The use of deep learning state-of-the-art architectures for oral epithelial dysplasia grading: A comparative appraisal.
Anna Luíza Damaceno AraújoViviane Mariano da SilvaMatheus Cardoso MoraesHenrique Alves de AmorimFlávia Sirotheau Correa PontesMaria Sissa Pereira Sant'AnaRicardo Alves MesquitaBruno Augusto Linhares Almeida MarizHelder Antonio Rebelo PontesLucas Lacerda de SouzaCristina Saldivia-SiracusaSyed Ali KhurramAlexander T PearsonManoela Domingues MartinsMárcio Ajudarte LopesPablo Agustin VargasLuiz Paulo KowalskiAlan Roger Dos Santos SilvaPublished in: Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology (2023)
The models were not able to generalize enough to be applied in real-life datasets due to an overlapping of features between the two classes (i.e., high risk and low risk of malignization).