Mostrar el registro sencillo del documento
A comprehensive study on classifcation of COVID‐19 on computed tomography with pretrained convolutional neural networks
dc.contributor.advisor | Pham, Tuan D. | |
dc.date.accessioned | 2020-10-13T15:29:58Z | |
dc.date.available | 2020-10-13T15:29:58Z | |
dc.date.created | 2020 | |
dc.identifier.issn | 2045-2322 | spa |
dc.identifier.other | https://doi.org/10.1038/s41598-020-74164-z | spa |
dc.identifier.uri | http://hdl.handle.net/20.500.12010/14377 | |
dc.description.abstract | The use of imaging data has been reported to be useful for rapid diagnosis of COVID-19.Although computed tomography (CT) scans show a variety of signs caused by the viral infection, given a large amount of images, these visual features are difcult and can take a long time to be recognized by radiologists.Artifcial intelligence methods for automated classifcation of COVID-19 on CT scans have been found to be very promising. However, current investigation of pretrained convolutional neural networks (CNNs) for COVID-19 diagnosis using CT data is limited.This study presents an investigation on 16 pretrained CNNs for classifcation of COVID-19 using a large public database of CT scans collected from COVID-19 patients and non-COVID-19 subjects.The results show that, using only 6 epochs for training, the CNNs achieved very high performance on the classifcation task.Among the 16 CNNs, DenseNet-201, which is the deepest net, is the best in terms of accuracy, balance between sensitivity and specifcity, F1 score, and area under curve. Furthermore, the implementation of transfer learning with the direct input of whole image slices and without the use of data augmentation provided better classifcation rates than the use of data augmentation. Such a fnding alleviates the task of data augmentation and manual extraction of regions of interest on CT images, which are adopted by current implementation of deep-learning models for COVID-19 classifcation. | spa |
dc.format.extent | 8 páginas | spa |
dc.format.mimetype | application/pdf | spa |
dc.language.iso | eng | spa |
dc.publisher | Scientific reports | spa |
dc.source | reponame:Expeditio Repositorio Institucional UJTL | spa |
dc.source | instname:Universidad de Bogotá Jorge Tadeo Lozano | spa |
dc.subject | Comprehensive study on classifcation | spa |
dc.subject | COVID‑19 | spa |
dc.subject | Computed tomography | spa |
dc.title | A comprehensive study on classifcation of COVID‐19 on computed tomography with pretrained convolutional neural networks | spa |
dc.type.local | Artículo | spa |
dc.subject.lemb | Síndrome respiratorio agudo grave | spa |
dc.subject.lemb | COVID-19 | spa |
dc.subject.lemb | SARS-CoV-2 | spa |
dc.subject.lemb | Coronavirus | spa |
dc.rights.accessrights | info:eu-repo/semantics/openAccess | spa |
dc.type.hasversion | info:eu-repo/semantics/acceptedVersion | spa |
dc.rights.local | Abierto (Texto Completo) | spa |
dc.identifier.doi | https://doi.org/10.1038/s41598-020-74164-z | spa |
dc.type.coar | http://purl.org/coar/resource_type/c_2df8fbb1 | spa |
Archivos en el documento
Archivos | Tamaño | Formato | Ver |
---|---|---|---|
No hay archivos asociados a este documento. |