An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19

Detalhes bibliográficos
Autor(a) principal: Miguel, Pedro Lucas [UNESP]
Data de Publicação: 2023
Outros Autores: Cansian, Adriano Mauro [UNESP], Rozendo, Guilherme Botazzo [UNESP], Medalha, Giuliano Cardozo, do Nascimento, Marcelo Zanchetta, Neves, Leandro Alves [UNESP]
Tipo de documento: Artigo de conferência
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://dx.doi.org/10.5220/0012038500003467
http://hdl.handle.net/11449/248922
Resumo: In this proposal, a study based on deep-learned features via transfer learning was developed to obtain a set of features and techniques for pattern recognition in the context of COVID-19 images. The proposal was based on the ResNet-50, DenseNet-201 and EfficientNet-b0 deep-learning models. In this work, the chosen layer for analysis was the avg pool layer from each model, with 2048 features from the ResNet-50, 1920 features from the DenseNet0201 and 1280 obtained features from the EfficientNet-b0. The most relevant descriptors were defined for the classification process, applying the ReliefF algorithm and two classification strategies: individually applied classifiers and employed an ensemble of classifiers using the score-level fusion approach. Thus, the two best combinations were identified, both using the DenseNet-201 model with the same subset of features. The first combination was defined via the SMO classifier (accuracy of 98.38%) and the second via the ensemble strategy (accuracy of 97.89%). The feature subset was composed of only 210 descriptors, representing only 10% of the original set. The strategies and information presented here are relevant contributions for the specialists interested in the study and development of computer-aided diagnosis in COVID-19 images.
id UNSP_573482ed81769bef0e077a59cadbeb17
oai_identifier_str oai:repositorio.unesp.br:11449/248922
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19Convolutional Neural NetworksCOVID-19Deep-Learned FeaturesRadiographic ImagesRelieFIn this proposal, a study based on deep-learned features via transfer learning was developed to obtain a set of features and techniques for pattern recognition in the context of COVID-19 images. The proposal was based on the ResNet-50, DenseNet-201 and EfficientNet-b0 deep-learning models. In this work, the chosen layer for analysis was the avg pool layer from each model, with 2048 features from the ResNet-50, 1920 features from the DenseNet0201 and 1280 obtained features from the EfficientNet-b0. The most relevant descriptors were defined for the classification process, applying the ReliefF algorithm and two classification strategies: individually applied classifiers and employed an ensemble of classifiers using the score-level fusion approach. Thus, the two best combinations were identified, both using the DenseNet-201 model with the same subset of features. The first combination was defined via the SMO classifier (accuracy of 98.38%) and the second via the ensemble strategy (accuracy of 97.89%). The feature subset was composed of only 210 descriptors, representing only 10% of the original set. The strategies and information presented here are relevant contributions for the specialists interested in the study and development of computer-aided diagnosis in COVID-19 images.Faculdade de Medicina de São José do Rio PretoConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG)Department of Computer Science and Statistics (DCCE) São Paulo State University (UNESP), Rua Cristóvão Colombo, 2265, SPWZTECH NETWORKS, Avenida Romeu Strazzi (room 503-B), 325, SPFaculty of Computer Science (FACOM) Federal University of Uberlândia (UFU), Avenida João Naves de Ávila 2121, Bl.B, MGDepartment of Computer Science and Statistics (DCCE) São Paulo State University (UNESP), Rua Cristóvão Colombo, 2265, SPCNPq: #120993/2020-1CNPq: #311404/2021-9CNPq: #313643/2021-0FAPEMIG: #APQ-00578-18Universidade Estadual Paulista (UNESP)WZTECH NETWORKSUniversidade Federal de Uberlândia (UFU)Miguel, Pedro Lucas [UNESP]Cansian, Adriano Mauro [UNESP]Rozendo, Guilherme Botazzo [UNESP]Medalha, Giuliano Cardozodo Nascimento, Marcelo ZanchettaNeves, Leandro Alves [UNESP]2023-07-29T13:57:25Z2023-07-29T13:57:25Z2023-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject675-682http://dx.doi.org/10.5220/0012038500003467International Conference on Enterprise Information Systems, ICEIS - Proceedings, v. 1, p. 675-682.2184-4992http://hdl.handle.net/11449/24892210.5220/00120385000034672-s2.0-85160769624Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengInternational Conference on Enterprise Information Systems, ICEIS - Proceedingsinfo:eu-repo/semantics/openAccess2023-07-29T13:57:25Zoai:repositorio.unesp.br:11449/248922Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T20:15:42.746291Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
