File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Detection of COVID-19 Using Deep Learning Algorithms on Chest Radiographs

TitleDetection of COVID-19 Using Deep Learning Algorithms on Chest Radiographs
Authors
KeywordsCOVID-19
chest radiographs
artificial intelligence
deep learning
pneumonia
Issue Date2020
PublisherLippincott Williams & Wilkins. The Journal's web site is located at http://www.thoracicimaging.com
Citation
Journal of Thoracic Imaging, 2020, v. 35 n. 6, p. 369-376 How to Cite?
AbstractPURPOSE: To evaluate the performance of a deep learning (DL) algorithm for the detection of COVID-19 on chest radiographs (CXR). MATERIALS AND METHODS: In this retrospective study, a DL model was trained on 112,120 CXR images with 14 labeled classifiers (ChestX-ray14) and fine-tuned using initial CXR on hospital admission of 509 patients, who had undergone COVID-19 reverse transcriptase-polymerase chain reaction (RT-PCR). The test set consisted of a CXR on presentation of 248 individuals suspected of COVID-19 pneumonia between February 16 and March 3, 2020 from 4 centers (72 RT-PCR positives and 176 RT-PCR negatives). The CXR were independently reviewed by 3 radiologists and using the DL algorithm. Diagnostic performance was compared with radiologists' performance and was assessed by area under the receiver operating characteristics (AUC). RESULTS: The median age of the subjects in the test set was 61 (interquartile range: 39 to 79) years (51% male). The DL algorithm achieved an AUC of 0.81, sensitivity of 0.85, and specificity of 0.72 in detecting COVID-19 using RT-PCR as the reference standard. On subgroup analyses, the model achieved an AUC of 0.79, sensitivity of 0.80, and specificity of 0.74 in detecting COVID-19 in patients presented with fever or respiratory systems and an AUC of 0.87, sensitivity of 0.85, and specificity of 0.81 in distinguishing COVID-19 from other forms of pneumonia. The algorithm significantly outperforms human readers (P<0.001 using DeLong test) with higher sensitivity (P=0.01 using McNemar test). CONCLUSIONS: A DL algorithm (COV19NET) for the detection of COVID-19 on chest radiographs can potentially be an effective tool in triaging patients, particularly in resource-stretched health-care systems.
Persistent Identifierhttp://hdl.handle.net/10722/294229
ISSN
2023 Impact Factor: 2.0
2023 SCImago Journal Rankings: 0.722
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorChiu, WHK-
dc.contributor.authorVardhanabhuti, V-
dc.contributor.authorPoplavskiy, D-
dc.contributor.authorYu, PLH-
dc.contributor.authorDu, R-
dc.contributor.authorYap, AYH-
dc.contributor.authorZhang, S-
dc.contributor.authorFong, HTA-
dc.contributor.authorChin, TWY-
dc.contributor.authorLee, JCY-
dc.contributor.authorLeung, ST-
dc.contributor.authorLo, CSY-
dc.contributor.authorLui, MMS-
dc.contributor.authorFang, BXH-
dc.contributor.authorNg, MY-
dc.contributor.authorKuo, MD-
dc.date.accessioned2020-11-23T08:28:16Z-
dc.date.available2020-11-23T08:28:16Z-
dc.date.issued2020-
dc.identifier.citationJournal of Thoracic Imaging, 2020, v. 35 n. 6, p. 369-376-
dc.identifier.issn0883-5993-
dc.identifier.urihttp://hdl.handle.net/10722/294229-
dc.description.abstractPURPOSE: To evaluate the performance of a deep learning (DL) algorithm for the detection of COVID-19 on chest radiographs (CXR). MATERIALS AND METHODS: In this retrospective study, a DL model was trained on 112,120 CXR images with 14 labeled classifiers (ChestX-ray14) and fine-tuned using initial CXR on hospital admission of 509 patients, who had undergone COVID-19 reverse transcriptase-polymerase chain reaction (RT-PCR). The test set consisted of a CXR on presentation of 248 individuals suspected of COVID-19 pneumonia between February 16 and March 3, 2020 from 4 centers (72 RT-PCR positives and 176 RT-PCR negatives). The CXR were independently reviewed by 3 radiologists and using the DL algorithm. Diagnostic performance was compared with radiologists' performance and was assessed by area under the receiver operating characteristics (AUC). RESULTS: The median age of the subjects in the test set was 61 (interquartile range: 39 to 79) years (51% male). The DL algorithm achieved an AUC of 0.81, sensitivity of 0.85, and specificity of 0.72 in detecting COVID-19 using RT-PCR as the reference standard. On subgroup analyses, the model achieved an AUC of 0.79, sensitivity of 0.80, and specificity of 0.74 in detecting COVID-19 in patients presented with fever or respiratory systems and an AUC of 0.87, sensitivity of 0.85, and specificity of 0.81 in distinguishing COVID-19 from other forms of pneumonia. The algorithm significantly outperforms human readers (P<0.001 using DeLong test) with higher sensitivity (P=0.01 using McNemar test). CONCLUSIONS: A DL algorithm (COV19NET) for the detection of COVID-19 on chest radiographs can potentially be an effective tool in triaging patients, particularly in resource-stretched health-care systems.-
dc.languageeng-
dc.publisherLippincott Williams & Wilkins. The Journal's web site is located at http://www.thoracicimaging.com-
dc.relation.ispartofJournal of Thoracic Imaging-
dc.subjectCOVID-19-
dc.subjectchest radiographs-
dc.subjectartificial intelligence-
dc.subjectdeep learning-
dc.subjectpneumonia-
dc.titleDetection of COVID-19 Using Deep Learning Algorithms on Chest Radiographs-
dc.typeArticle-
dc.identifier.emailChiu, WHK: kwhchiu@hku.hk-
dc.identifier.emailVardhanabhuti, V: varv@hku.hk-
dc.identifier.emailYu, PLH: plhyu@hku.hk-
dc.identifier.emailZhang, S: sailong@hku.hk-
dc.identifier.emailFong, HTA: ahtfong@hku.hk-
dc.identifier.emailNg, MY: myng2@hku.hk-
dc.identifier.authorityChiu, WHK=rp02074-
dc.identifier.authorityVardhanabhuti, V=rp01900-
dc.identifier.authorityYu, PLH=rp00835-
dc.identifier.authorityNg, MY=rp01976-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1097/rti.0000000000000559-
dc.identifier.pmid32969949-
dc.identifier.scopuseid_2-s2.0-85095600045-
dc.identifier.hkuros320301-
dc.identifier.volume35-
dc.identifier.issue6-
dc.identifier.spage369-
dc.identifier.epage376-
dc.identifier.isiWOS:000583403600010-
dc.publisher.placeUnited States-
dc.identifier.issnl0883-5993-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats