Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context
Summary: 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 L...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | The Lancet Regional Health. Americas |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667193X25001486 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850183058486984704 |
|---|---|
| author | Cristina Saldivia-Siracusa Eduardo Santos Carlos de Souza Arnaldo Vitor Barros da Silva Anna Luíza Damaceno Araújo Caíque Mariano Pedroso Tarcília Aparecida da Silva Maria Sissa Pereira Sant'Ana Felipe Paiva Fonseca Hélder Antônio Rebelo Pontes Marcos G. Quiles Marcio Ajudarte Lopes Pablo Agustin Vargas Syed Ali Khurram Alexander T. Pearson Mark W. Lingen Luiz Paulo Kowalski Keith D. Hunter André Carlos Ponce de Leon Ferreira de Carvalho Alan Roger Santos-Silva |
| author_facet | Cristina Saldivia-Siracusa Eduardo Santos Carlos de Souza Arnaldo Vitor Barros da Silva Anna Luíza Damaceno Araújo Caíque Mariano Pedroso Tarcília Aparecida da Silva Maria Sissa Pereira Sant'Ana Felipe Paiva Fonseca Hélder Antônio Rebelo Pontes Marcos G. Quiles Marcio Ajudarte Lopes Pablo Agustin Vargas Syed Ali Khurram Alexander T. Pearson Mark W. Lingen Luiz Paulo Kowalski Keith D. Hunter André Carlos Ponce de Leon Ferreira de Carvalho Alan Roger Santos-Silva |
| author_sort | Cristina Saldivia-Siracusa |
| collection | DOAJ |
| description | Summary: 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 explainability 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. Funding: This work was supported by FAPESP (2022/13069-8, 2022/07276-0, 2021/14585-7 and 2024/20694-1), CAPES, CNPq (307604/2023-3) and FAPEMIG. |
| format | Article |
| id | doaj-art-d1e7e281e4ee489f82489f8f4f19de97 |
| institution | OA Journals |
| issn | 2667-193X |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | The Lancet Regional Health. Americas |
| spelling | doaj-art-d1e7e281e4ee489f82489f8f4f19de972025-08-20T02:17:28ZengElsevierThe Lancet Regional Health. Americas2667-193X2025-07-014710113810.1016/j.lana.2025.101138Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in contextCristina Saldivia-Siracusa0Eduardo Santos Carlos de Souza1Arnaldo Vitor Barros da Silva2Anna Luíza Damaceno Araújo3Caíque Mariano Pedroso4Tarcília Aparecida da Silva5Maria Sissa Pereira Sant'Ana6Felipe Paiva Fonseca7Hélder Antônio Rebelo Pontes8Marcos G. Quiles9Marcio Ajudarte Lopes10Pablo Agustin Vargas11Syed Ali Khurram12Alexander T. Pearson13Mark W. Lingen14Luiz Paulo Kowalski15Keith D. Hunter16André Carlos Ponce de Leon Ferreira de Carvalho17Alan Roger Santos-Silva18Departamento de Diagnóstico Oral, Faculdade de Odontologia de Piracicaba, Universidade Estadual de Campinas, Piracicaba, São Paulo, BrazilInstitute of Mathematics and Computer Sciences, University of São Paulo, São Carlos, São Paulo, BrazilInstitute of Science and Technology, Federal University of São Paulo, São José dos Campos, São Paulo, BrazilHead and Neck Surgery Department, University of São Paulo Medical School, São Paulo, Brazil; Hospital Israelita Albert Einstein, São Paulo, BrazilDepartamento de Diagnóstico Oral, Faculdade de Odontologia de Piracicaba, Universidade Estadual de Campinas, Piracicaba, São Paulo, BrazilDepartment of Oral Surgery and Pathology, School of Dentistry, Federal University of Minas Gerais (UFMG), Belo Horizonte, Minas Gerais, BrazilDepartment of Oral Surgery and Pathology, School of Dentistry, Federal University of Minas Gerais (UFMG), Belo Horizonte, Minas Gerais, BrazilDepartment of Oral Surgery and Pathology, School of Dentistry, Federal University of Minas Gerais (UFMG), Belo Horizonte, Minas Gerais, BrazilService of Oral Pathology, João de Barros Barreto University Hospital, Federal University of Pará, Belém, BrazilInstitute of Science and Technology, Federal University of São Paulo, São José dos Campos, São Paulo, BrazilDepartamento de Diagnóstico Oral, Faculdade de Odontologia de Piracicaba, Universidade Estadual de Campinas, Piracicaba, São Paulo, BrazilDepartamento de Diagnóstico Oral, Faculdade de Odontologia de Piracicaba, Universidade Estadual de Campinas, Piracicaba, São Paulo, BrazilUnit of Oral and Maxillofacial Pathology, School of Clinical Dentistry, University of Sheffield, Sheffield, UKSection of Haematology/Oncology, Department of Medicine, University of Chicago, Chicago, IL, USA; University of Chicago Comprehensive Cancer Center, Chicago, IL, USADepartment of Pathology, The