Objectives: Oral squamous cell carcinoma (OSCC) presents a significant global health issue due to its high morbidity and mortality rates, mainly resulting from late-stage diagnoses. Artificial intelligence (AI) technologies, including deep learning algorithms and imaging techniques, offer transformative potential for the early detection and management of OSCC. This review assesses the feasibility, diagnostic performance, and future directions of AI-driven approaches in OSCC care.

Materials and Methods: A review of 19 articles was conducted using several databases. The search strategy included keywords such as "oral cancer," "artificial intelligence," and "deep learning." The inclusion criteria followed the PECOS framework, focusing on studies related to AI applications in OSCC diagnosis and management. Diagnostic accuracy, sensitivity, and specificity data were extracted and analysed for clinical relevance.

Results: The review indicated a diagnostic accuracy of 92.2%, with sensitivity and specificity reaching 100%. Advanced AI tools, including deep learning algorithms and imaging techniques, were used to analyse histopathological and photographic data. However, differences in datasets and methodologies limited direct comparison across studies, emphasising the need for standardisation. The findings highlight the importance of standardised datasets and validation protocols to improve the reliability and scalability of AI in OSCC detection. Emerging techniques, such as multi-task learning and ensemble models, show promise for enhancing diagnostic precision. 

Conclusions: Incorporating AI into interdisciplinary care models can further facilitate early diagnosis and optimise patient outcomes. AI-driven technologies have the potential to revolutionise OSCC detection by improving diagnostic accuracy and enabling early intervention. Ongoing research and development are crucial to refining AI applications, ensuring effective integration into clinical practice, and significantly improving patient outcomes.