dc.contributor.author
Bressem, Keno K.
dc.contributor.author
Adams, Lisa C.
dc.contributor.author
Erxleben, Christoph
dc.contributor.author
Hamm, Bernd
dc.contributor.author
Niehues, Stefan M.
dc.contributor.author
Vahldiek, Janis L.
dc.date.accessioned
2020-09-21T08:49:09Z
dc.date.available
2020-09-21T08:49:09Z
dc.identifier.uri
https://refubium.fu-berlin.de/handle/fub188/28261
dc.identifier.uri
http://dx.doi.org/10.17169/refubium-28011
dc.description.abstract
Chest radiographs are among the most frequently acquired images in radiology and are often the subject of computer vision research. However, most of the models used to classify chest radiographs are derived from openly available deep neural networks, trained on large image datasets. These datasets differ from chest radiographs in that they are mostly color images and have substantially more labels. Therefore, very deep convolutional neural networks (CNN) designed for ImageNet and often representing more complex relationships, might not be required for the comparably simpler task of classifying medical image data. Sixteen different architectures of CNN were compared regarding the classification performance on two openly available datasets, the CheXpert and COVID-19 Image Data Collection. Areas under the receiver operating characteristics curves (AUROC) between 0.83 and 0.89 could be achieved on the CheXpert dataset. On the COVID-19 Image Data Collection, all models showed an excellent ability to detect COVID-19 and non-COVID pneumonia with AUROC values between 0.983 and 0.998. It could be observed, that more shallow networks may achieve results comparable to their deeper and more complex counterparts with shorter training times, enabling classification performances on medical image data close to the state-of-the-art methods even when using limited hardware.
en
dc.rights.uri
https://creativecommons.org/licenses/by/4.0/
dc.subject
Betacoronavirus
en
dc.subject
Coronavirus Infections
en
dc.subject
Deep Learning
en
dc.subject
Diagnosis, Computer-Assisted
en
dc.subject
Neural Networks, Computer
en
dc.subject
Pneumonia, Viral
en
dc.subject
Radiography, Thoracic
en
dc.subject
Sensitivity and Specificity
en
dc.subject.ddc
600 Technik, Medizin, angewandte Wissenschaften::610 Medizin und Gesundheit::610 Medizin und Gesundheit
dc.title
Comparing different deep learning architectures for classification of chest radiographs
dc.type
Wissenschaftlicher Artikel
dcterms.bibliographicCitation.articlenumber
13590
dcterms.bibliographicCitation.doi
10.1038/s41598-020-70479-z
dcterms.bibliographicCitation.journaltitle
Scientific Reports
dcterms.bibliographicCitation.originalpublishername
Nature Research
dcterms.bibliographicCitation.volume
10
refubium.affiliation
Charité - Universitätsmedizin Berlin
refubium.resourceType.isindependentpub
no
dcterms.accessRights.openaire
open access
dcterms.bibliographicCitation.pmid
32788602
dcterms.isPartOf.eissn
2045-2322