Convolutional neural network misclassification analysis in oral lesions: an error evaluation criterion by image characteristics

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorFaculdade de Odontologia–Federal University of Rio Grande do Sul–UFRGS-
Autor(es): dc.contributorUniversity of Vale do Rio dos Sinos–UNISINOS-
Autor(es): dc.contributorFederal University of Rio Grande do Sul-
Autor(es): dc.contributorHospital de Clínicas de Porto Alegre (HCPA)-
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.creatorGomes, Rita Fabiane Teixeira-
Autor(es): dc.creatorSchmith, Jean-
Autor(es): dc.creatorde Figueiredo, Rodrigo Marques-
Autor(es): dc.creatorFreitas, Samuel Armbrust-
Autor(es): dc.creatorMachado, Giovanna Nunes-
Autor(es): dc.creatorRomanini, Juliana-
Autor(es): dc.creatorAlmeida, Janete Dias-
Autor(es): dc.creatorPereira, Cassius Torres-
Autor(es): dc.creatorRodrigues, Jonas de Almeida-
Autor(es): dc.creatorCarrard, Vinicius Coelho-
Data de aceite: dc.date.accessioned2025-08-21T21:55:37Z-
Data de disponibilização: dc.date.available2025-08-21T21:55:37Z-
Data de envio: dc.date.issued2025-04-29-
Data de envio: dc.date.issued2024-03-01-
Fonte completa do material: dc.identifierhttp://dx.doi.org/10.1016/j.oooo.2023.10.003-
Fonte completa do material: dc.identifierhttps://hdl.handle.net/11449/298258-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/298258-
Descrição: dc.descriptionObjective: This retrospective study analyzed the errors generated by a convolutional neural network (CNN) when performing automated classification of oral lesions according to their clinical characteristics, seeking to identify patterns in systemic errors in the intermediate layers of the CNN. Study Design: A cross-sectional analysis nested in a previous trial in which automated classification by a CNN model of elementary lesions from clinical images of oral lesions was performed. The resulting CNN classification errors formed the dataset for this study. A total of 116 real outputs were identified that diverged from the estimated outputs, representing 7.6% of the total images analyzed by the CNN. Results: The discrepancies between the real and estimated outputs were associated with problems relating to image sharpness, resolution, and focus; human errors; and the impact of data augmentation. Conclusions: From qualitative analysis of errors in the process of automated classification of clinical images, it was possible to confirm the impact of image quality, as well as identify the strong impact of the data augmentation process. Knowledge of the factors that models evaluate to make decisions can increase confidence in the high classification potential of CNNs.-
Descrição: dc.descriptionDepartment of Oral Pathology Faculdade de Odontologia–Federal University of Rio Grande do Sul–UFRGS-
Descrição: dc.descriptionPolytechnic School University of Vale do Rio dos Sinos–UNISINOS-
Descrição: dc.descriptionTechnology in Automation and Electronics Laboratory–TECAE Lab University of Vale do Rio dos Sinos–UNISINOS-
Descrição: dc.descriptionDepartment of Applied Computing University of Vale do Rio dos Sinos–UNISINOS-
Descrição: dc.descriptionTelessaudeRS–UFRGS Federal University of Rio Grande do Sul, Rio Grande do Sul-
Descrição: dc.descriptionOral Medicine Otorhynolaringology Service Hospital de Clínicas de Porto Alegre (HCPA), Rio Grande do Sul-
Descrição: dc.descriptionDepartment of Biosciences and Oral Diagnostics São Paulo State University Campus São José dos Campos-
Descrição: dc.descriptionDepartment of Stomatology. Federal University of Paraná-
Descrição: dc.descriptionDepartment of Surgery and Orthopaedics Faculdade de Odontologia–Federal University of Rio Grande do Sul–UFRGS-
Descrição: dc.descriptionDepartment of Biosciences and Oral Diagnostics São Paulo State University Campus São José dos Campos-
Formato: dc.format243-252-
Idioma: dc.languageen-
Relação: dc.relationOral Surgery, Oral Medicine, Oral Pathology and Oral Radiology-
???dc.source???: dc.sourceScopus-
Título: dc.titleConvolutional neural network misclassification analysis in oral lesions: an error evaluation criterion by image characteristics-
Tipo de arquivo: dc.typelivro digital-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.