Segmentation of cervical nuclei using convolutional neural network for conventional cytology.

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.creatorTeixeira, Júlia Beatriz Azevedo-
Autor(es): dc.creatorRezende, Mariana Trevisan-
Autor(es): dc.creatorDiniz, Débora Nasser-
Autor(es): dc.creatorCarneiro, Cláudia Martins-
Autor(es): dc.creatorLuz, Eduardo José da Silva-
Autor(es): dc.creatorSouza, Marcone Jamilson Freitas-
Autor(es): dc.creatorUshizima, Daniela Mayumi-
Autor(es): dc.creatorMedeiros, Fátima Nelsizeuma Sombra de-
Autor(es): dc.creatorBianchi, Andrea Gomes Campos-
Data de aceite: dc.date.accessioned2025-08-21T15:23:12Z-
Data de disponibilização: dc.date.available2025-08-21T15:23:12Z-
Data de envio: dc.date.issued2025-01-13-
Data de envio: dc.date.issued2025-01-13-
Data de envio: dc.date.issued2022-
Fonte completa do material: dc.identifierhttps://www.repositorio.ufop.br/handle/123456789/19516-
Fonte completa do material: dc.identifierhttps://www.tandfonline.com/doi/epdf/10.1080/21681163.2023.2197086?needAccess=true-
Fonte completa do material: dc.identifierhttps://doi.org/10.1080/21681163.2023.2197086© 2023 Informa UK Limited, trading as Taylor & Francis Group-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/capes/1012804-
Descrição: dc.descriptionAlthough implementing the Pap smear has drastically reduced the mortality rates from cervical cancer, false positives and negatives are related to the quality of the analysis and the cytopathologist experience. An alternative is the insertion of digital cytology in the quality monitoring to assist the screening. However, conventional cytology is still a major challenge, as it presents a lot of cellular overlap and several epithelial structures that make it difficult to implement computational methodologies. This article compares the performance of U-net and SegNet neural networks for nuclei segmentation in cervical images. Experiments were performed with different activation functions, batch sizes, and datasets, ISBI (synthetic images from liquid cytology) and CRIC Cervix-Seg (conventional cytology real images). The models achieved a Dice coefficient of 0.9783 for ISBI2015 and 0.9429 for CRIC Cervix-Seg. These results suggest a methodology capable of segmenting real images of cervical nuclei with quality, even in situations of overlap and artefacts, advancing efforts towards the automation of tasks as part of the cytopathological analysis in the laboratory work routine.-
Formato: dc.formatapplication/pdf-
Idioma: dc.languageen-
Direitos: dc.rightsrestrito-
Palavras-chave: dc.subjectCervical nuclei-
Palavras-chave: dc.subjectSegmentation-
Palavras-chave: dc.subjectConvolutional neural network-
Palavras-chave: dc.subjectPap smear-
Título: dc.titleSegmentation of cervical nuclei using convolutional neural network for conventional cytology.-
Aparece nas coleções:Repositório Institucional - UFOP

Não existem arquivos associados a este item.