Logo do repositório
 
Publicação

Shallow and deep learning approaches to classify melanoma and non-melanocytic skin lesions

2026-01-13Artigo científico Acesso restrito
Ver registo simples
datacite.subject.fosCiências Médicas::Ciências da Saúde
datacite.subject.sdg03:Saúde de Qualidade
dc.contributor.authorSpolaôr, Newton
dc.contributor.authorLee, Huei Diana
dc.contributor.authorTakaki, Weber Shoity Resende
dc.contributor.authorMendes, Ana Isabel Gonçalves
dc.contributor.authorFonseca-Pinto, Rui
dc.contributor.authorNogueira, Conceição Veloso
dc.contributor.authorCoy, Claudio Saddy Rodrigues
dc.contributor.authorWu, Feng Chung
dc.date.accessioned2026-04-22T14:30:00Z
dc.date.available2026-04-22T14:30:00Z
dc.date.issued2026-01-13
dc.descriptionWe would like to acknowledge the Secretaria de Estado da Ciência, Tecnologia e Ensino Superior—SETI—Fundo Paraná through a research scholarship by the Project 49/2024 to some LABI members. We also thank eurekaSD: Enhancing University Research and Education in Areas Useful for Sustainable Development—Grants EK14AC0037 and EK15AC0264. The Portuguese team was partially supported by Fundação para a Ciência e a Tecnologia (FCT). R Fonseca-Pinto was financed by the Projects UID/05704/2023 and UID/50008/2023, and C V Nogueira was financed by the Project UID/00013/2025. The funding agencies did not have any further involvement in this paper. We also thank PGEEC and PPGCAS at UNIOESTE for their cooperation in this research.
dc.description.abstractSeveral image processing methods in Dermatology are grounded in shallow and deep learning approaches. These solutions are relevant to assist health experts in decision-making processes related to harmful melanoma—a malignant melanocytic condition—and other skin lesions. This work aims to compare these approaches in a specific classification problem: malignant melanocytic lesions versus non-melanocytic ones. We developed 39 learning method configurations, including three original ones based on fine-tuned deep neural networks. Some implemented settings incorporate auxiliary procedures, such as oversampling, feature selection and data augmentation. An experimental evaluation in the public Derm7pt dermoscopic database suggests that the best original setting performance was competitive against the leading results reported by recent literature alternatives. In particular, the proposal reached average accuracy and sensitivity of 0.9909 and 0.9976, respectively. These results were averaged across three runs of the stratified nested cross-validation strategy. Moreover, our 39 configurations outperformed an experimental baseline derived from the majority class error. Thus, this work can be helpful in inspiring computational systems that could act as preliminary filters to support the detection of a harmful form of skin cancer and its separation from other lesions.eng
dc.description.sponsorship- Secretaria de Estado da Ciência, Tecnologia e Ensino Superior—SETI—Fundo Paraná: Project 49/2024 . —Grants EK14AC0037 and EK15AC0264. The Portuguese team was partially supported by Fundação para a Ciência e a Tecnologia (FCT). R Fonseca-Pinto was financed by the Projects UID/05704/2023 and UID/50008/2023, and C V Nogueira was financed by the Project UID/00013/2025. The funding agencies did not have any further involvement in this paper. We also thank PGEEC and PPGCAS at UNIOESTE for their cooperation in this research.
dc.identifier.citationSpolaôr, N., Cherman, E. A., Igawa, R. A., Sato, A. K., Nanni, M. R., & Falcão, A. X. (2026). Shallow and deep learning approaches to classify melanoma and non-melanocytic skin lesions. Medical Engineering & Physics, 147, 015014. https://doi.org/10.1088/1873-4030/ac290b
dc.identifier.doi10.1088/1873-4030/ae290b
dc.identifier.issn1873-4030
dc.identifier.urihttp://hdl.handle.net/10400.8/16175
dc.language.isoeng
dc.peerreviewedyes
dc.publisherIOP Publishing
dc.relationCenter for Innovative Care and Health Technology
dc.relationInstituto de Telecomunicações
dc.relationCenter of Mathematics of the University of Minho
dc.relation.hasversionhttps://iopscience.iop.org/article/10.1088/1873-4030/ae290b
dc.relation.ispartofMedical Engineering & Physics
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectArtificial intelligence
dc.subjectBiomedical engineering
dc.subjectFeature learning
dc.subjectMachine learning
dc.subjectStatistical test
dc.subjectTransfer learning
dc.titleShallow and deep learning approaches to classify melanoma and non-melanocytic skin lesionseng
dc.typejournal article
dspace.entity.typePublication
oaire.awardNumberUIDB/05704/2020
oaire.awardNumberUID/50008/2025
oaire.awardNumberUID/00013/2025
oaire.awardTitleCenter for Innovative Care and Health Technology
oaire.awardTitleInstituto de Telecomunicações
oaire.awardTitleCenter of Mathematics of the University of Minho
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F05704%2F2020/PT
oaire.awardURIhttp://hdl.handle.net/10400.8/15747
oaire.awardURIhttp://hdl.handle.net/10400.8/16174
oaire.citation.endPage13
oaire.citation.issue1
oaire.citation.startPage1
oaire.citation.titleMedical Engineering & Physics
oaire.citation.volume147
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStreamAvaliação UID 2023/2024
oaire.fundingStreamAvaliação UID 2023/2024
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameMendes
person.familyNameFonseca-Pinto
person.familyNameNogueira
person.givenNameAna
person.givenNameRui
person.givenNameConceição
person.identifier.ciencia-id3C19-42EA-DDBD
person.identifier.ciencia-id681D-C547-B184
person.identifier.ciencia-id9C14-CD1D-1298
person.identifier.orcid0000-0002-4161-6130
person.identifier.orcid0000-0001-6774-5363
person.identifier.orcid0000-0002-9269-2221
person.identifier.ridK-9449-2014
person.identifier.scopus-author-id55873199300
person.identifier.scopus-author-id26039086400
person.identifier.scopus-author-id24339135500
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
relation.isAuthorOfPublication3523e160-c260-47a6-89cc-eb3fca0059d9
relation.isAuthorOfPublication7eb9d123-1800-4afd-a2f6-91043353011b
relation.isAuthorOfPublicationb44ae8f7-d99a-4098-ae0b-bbf22db8f2a0
relation.isAuthorOfPublication.latestForDiscovery7eb9d123-1800-4afd-a2f6-91043353011b
relation.isProjectOfPublicationd421b07d-3471-4026-aa43-def80b8e142b
relation.isProjectOfPublicationfb51562e-8a49-4d10-b530-5f39dca8aa62
relation.isProjectOfPublication5f22f5c1-8776-4486-a4be-8f9c8a76b6b5
relation.isProjectOfPublication.latestForDiscoveryd421b07d-3471-4026-aa43-def80b8e142b

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura indisponível
Nome:
Shallow and deep learning approaches to classify melanoma and non-melanocytic skin lesions.pdf
Tamanho:
12.4 MB
Formato:
Adobe Portable Document Format
Ver/Abrir
Licença
A mostrar 1 - 1 de 1
Miniatura indisponível
Nome:
license.txt
Tamanho:
1.32 KB
Formato:
Item-specific license agreed upon to submission
Descrição:
Ver/Abrir

Coleções

ciTechCare - Artigos
ESSLei - Artigos em revistas internacionais
ESTG - Artigos em revistas internacionais