Applications of AI-based Deep Learning Models for Detecting Dental Caries on Intraoral Images – A Systematic Review

Abstract

The detection of dental caries, a prevalent oral health issue, remains a critical component of effective dental care. Traditional methods of diagnosis, such as visual inspection and radiographic analysis, are often limited by subjectivity and variability. In recent years, the integration of Artificial Intelligence (AI) and deep learning models has shown significant promise in enhancing the accuracy, speed, and consistency of dental caries detection. This paper systematically reviews the application of AI-based models, particularly Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), and Transfer Learning models, in detecting dental caries from intraoral images. The review highlights the strengths and limitations of these models, providing a comprehensive analysis of their performance metrics, including accuracy, sensitivity, specificity, and area under the curve (AUC). While AI models, especially CNNs, have demonstrated superior performance compared to traditional diagnostic methods, challenges such as dataset bias, generalization, and model interpretability persist. The analysis also emphasizes the need for larger and more diverse datasets to improve model robustness and reduce bias across different demographic groups. Furthermore, the review explores the ethical considerations surrounding AI in dental diagnostics, including the importance of data privacy and transparency in decision-making. In conclusion, AI-based models hold transformative potential in dental caries detection, offering faster, more accurate diagnoses and improving clinical workflows. However, overcoming current limitations, such as dataset diversity and model explainability, is essential for broader adoption in clinical practice. Future research should focus on expanding datasets, improving model interpretability, and developing hybrid models to enhance performance across diverse clinical settings. Ultimately, AI has the potential to significantly enhance global dental health and make caries detection more accessible to underserved populations, improving both clinical outcomes and patient care.