Please use this identifier to cite or link to this item: http://repository.aaup.edu/jspui/handle/123456789/1468
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKarajah, Eman Naser $AAUP$Palestinian-
dc.contributor.authorAwad, Mohammed $AAUP$Palestinian-
dc.date.accessioned2022-02-04T07:41:54Z-
dc.date.available2022-02-04T07:41:54Z-
dc.date.issued2022-01-31-
dc.identifier.citationInternational Conference on Promising Electronic Technologies (ICPET)en_US
dc.identifier.urihttp://repository.aaup.edu/jspui/handle/123456789/1468-
dc.description.abstractCovid-19 is a newly discovered coronavirus that the World Health Organization has officially announced in March 2020 as a pandemic. It is a new virus in the medical field with no specific treatment. Besides, they have not discovered all the symptoms but only some of them. Covid-19 is spreading very fast as the medical systems worldwide cannot hospitalize all the patients, which leads to an increase in the number of the virus death. This work will use the power of deep learning and transfer learning to give faster diagnoses for infection in Covid-19 using X-ray images. The proposed approach is a modified version from Visual Geometry Group (VGG 16). It uses the architecture of the VGG16 with modification to achieve higher accuracy. The model was trained to classify X-ray images into two classes; normal (healthy) and Covid-19 (sick) classes. The model can then predict any uploaded X-ray image class as normal or Covid 19. The achieved accuracy by modified VGG 16 is 99.7%. The model is evaluated through a confusion matrix, precision, accuracy, recall, and f measure.en_US
dc.language.isoenen_US
dc.publisherIEEE Xploreen_US
dc.relation.ispartofseries(ICPET), 2021;, pp. 46-51-
dc.subjectCoronavirus, Covid-19en_US
dc.subjectDeep Learningen_US
dc.subjectTransfer Learningen_US
dc.subjectVisual Geometry Group (VGG 16)en_US
dc.titleCovid-19 Detection From Chest X-Rays Using Modified VGG 16 Modelen_US
dc.typeArticleen_US
Appears in Collections:Faculty & Staff Scientific Research publications

Files in This Item:
There are no files associated with this item.
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Admin Tools