Abstract
Artificial intelligence (AI) as a means of enhancing and improve the early detection and diagnosis of oral cancer. While traditional diagnostic methods are effective, they are inherently subjective, not widely accessible, and often result in detection only at later stages of disease progression. The literature over the past few years has importantly identified that AI models, in particular deep learning and convolutional neural networks, displayed high diagnostic accuracy on clinical, histopathological and optical imaging data. However, challenges exist in the form of variability in the data sets, limited external validation in clinical practice and or interpretability of AI models for clinical use. In conclusion, AI presents a compelling opportunity as a supportive adjunct for oral oncology, although standardized validation and real-world implementation should occur before widespread utilization.