Artificial Intelligence-Based Evaluation of Oral Potentially Malignant Disorders Using Photographic and Clinical Data: A Retrospective Exploratory Diagnostic Accuracy Study.
- 1Interdisciplinary Department of Medicine, University of Bari "Aldo Moro", 70124 Bari, Italy.
- 2Multidisciplinary Department of Medical-Surgical and Dental Specialties, University of Campania "Luigi Vanvitelli", 80138 Naples, Italy.
Abstract
Objective: The objective of this study was to evaluate the diagnostic performance of GPT-5.2 Thinking in identifying oral potentially malignant disorders within a patient sample that also included oral carcinomas and other oral cavity lesions, using intraoral photographs and associated clinical data. Methods: This retrospective exploratory diagnostic accuracy study included 103 oral lesions: 69 oral potentially malignant disorders, 25 oral carcinomas, and 9 other oral pathologies. Each case was assessed in three independent sessions, and the majority diagnosis was compared with the final clinical and/or histopathological reference diagnosis. Results: For oral potentially malignant disorders, sensitivity was 92.75%, specificity was 88.24%, accuracy was 91.26%, positive predictive value was 94.12%, negative predictive value was 85.71%, and Cohen's kappa was 0.804. Sensitivity was 81.82% for homogeneous leukoplakia, 78.57% for non-homogeneous leukoplakia, and 86.36% for oral lichen planus. For oral carcinomas, sensitivity was 84.00%, specificity was 97.44%, accuracy was 94.17%, positive predictive value was 91.30%, negative predictive value was 95.00%, and Cohen's kappa was 0.837. Conclusions: GPT-5.2 Thinking showed promising diagnostic performance in this exploratory case-enriched dataset, but predictive values require validation in representative clinical populations.
Related
- PO022 | ARTIFICIAL INTELLIGENCE–ASSISTED PREDICTION OF THROMBOTIC AND HEMORRHAGIC COMPLICATIONS IN PATIENTS TREATED WITH VITAMIN K ANTAGONISTS: A RETROSPECTIVE EXPLORATORY STUDY
- Dentists’ acceptance of artificial intelligence-assisted decision support for oral potentially malignant disorder detection: a structural equation modeling study
- Convergent Validity and Exploratory Discrimination of Two-Dimensional Facial Photographic Measures for Clinician-Rated Facial Asymmetry in Young Mongolian Adults
- Intelligent Cloud Resource Usage Potentially to Improve Task Scheduling by the use of Artificial Intelligence.
- Self-Perceived Knowledge, Self-Efficacy, and Reported Integration of Generative Artificial Intelligence Among University Faculty: An Exploratory Profile Analysis
BibTeX
@article{falco2026artificial,
title = {{Artificial Intelligence-Based Evaluation of Oral Potentially Malignant Disorders Using Photographic and Clinical Data: A Retrospective Exploratory Diagnostic Accuracy Study.}},
author = {Domenico De Falco and Fabiana Pinto and Alberta Lucchese and Massimo Petruzzi},
journal = {Dentistry journal},
year = {2026},
month = oct,
doi = {10.3390/dj14100657},
url = {https://europepmc.org/article/MED/42930423},
note = {CC BY},
}
RIS
TY - JOUR
TI - Artificial Intelligence-Based Evaluation of Oral Potentially Malignant Disorders Using Photographic and Clinical Data: A Retrospective Exploratory Diagnostic Accuracy Study.
AU - Domenico De Falco
AU - Fabiana Pinto
AU - Alberta Lucchese
AU - Massimo Petruzzi
PY - 2026
DA - 2026/10/10
JO - Dentistry journal
DO - 10.3390/dj14100657
UR - https://europepmc.org/article/MED/42930423
AB - Objective: The objective of this study was to evaluate the diagnostic performance of GPT-5.2 Thinking in identifying oral potentially malignant disorders within a patient sample that also included oral carcinomas and other oral cavity lesions, using intraoral photographs and associated clinical data. Methods: This retrospective exploratory diagnostic accuracy study included 103 oral lesions: 69 oral potentially malignant disorders, 25 oral carcinomas, and 9 other oral pathologies. Each case was assessed in three independent sessions, and the majority diagnosis was compared with the final clinical and/or histopathological reference diagnosis. Results: For oral potentially malignant disorders, sensitivity was 92.75%, specificity was 88.24%, accuracy was 91.26%, positive predictive value was 94.12%, negative predictive value was 85.71%, and Cohen's kappa was 0.804. Sensitivity was 81.82% for homogeneous leukoplakia, 78.57% for non-homogeneous leukoplakia, and 86.36% for oral lichen planus. For oral carcinomas, sensitivity was 84.00%, specificity was 97.44%, accuracy was 94.17%, positive predictive value was 91.30%, negative predictive value was 95.00%, and Cohen's kappa was 0.837. Conclusions: GPT-5.2 Thinking showed promising diagnostic performance in this exploratory case-enriched dataset, but predictive values require validation in representative clinical populations.
SN - 2304-6767
ER -