Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions

Oral cancer is a frequently malignant tumor that can be detected during an oral examination. Unfortunately, it is often diagnosed in advanced stages, which leads to low survival rates of about 50% at five years. Due to the low survival rate, it is crucial to develop automated systems that allow the...

ver descrição completa

Detalhes bibliográficos
Autores: Redondo, Alejandro, Ivaylova, Katerina, Bachiller Mayoral, Margarita, Rincón Zamorano, Mariano, Cuadra Troncoso, José Manuel, Tamimi, Faleh, López Cedrún, José Luis, Diniz Freitas, Márcio, Lago Méndez, Lucía, Rubín Roger, Guillermo, Torres, Jesús, Bagán, Leticia, Hernández, Gonzalo, López-Pintor, Rosa María
Formato: artículo
Fecha de publicación:2025
País:España
Recursos:Universidad Nacional de Educación a Distancia
Repositorio:e-spacio. Repositorio Institucional de la UNED
Idioma:inglés
OAI Identifier:oai:e-spacio.uned.es:20.500.14468/31698
Acesso em linha:https://hdl.handle.net/20.500.14468/31698
Access Level:acceso abierto
Palavra-chave:1203.17 Informática
images classification
oral cancer
oral potentially malignant disorders
deep learning
convolutional neural network
skip connection networks
visual transformers
convNeXt
ODS 3 - Salud y bienestar
ODS 9 - Industria, innovación e infraestructura
id ES_69a5eb2d79bf94d51230133d4bd9b853
oai_identifier_str oai:e-spacio.uned.es:20.500.14468/31698
network_acronym_str ES
network_name_str España
repository_id_str
spelling Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictionsRedondo, AlejandroIvaylova, KaterinaBachiller Mayoral, MargaritaRincón Zamorano, MarianoCuadra Troncoso, José ManuelTamimi, FalehLópez Cedrún, José LuisDiniz Freitas, MárcioLago Méndez, LucíaRubín Roger, GuillermoTorres, JesúsBagán, LeticiaHernández, GonzaloLópez-Pintor, Rosa María1203.17 Informáticaimages classificationoral canceroral potentially malignant disordersdeep learningconvolutional neural networkskip connection networksvisual transformersconvNeXtODS 3 - Salud y bienestarODS 9 - Industria, innovación e infraestructuraOral cancer is a frequently malignant tumor that can be detected during an oral examination. Unfortunately, it is often diagnosed in advanced stages, which leads to low survival rates of about 50% at five years. Due to the low survival rate, it is crucial to develop automated systems that allow the classification of oral lesions according to their severity, aiding in the early diagnosis of oral cancer. This study aims to investigate the effectiveness of using clinical images and deep learning based models to perform a multiclass classification of oral mucosal lesions in color photographs taken without following any acquisition protocol. The classification differentiated four classes: malignant, potentially malignant, benign and healthy. The dataset included a total of 3246 images from 1013 patients, with 40 different categories of oral lesions, including healthy oral mucosa. The images showed different areas of the oral cavity and were captured from different perspectives by diverse dentists and maxillofacial surgeons in the practice. For the classification, different deep learning architectures were applied and compared, from the best known convolutional neural networks (CNN) and skip connection networks (SCN), to more innovative architectures such as visual transformers and a recent hybrid architecture, ConvNeXt v2. The ConvNeXt v2 Tiny architecture, with 85.53% accuracy, 85.02% precision, 85.50% recall, 84.92% F1-score, and 97.40% ROC AUC for an input image size of 354 × 354 pixels, outperformed the other architectures on the same database. The present model improved on previous proposals by considering a greater number of oral lesions and output classes.ELSEVIERInstituto de Salud Carlos III (ISCIII)Agencia Estatal de Investigacióne-Spacio UNED20262026-02-0420252025-07-1920252025-07-19journal articlehttp://purl.org/coar/resource_type/c_6501info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/20.500.14468/31698reponame:e-spacio. Repositorio Institucional de la UNEDinstname:Universidad Nacional de Educación a DistanciaInglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by-nc-nd/4.0/deed.esoai:e-spacio.uned.es:20.500.14468/316982026-06-06T12:38:31Z
dc.title.none.fl_str_mv Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
title Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
spellingShingle Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
Redondo, Alejandro
1203.17 Informática
images classification
oral cancer
oral potentially malignant disorders
deep learning
convolutional neural network
skip connection networks
visual transformers
convNeXt
ODS 3 - Salud y bienestar