title An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
spellingShingle An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
Miguel, Pedro Lucas [UNESP]
Convolutional Neural Networks
COVID-19
Deep-Learned Features
Radiographic Images
RelieF
title_short An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
title_full An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
title_fullStr An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
title_full_unstemmed An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
title_sort An Investigation of Deep-Learned Features for Classifying Radiographic Images of COVID-19
author Miguel, Pedro Lucas [UNESP]
author_facet Miguel, Pedro Lucas [UNESP]
Cansian, Adriano Mauro [UNESP]
Rozendo, Guilherme Botazzo [UNESP]
Medalha, Giuliano Cardozo
do Nascimento, Marcelo Zanchetta
Neves, Leandro Alves [UNESP]
author_role author
author2 Cansian, Adriano Mauro [UNESP]
Rozendo, Guilherme Botazzo [UNESP]
Medalha, Giuliano Cardozo
do Nascimento, Marcelo Zanchetta
Neves, Leandro Alves [UNESP]
author2_role author
author
author
author
author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (UNESP)
WZTECH NETWORKS
Universidade Federal de Uberlândia (UFU)
dc.contributor.author.fl_str_mv Miguel, Pedro Lucas [UNESP]
Cansian, Adriano Mauro [UNESP]
Rozendo, Guilherme Botazzo [UNESP]
Medalha, Giuliano Cardozo
do Nascimento, Marcelo Zanchetta
Neves, Leandro Alves [UNESP]
dc.subject.por.fl_str_mv Convolutional Neural Networks
COVID-19
Deep-Learned Features
Radiographic Images
RelieF
topic Convolutional Neural Networks
COVID-19
Deep-Learned Features
Radiographic Images
RelieF
description In this proposal, a study based on deep-learned features via transfer learning was developed to obtain a set of features and techniques for pattern recognition in the context of COVID-19 images. The proposal was based on the ResNet-50, DenseNet-201 and EfficientNet-b0 deep-learning models. In this work, the chosen layer for analysis was the avg pool layer from each model, with 2048 features from the ResNet-50, 1920 features from the DenseNet0201 and 1280 obtained features from the EfficientNet-b0. The most relevant descriptors were defined for the classification process, applying the ReliefF algorithm and two classification strategies: individually applied classifiers and employed an ensemble of classifiers using the score-level fusion approach. Thus, the two best combinations were identified, both using the DenseNet-201 model with the same subset of features. The first combination was defined via the SMO classifier (accuracy of 98.38%) and the second via the ensemble strategy (accuracy of 97.89%). The feature subset was composed of only 210 descriptors, representing only 10% of the original set. The strategies and information presented here are relevant contributions for the specialists interested in the study and development of computer-aided diagnosis in COVID-19 images.
publishDate 2023
dc.date.none.fl_str_mv 2023-07-29T13:57:25Z
2023-07-29T13:57:25Z
2023-01-01
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/conferenceObject
format conferenceObject
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://dx.doi.org/10.5220/0012038500003467
International Conference on Enterprise Information Systems, ICEIS - Proceedings, v. 1, p. 675-682.
2184-4992
http://hdl.handle.net/11449/248922
10.5220/0012038500003467
2-s2.0-85160769624
url http://dx.doi.org/10.5220/0012038500003467
http://hdl.handle.net/11449/248922
identifier_str_mv International Conference on Enterprise Information Systems, ICEIS - Proceedings, v. 1, p. 675-682.
2184-4992
10.5220/0012038500003467
2-s2.0-85160769624
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv International Conference on Enterprise Information Systems, ICEIS - Proceedings
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 675-682
dc.source.none.fl_str_mv Scopus
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1808129180706013184