University of Chicago Medicine, Chicago, IL, USAHead and Neck Surgery Department, University of São Paulo Medical School, São Paulo, Brazil; Department of Head and Neck Surgery and Otorhinolaryngology, A.C. Camargo Cancer Center, São Paulo, State of São Paulo, BrazilLiverpool Head and Neck Center, ISMIB, University of Liverpool, Liverpool, UKInstitute of Mathematics and Computer Sciences, University of São Paulo, São Carlos, São Paulo, BrazilDepartamento de Diagnóstico Oral, Faculdade de Odontologia de Piracicaba, Universidade Estadual de Campinas, Piracicaba, São Paulo, Brazil; Corresponding author. Oral Diagnosis Department, Piracicaba Dental School, State University of Campinas (UNICAMP), Av. Limeira, no 901, Areão, Piracicaba, São Paulo, 13414-903, Brazil.Summary: 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 explainability 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. Funding: This work was supported by FAPESP (2022/13069-8, 2022/07276-0, 2021/14585-7 and 2024/20694-1), CAPES, CNPq (307604/2023-3) and FAPEMIG.http://www.sciencedirect.com/science/article/pii/S2667193X25001486Artificial intelligenceDeep learningArtificial neural networkOral cancerHead and neck cancerPrecancerous conditions |
| spellingShingle | Cristina Saldivia-Siracusa Eduardo Santos Carlos de Souza Arnaldo Vitor Barros da Silva Anna Luíza Damaceno Araújo Caíque Mariano Pedroso Tarcília Aparecida da Silva Maria Sissa Pereira Sant'Ana Felipe Paiva Fonseca Hélder Antônio Rebelo Pontes Marcos G. Quiles Marcio Ajudarte Lopes Pablo Agustin Vargas Syed Ali Khurram Alexander T. Pearson Mark W. Lingen Luiz Paulo Kowalski Keith D. Hunter André Carlos Ponce de Leon Ferreira de Carvalho Alan Roger Santos-Silva Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context The Lancet Regional Health. Americas Artificial intelligence Deep learning Artificial neural network Oral cancer Head and neck cancer Precancerous conditions |
| title | Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context |
| title_full | Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context |
| title_fullStr | Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context |
| title_full_unstemmed | Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context |
| title_short | Automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework: a cross-sectional studyResearch in context |
| title_sort | automated classification of oral potentially malignant disorders and oral squamous cell carcinoma using a convolutional neural network framework a cross sectional studyresearch in context |
| topic | Artificial intelligence Deep learning Artificial neural network Oral cancer Head and neck cancer Precancerous conditions |
| url | http://www.sciencedirect.com/science/article/pii/S2667193X25001486 |
| work_keys_str_mv | AT cristinasaldiviasiracusa automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT eduardosantoscarlosdesouza automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT arnaldovitorbarrosdasilva automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT annaluizadamacenoaraujo automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT caiquemarianopedroso automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT tarciliaaparecidadasilva automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT mariasissapereirasantana automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT felipepaivafonseca automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT helderantoniorebelopontes automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT marcosgquiles automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT marcioajudartelopes automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT pabloagustinvargas automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT syedalikhurram automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT alexandertpearson automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT markwlingen automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT luizpaulokowalski automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT keithdhunter automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT andrecarlosponcedeleonferreiradecarvalho automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext AT alanrogersantossilva automatedclassificationoforalpotentiallymalignantdisordersandoralsquamouscellcarcinomausingaconvolutionalneuralnetworkframeworkacrosssectionalstudyresearchincontext |