ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification

Autores
Gramática, Martín Nicolás; García, Mario Alejandro; Silva Fiezzy, Tomás; Ricapito, Juan Pablo
Año de publicación
2023
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Cervical cancer (CC) remains a significant global health concern, and one of the key tools for its early detection and diagnosis is the Papanicolaou test, commonly known as the Pap test. This test involves the examination of cells from the cervix to identify abnormal cellular changes that could indicate the presence of pre-cancerous or cancerous conditions. Automating and improving the accuracy of cell detection and classification in Pap test samples can aid healthcare professionals, particularly pathologists, in making more precise and timely diagnoses. The study focuses on two main approaches: one uses a pure YOLOv7 object detection model to locate and classify cells in a single stage, while the other employs a two-step approach that combines YOLOv7 for localization and EfficientNetB0 for classification. The results show that the pure YOLOv7 model provides acceptable performance for cell detection in cytological samples, which could assist pathologists in identifying regions of interest with cells showing a certain degree of alteration in the samples. On the other hand, the two-step approach, although promising in theory, fails to outperform the pure YOLOv7 model and adds complexity to the process. This work also introduces a web interface and an API that allow real-time inference using the trained models, which could be a useful tool for pathologists in their daily CC diagnosis work.
Fil: Gramática, Martín Nicolás . Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: García, Mario Alejandro. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: Silva Fiezzy, Tomás. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: Ricapito, Juan Pablo. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Peer Reviewed
Materia
Pap test
Cervical cancer
Deep learning
Digital pathology
Nivel de accesibilidad
acceso abierto
Condiciones de uso
Attribution-NonCommercial-NoDerivs 2.5 Argentina
Repositorio
Repositorio Institucional Abierto (UTN)
Institución
Universidad Tecnológica Nacional
OAI Identificador
oai:ria.utn.edu.ar:20.500.12272/15580
Acceder
id RIAUTN_b1b2760a94445787a0f0fb172130f2bf
oai_identifier_str oai:ria.utn.edu.ar:20.500.12272/15580
network_acronym_str RIAUTN
repository_id_str a
network_name_str Repositorio Institucional Abierto (UTN)
spelling ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classificationGramática, Martín NicolásGarcía, Mario AlejandroSilva Fiezzy, TomásRicapito, Juan PabloPap testCervical cancerDeep learningDigital pathologyCervical cancer (CC) remains a significant global health concern, and one of the key tools for its early detection and diagnosis is the Papanicolaou test, commonly known as the Pap test. This test involves the examination of cells from the cervix to identify abnormal cellular changes that could indicate the presence of pre-cancerous or cancerous conditions. Automating and improving the accuracy of cell detection and classification in Pap test samples can aid healthcare professionals, particularly pathologists, in making more precise and timely diagnoses. The study focuses on two main approaches: one uses a pure YOLOv7 object detection model to locate and classify cells in a single stage, while the other employs a two-step approach that combines YOLOv7 for localization and EfficientNetB0 for classification. The results show that the pure YOLOv7 model provides acceptable performance for cell detection in cytological samples, which could assist pathologists in identifying regions of interest with cells showing a certain degree of alteration in the samples. On the other hand, the two-step approach, although promising in theory, fails to outperform the pure YOLOv7 model and adds complexity to the process. This work also introduces a web interface and an API that allow real-time inference using the trained models, which could be a useful tool for pathologists in their daily CC diagnosis work.Fil: Gramática, Martín Nicolás . Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.Fil: García, Mario Alejandro. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.Fil: Silva Fiezzy, Tomás. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.Fil: Ricapito, Juan Pablo. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.Peer ReviewededUTecNe2026-09-15T19:15:44Z2023-11-02info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articulopdfapplication/pdfXI Congreso Nacional de Ingeniería en Informática y Sistemas de Información. CoNaIISI 2023978-987-8992-38-9https://hdl.handle.net/20.500.12272/15580enginfo:eu-repo/semantics/openAccessAttribution-NonCommercial-NoDerivs 2.5 Argentinahttp://creativecommons.org/licenses/by-nc-nd/2.5/ar/Gramática, Martín Nicolás; García, Mario Alejandro; Silva Fiezzy, Tomás; Ricapito, Juan Pablohttps://creativecommons.org/licenses/by-nc-nd/4.0/reponame:Repositorio Institucional Abierto (UTN)instname:Universidad Tecnológica Nacional2026-09-24T12:45:24Zoai:ria.utn.edu.ar:20.500.12272/15580instacron:UTNInstitucionalhttp://ria.utn.edu.ar/Universidad públicaNo correspondehttp://ria.utn.edu.ar/oaigestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:a2026-09-24 12:45:24.851Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacionalfalse
dc.title.none.fl_str_mv ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
title ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
spellingShingle ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
Gramática, Martín Nicolás
Pap test
Cervical cancer
Deep learning
Digital pathology
title_short ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
