Association of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.contributorUniversity of Porto-
Autor(es): dc.contributorUniversidade de São Paulo (USP)-
Autor(es): dc.contributorUniversidade Federal de Uberlândia (UFU)-
Autor(es): dc.creatorLopes, Thales R. S.-
Autor(es): dc.creatorRoberto, Guilherme F.-
Autor(es): dc.creatorSoares, Carlos-
Autor(es): dc.creatorTosta, Thaína A. A.-
Autor(es): dc.creatorSilva, Adriano B.-
Autor(es): dc.creatorLoyola, Adriano M.-
Autor(es): dc.creatorCardoso, Sérgio V.-
Autor(es): dc.creatorde Faria, Paulo R.-
Autor(es): dc.creatordo Nascimento, Marcelo Z.-
Autor(es): dc.creatorNeves, Leandro A.-
Data de aceite: dc.date.accessioned2025-08-21T16:32:53Z-
Data de disponibilização: dc.date.available2025-08-21T16:32:53Z-
Data de envio: dc.date.issued2025-04-29-
Data de envio: dc.date.issued2023-12-31-
Fonte completa do material: dc.identifierhttp://dx.doi.org/10.5220/0012358200003660-
Fonte completa do material: dc.identifierhttps://hdl.handle.net/11449/309503-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/309503-
Descrição: dc.descriptionIn this work, a method based on the use of explainable artificial intelligence techniques with multiscale and multidimensional fractal techniques is presented in order to investigate histological images stained with Hematoxylin-Eosin. The CNN GoogLeNet neural activation patterns were explored, obtained from the gradient-weighted class activation mapping and locally-interpretable model-agnostic explanation techniques. The feature vectors were generated with multiscale and multidimensional fractal techniques, specifically fractal dimension, lacunarity and percolation. The features were evaluated by ranking each entry, using the ReliefF algorithm. The discriminative power of each solution was defined via classifiers with different heuristics. The best results were obtained from LIME, with a significant increase in accuracy and AUC rates when compared to those provided by GoogLeNet. The details presented here can contribute to the development of models aimed at the classification of histological images.-
Descrição: dc.descriptionDepartment of Computer Science and Statistics São Paulo State University, SP-
Descrição: dc.descriptionFaculty of Engineering University of Porto-
Descrição: dc.descriptionInstitute of Science and Technology Federal University of São Paulo, SP-
Descrição: dc.descriptionFaculty of Computer Science Federal University of Uberlândia, MG-
Descrição: dc.descriptionArea of Oral Pathology School of Dentistry Federal University of Uberlândia, MG-
Descrição: dc.descriptionDepartment of Histology and Morphology Institute of Biomedical Science Federal University of Uberlândia, MG-
Descrição: dc.descriptionDepartment of Computer Science and Statistics São Paulo State University, SP-
Formato: dc.format441-447-
Idioma: dc.languageen-
Relação: dc.relationProceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications-
???dc.source???: dc.sourceScopus-
Palavras-chave: dc.subjectDeep Learning-
Palavras-chave: dc.subjectExplainable Artificial Intelligence-
Palavras-chave: dc.subjectFractal Features-
Palavras-chave: dc.subjectHistological Images-
Título: dc.titleAssociation of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images-
Tipo de arquivo: dc.typeaula digital-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.