Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional study

Author(s):Show less -

Saldivia-Siracusa, Cristina ; Souza, Eduardo Santos Carlos de ; Silva, Arnaldo Vitor Barros da ; Araujo, Anna Luiza Damaceno ; Pedroso, Caique Mariano ; Silva, Tarcilia Aparecida da ; Ana, Maria Sissa Pereira Sant ' ; Fonseca, Felipe Paiva ; Pontes, Helder Antonio Rebelo ; Quiles, Marcos G. ; Lopes, Marcio Ajudarte ; Vargas, Pablo Agustin ; Khurram, Syed Ali ; Pearson, Alexander T. ; Lingen, Mark W. ; Kowalski, Luiz Paulo ; Hunter, Keith D. ; Carvalho, Andre Carlos Ponce de Leon Ferreira de ; Santos-Silva, Alan Roger

• Document type: Journal article
• Source: LANCET REGIONAL HEALTH-AMERICAS; v. 47, p. 13-pg., 2025-07-01.

Abstract

Background Artificial Intelligence (AI) models hold promise as useful tools in healthcare practice. We aimed to develop and assess AI models for automatic classification of oral potentially malignant disorders (OPMD) and oral squamous cell carcinoma (OSCC) clinical images through a Deep Learning (DL) approach, and to explore explain-ability using Gradient-weighted Class Activation Mapping (Grad-CAM). Methods This study assessed a dataset of 778 clinical images of OPMD and OSCC, divided into training, model optimization, and internal testing subsets with an 8:1:1 proportion. Transfer learning strategies were applied to pre-train 8 convolutional neural networks (CNN). Performance was evaluated by mean accuracy, precision, recall, specificity, F1-score and area under the receiver operating characteristic (AUROC) values. Grad-CAM qualitative appraisal was performed to assess explainability. Findings ConvNeXt and MobileNet CNNs showed the best performance. Transfer learning strategies enhanced performance for both algorithms, and the greatest model achieved mean accuracy, precision, recall, F1-score and AUROC of 0.799, 0.837, 0.756, 0.794 and 0.863 during internal testing, respectively. MobileNet displayed the lowest computational cost. Grad-CAM analysis demonstrated discrepancies between the best-performing model and the highest explainability model. Interpretation ConvNeXt and MobileNet DL models accurately distinguished OSCC from OPMD in clinical photographs taken with different types of image-capture devices. Grad-CAM proved to be an outstanding tool to improve performance interpretation. Obtained results suggest that the adoption of DL models in healthcare could aid in diagnostic assistance and decision-making during clinical practice. (AU)

• FAPESP's process: 21/14585-7 - Artificial intelligence applied to the clinical and histopathological diagnosis of Head and Neck Cancer
• Grantee: Anna Luiza Damaceno Araujo
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• FAPESP's process: 22/07276-0 - Clinicopathological and digital predictors of recurrence and malignancy of oral leukoplakia and proliferative verrucous leukoplakia: a clinical trial associated with the use of artificial intelligence
• Grantee: Caique Mariano Pedroso
• Support Opportunities: Scholarships in Brazil - Doctorate
• FAPESP's process: 22/13069-8 - ARTIFICIAL INTELLIGENCE FOR CLINICAL AND HISTOPATHOLOGICAL DIAGNOSIS OF INCIPIENT ORAL SQUAMOUS CELL CARCINOMA: A MULTICENTRIC INTERNATIONAL STUDY
• Grantee: Cristina Saldivia Siracusa
• Support Opportunities: Scholarships in Brazil - Doctorate
• FAPESP's process: 24/20694-1 - Development of a Screening Program for Oral Potentially Malignant Disorders and Squamous Cell Carcinoma of the Oral Cavity in Latin America and the Caribbean.
• Grantee: Alan Roger dos Santos Silva
• Support Opportunities: Scholarships abroad - Research