ODS 9 - Industria, innovación e infraestructura
title_short Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
title_full Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
title_fullStr Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
title_full_unstemmed Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
title_sort Multiclass classification of oral mucosal lesions by deep learning from clinical images without performing any restrictions
dc.creator.none.fl_str_mv Redondo, Alejandro
Ivaylova, Katerina
Bachiller Mayoral, Margarita
Rincón Zamorano, Mariano
Cuadra Troncoso, José Manuel
Tamimi, Faleh
López Cedrún, José Luis
Diniz Freitas, Márcio
Lago Méndez, Lucía
Rubín Roger, Guillermo
Torres, Jesús
Bagán, Leticia
Hernández, Gonzalo
López-Pintor, Rosa María
author Redondo, Alejandro
author_facet Redondo, Alejandro
Ivaylova, Katerina
Bachiller Mayoral, Margarita
Rincón Zamorano, Mariano
Cuadra Troncoso, José Manuel
Tamimi, Faleh
López Cedrún, José Luis
Diniz Freitas, Márcio
Lago Méndez, Lucía
Rubín Roger, Guillermo
Torres, Jesús
Bagán, Leticia
Hernández, Gonzalo
López-Pintor, Rosa María
author_role author
author2 Ivaylova, Katerina
Bachiller Mayoral, Margarita
Rincón Zamorano, Mariano
Cuadra Troncoso, José Manuel
Tamimi, Faleh
López Cedrún, José Luis
Diniz Freitas, Márcio
Lago Méndez, Lucía
Rubín Roger, Guillermo
Torres, Jesús
Bagán, Leticia
Hernández, Gonzalo
López-Pintor, Rosa María
author2_role author
author
author
author
author
author
author
author
author
author
author
author
author
dc.contributor.none.fl_str_mv Instituto de Salud Carlos III (ISCIII)
Agencia Estatal de Investigación
e-Spacio UNED
dc.subject.none.fl_str_mv 1203.17 Informática
images classification
oral cancer
oral potentially malignant disorders
deep learning
convolutional neural network
skip connection networks
visual transformers
convNeXt
ODS 3 - Salud y bienestar
ODS 9 - Industria, innovación e infraestructura
topic 1203.17 Informática
images classification
oral cancer
oral potentially malignant disorders
deep learning
convolutional neural network
skip connection networks
visual transformers
convNeXt
ODS 3 - Salud y bienestar
ODS 9 - Industria, innovación e infraestructura
description Oral cancer is a frequently malignant tumor that can be detected during an oral examination. Unfortunately, it is often diagnosed in advanced stages, which leads to low survival rates of about 50% at five years. Due to the low survival rate, it is crucial to develop automated systems that allow the classification of oral lesions according to their severity, aiding in the early diagnosis of oral cancer. This study aims to investigate the effectiveness of using clinical images and deep learning based models to perform a multiclass classification of oral mucosal lesions in color photographs taken without following any acquisition protocol. The classification differentiated four classes: malignant, potentially malignant, benign and healthy. The dataset included a total of 3246 images from 1013 patients, with 40 different categories of oral lesions, including healthy oral mucosa. The images showed different areas of the oral cavity and were captured from different perspectives by diverse dentists and maxillofacial surgeons in the practice. For the classification, different deep learning architectures were applied and compared, from the best known convolutional neural networks (CNN) and skip connection networks (SCN), to more innovative architectures such as visual transformers and a recent hybrid architecture, ConvNeXt v2. The ConvNeXt v2 Tiny architecture, with 85.53% accuracy, 85.02% precision, 85.50% recall, 84.92% F1-score, and 97.40% ROC AUC for an input image size of 354 × 354 pixels, outperformed the other architectures on the same database. The present model improved on previous proposals by considering a greater number of oral lesions and output classes.
publishDate 2025
dc.date.none.fl_str_mv 2025
2025-07-19
2025
2025-07-19
2026
2026-02-04
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/20.500.14468/31698
url https://hdl.handle.net/20.500.14468/31698
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.es
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.es
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv ELSEVIER
publisher.none.fl_str_mv ELSEVIER
dc.source.none.fl_str_mv reponame:e-spacio. Repositorio Institucional de la UNED
instname:Universidad Nacional de Educación a Distancia
instname_str Universidad Nacional de Educación a Distancia
reponame_str e-spacio. Repositorio Institucional de la UNED
collection e-spacio. Repositorio Institucional de la UNED
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869410044477440000
score 15,198674