Adversarial Representation Learning for Medical Imaging

Detalhes bibliográficos
Autor(a) principal: Domingues, José David Miranda Barreira
Data de Publicação: 2022
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10451/56666
Resumo: Tese de mestrado, Engenharia Informática, 2022, Universidade de Lisboa, Faculdade de Ciências
id RCAP_d999814602ea7e4bc185fed63c757827
oai_identifier_str oai:repositorio.ul.pt:10451/56666
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Adversarial Representation Learning for Medical ImagingTeses de mestrado - 2022Departamento de InformáticaTese de mestrado, Engenharia Informática, 2022, Universidade de Lisboa, Faculdade de CiênciasBreast cancer is a significant cause of death worldwide, especially among women, being one of the hottest topics in the medical area. In 2020, according to the World Health Organization, there were 2 million women diagnosed with this disease and 685.000 deaths globally. Thus, demonstrating the enormous impact that this disease has and hence the theme of this work being focused on breast cancer. Nowadays, most medical cases use CAD (Computer-Aided Diagnosis) systems in various ways to prevent and help doctors attenuate the impact of cancer by combining their expertise and the advanced technology we have today to perform various tasks. These learning-based systems use many high-quality datasets to extract and identify core aspects and execute multiple tasks. However, there is significant difficulty accessing these datasets because of data protection rules or even different data sharing policies, allied to the nonexistence of suitable enough public datasets and labelled data. Regarding this problem and the growing use of CADs systems in the breast cancer topic, this work proposes generating mammograms based on a single mammogram allowing health entities to generate their mammograms and, thus, a highquality dataset. With that goal, this project uses the base work of ConSinGAN to generate images based on a single one and an innovative way of gaining more image variability by using single image composition harmonisation. The results underwent a validation process, where the images’ quality, diversity and impact were analysed. In terms of real-life usage, there is still a long way to go since such images need to be validated by real doctors and generated at much higher resolutions. However, for now, it is already a significant step toward this purpose.Garcia, Nuno Ricardo da CruzTomás, Helena Isabel Aidos LopesRepositório da Universidade de LisboaDomingues, José David Miranda Barreira2023-03-15T10:55:12Z202220222022-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10451/56666enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-08T17:04:31Zoai:repositorio.ul.pt:10451/56666Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:07:13.754441Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Adversarial Representation Learning for Medical Imaging
title Adversarial Representation Learning for Medical Imaging
spellingShingle Adversarial Representation Learning for Medical Imaging
Domingues, José David Miranda Barreira
Teses de mestrado - 2022
Departamento de Informática
title_short Adversarial Representation Learning for Medical Imaging
title_full Adversarial Representation Learning for Medical Imaging
title_fullStr Adversarial Representation Learning for Medical Imaging
title_full_unstemmed Adversarial Representation Learning for Medical Imaging
title_sort Adversarial Representation Learning for Medical Imaging
author Domingues, José David Miranda Barreira
author_facet Domingues, José David Miranda Barreira
author_role author
dc.contributor.none.fl_str_mv Garcia, Nuno Ricardo da Cruz
Tomás, Helena Isabel Aidos Lopes
Repositório da Universidade de Lisboa
dc.contributor.author.fl_str_mv Domingues, José David Miranda Barreira
dc.subject.por.fl_str_mv Teses de mestrado - 2022
Departamento de Informática
topic Teses de mestrado - 2022
Departamento de Informática
description Tese de mestrado, Engenharia Informática, 2022, Universidade de Lisboa, Faculdade de Ciências
publishDate 2022
dc.date.none.fl_str_mv 2022
2022
2022-01-01T00:00:00Z
2023-03-15T10:55:12Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10451/56666
url http://hdl.handle.net/10451/56666
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799134625516224512

Registros relacionados