title_full ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
title_fullStr ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
title_full_unstemmed ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
title_sort ConvPapDetect: Enhancing conventional pap smear analysis with deep learning for cell detection and classification
dc.creator.none.fl_str_mv Gramática, Martín Nicolás
García, Mario Alejandro
Silva Fiezzy, Tomás
Ricapito, Juan Pablo
author Gramática, Martín Nicolás
author_facet Gramática, Martín Nicolás
García, Mario Alejandro
Silva Fiezzy, Tomás
Ricapito, Juan Pablo
author_role author
author2 García, Mario Alejandro
Silva Fiezzy, Tomás
Ricapito, Juan Pablo
author2_role author
author
author
dc.subject.none.fl_str_mv Pap test
Cervical cancer
Deep learning
Digital pathology
topic Pap test
Cervical cancer
Deep learning
Digital pathology
dc.description.none.fl_txt_mv Cervical cancer (CC) remains a significant global health concern, and one of the key tools for its early detection and diagnosis is the Papanicolaou test, commonly known as the Pap test. This test involves the examination of cells from the cervix to identify abnormal cellular changes that could indicate the presence of pre-cancerous or cancerous conditions. Automating and improving the accuracy of cell detection and classification in Pap test samples can aid healthcare professionals, particularly pathologists, in making more precise and timely diagnoses. The study focuses on two main approaches: one uses a pure YOLOv7 object detection model to locate and classify cells in a single stage, while the other employs a two-step approach that combines YOLOv7 for localization and EfficientNetB0 for classification. The results show that the pure YOLOv7 model provides acceptable performance for cell detection in cytological samples, which could assist pathologists in identifying regions of interest with cells showing a certain degree of alteration in the samples. On the other hand, the two-step approach, although promising in theory, fails to outperform the pure YOLOv7 model and adds complexity to the process. This work also introduces a web interface and an API that allow real-time inference using the trained models, which could be a useful tool for pathologists in their daily CC diagnosis work.
Fil: Gramática, Martín Nicolás . Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: García, Mario Alejandro. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: Silva Fiezzy, Tomás. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Fil: Ricapito, Juan Pablo. Universidad Tecnológica Nacional. Facultad Regional Córdoba. Grupo de Investigación en Inteligencia Artificial; Argentina.
Peer Reviewed
description Cervical cancer (CC) remains a significant global health concern, and one of the key tools for its early detection and diagnosis is the Papanicolaou test, commonly known as the Pap test. This test involves the examination of cells from the cervix to identify abnormal cellular changes that could indicate the presence of pre-cancerous or cancerous conditions. Automating and improving the accuracy of cell detection and classification in Pap test samples can aid healthcare professionals, particularly pathologists, in making more precise and timely diagnoses. The study focuses on two main approaches: one uses a pure YOLOv7 object detection model to locate and classify cells in a single stage, while the other employs a two-step approach that combines YOLOv7 for localization and EfficientNetB0 for classification. The results show that the pure YOLOv7 model provides acceptable performance for cell detection in cytological samples, which could assist pathologists in identifying regions of interest with cells showing a certain degree of alteration in the samples. On the other hand, the two-step approach, although promising in theory, fails to outperform the pure YOLOv7 model and adds complexity to the process. This work also introduces a web interface and an API that allow real-time inference using the trained models, which could be a useful tool for pathologists in their daily CC diagnosis work.
publishDate 2023
dc.date.none.fl_str_mv 2023-11-02
2026-09-15T19:15:44Z
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv XI Congreso Nacional de Ingeniería en Informática y Sistemas de Información. CoNaIISI 2023
978-987-8992-38-9
https://hdl.handle.net/20.500.12272/15580
identifier_str_mv XI Congreso Nacional de Ingeniería en Informática y Sistemas de Información. CoNaIISI 2023
978-987-8992-38-9
url https://hdl.handle.net/20.500.12272/15580
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
Attribution-NonCommercial-NoDerivs 2.5 Argentina
http://creativecommons.org/licenses/by-nc-nd/2.5/ar/
Gramática, Martín Nicolás; García, Mario Alejandro; Silva Fiezzy, Tomás; Ricapito, Juan Pablo
https://creativecommons.org/licenses/by-nc-nd/4.0/
eu_rights_str_mv openAccess
rights_invalid_str_mv Attribution-NonCommercial-NoDerivs 2.5 Argentina
http://creativecommons.org/licenses/by-nc-nd/2.5/ar/
Gramática, Martín Nicolás; García, Mario Alejandro; Silva Fiezzy, Tomás; Ricapito, Juan Pablo
https://creativecommons.org/licenses/by-nc-nd/4.0/
dc.format.none.fl_str_mv pdf
application/pdf
dc.publisher.none.fl_str_mv edUTecNe
publisher.none.fl_str_mv edUTecNe
dc.source.none.fl_str_mv reponame:Repositorio Institucional Abierto (UTN)
instname:Universidad Tecnológica Nacional
reponame_str Repositorio Institucional Abierto (UTN)
collection Repositorio Institucional Abierto (UTN)
instname_str Universidad Tecnológica Nacional
repository.name.fl_str_mv Repositorio Institucional Abierto (UTN) - Universidad Tecnológica Nacional
repository.mail.fl_str_mv gestionria@rec.utn.edu.ar; fsuarez@rec.utn.edu.ar
_version_ 1877230897667768320
score 13.265058

Publicaciones similares