dc.contributor.authorCheng, F-
dc.contributor.authorOuyang, W-
dc.date.accessioned2021-09-23T08:53:57Z-
dc.date.available2021-09-23T08:53:57Z-
dc.date.issued2021-
dc.identifier.citationArtificial Intelligence in Medicine, 2021, v. 117, p. article no. 102082-
dc.identifier.issn0933-3657-
dc.identifier.urihttp://hdl.handle.net/10722/304008-
dc.description.abstractDuring pandemics (e.g., COVID-19) physicians have to focus on diagnosing and treating patients, which often results in that only a limited amount of labeled CT images is available. Although recent semi-supervised learning algorithms may alleviate the problem of annotation scarcity, limited real-world CT images still cause those algorithms producing inaccurate detection results, especially in real-world COVID-19 cases. Existing models often cannot detect the small infected regions in COVID-19 CT images, such a challenge implicitly causes that many patients with minor symptoms are misdiagnosed and develop more severe symptoms, causing a higher mortality. In this paper, we propose a new method to address this challenge. Not only can we detect severe cases, but also detect minor symptoms using real-world COVID-19 CT images in which the source domain only includes limited labeled CT images but the target domain has a lot of unlabeled CT images. Specifically, we adopt Network-in-Network and Instance Normalization to build a new module (we term it NI module) and extract discriminative representations from CT images from both source and target domains. A domain classifier is utilized to implement infected region adaptation from source domain to target domain in an Adversarial Learning manner, and learns domain-invariant region proposal network (RPN) in the Faster R-CNN model. We call our model NIA-Network (Network-in-Network, Instance Normalization and Adversarial Learning), and conduct extensive experiments on two COVID-19 datasets to validate our approach. The experimental results show that our model can effectively detect infected regions with different sizes and achieve the highest diagnostic accuracy compared with existing SOTA methods.-
dc.languageeng-
dc.publisherElsevier BV. The Journal's web site is located at http://www.elsevier.com/locate/artmed-
dc.relation.ispartofArtificial Intelligence in Medicine-
dc.subjectCOVID-19 diagnosis-
dc.subjectSemi-supervised learning-
dc.subjectAdversarial learning-
dc.subjectNetwork-in-Network-
dc.subjectInstance normalization-
dc.titleNIA-Network: Towards improving lung CT infection detection for COVID-19 diagnosis-
dc.typeArticle-
dc.identifier.emailYu, L: lqyu@hku.hk-
dc.identifier.authorityYu, L=rp02814-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.artmed.2021.102082-
dc.identifier.scopuseid_2-s2.0-85106430602-
dc.identifier.hkuros325075-
dc.identifier.volume117-
dc.identifier.spagearticle no. 102082-
dc.identifier.epagearticle no. 102082-
dc.identifier.isiWOS:000661230700006-
dc.publisher.placeNetherlands-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats