Hamadan University of Medical Sciences Avicenna Journal of Dental Research 2423-7582 18 2 2026 06 30 Using Artificial Intelligence in the Diagnosis, Classification, and Prognosis of Oral Squamous Cell Carcinoma: A Systematic Review and Meta-Analysis 107 117 10.34172/ajdr.4429 EN Mohammad Mahdi Maleki https://orcid.org/0009-0009-4434-4634 Samaneh Vaziriamjad https://orcid.org/0000-0003-2220-9975 Fatemeh Shahbazi https://orcid.org/0000-0003-2320-4885 REVIEW 10.34172/ajdr.4429 2025 06 16 2026 05 31 Introduction: Oral squamous cell carcinoma (OSCC) is a prevalent malignancy within the head and neck region. Nonetheless, the overall prognosis remains poor, primarily due to the late-stage diagnosis in a substantial proportion of patients. Thus, this study aimed to critically evaluate the application of artificial intelligence (AI) in the diagnostic, classificatory, and prognostic processes of OSCC. Methods: PubMed, Scopus, Web of Science and Embase databases were searched up to December 2024. Included studies reported relevant metrics (true positive, false positive, true negative, false negative, sensitivity, specificity, and area under the curve [AUC]). The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Results: The search identified 1,863 articles, with 27 meeting the inclusion criteria. AI models achieved a pooled sensitivity of 0.90 (95% CI: 0.81–0.95), specificity of 0.89 (95% CI: 0.83–0.93), and an AUC of 0.733 (95% CI: 0.679–0.786). Diagnostic applications had the highest sensitivity, while classification tasks demonstrated superior specificity. Ultimately, prognostic models displayed lower sensitivity and specificity, indicating variability in performance across different AI applications. Conclusion: Rapid development of AI algorithms has facilitated their application in OSCC diagnosis, classification, and prognosis. Additionally, AI models have shown high sensitivity and specificity in diagnostic applications, promising earlier detection and improved classification accuracy, which can lead to superior therapeutic outcomes. Moreover, their non-invasive nature compared to histopathological biopsy enhances patient comfort while reducing costs. Overall, AI holds significant potential to transform OSCC diagnosis and prognosis, finally affecting community oral health.