Deep learning applied to the histopathological diagnosis of ameloblastomas and ameloblastic carcinomas.
Daniela Giraldo RoldánErin Crespo Cordeiro RibeiroAnna Luíza Damaceno AraújoPaulo Victor Mendes PenafortViviane Mariano da SilvaJeconias CâmaraHelder Antonio Rebelo PontesManoela Domingues MartinsMárcio Campos OliveiraAlan Roger Dos Santos SilvaMárcio Ajudarte LopesLuiz Paulo KowalskiMatheus Cardoso MoraesPablo Agustin VargasPublished 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 demonstrated a strong potential of learning, but lack of generalization ability. The models learn fast, reaching a training accuracy of 98%. The evaluation process showed instability in validation; however, acceptable performance in the testing process, which may be due to the small data set. This first investigation opens an opportunity for expanding collaboration to incorporate more complementary data; as well as, developing and evaluating new alternative models.