The Detection and Diagnosis of Dental Caries through Artificial Intelligence: A Comprehensive Systematic Review of Current Models and Their Clinical Applications

Abstract

Background: Dental caries (DC) is a prevalent oral disorder affecting billions worldwide, leading to significant health burdens and economic costs. Traditional methods of detecting and diagnosing DC, including visual examinations and radiographic evaluations, often suffer from variability in accuracy and reliability. Recent advancements in artificial intelligence (AI) have shown promise in enhancing diagnostic capabilities in various medical fields, prompting interest in its application for dental caries.

Methods: This systematic review analyzed literature from 2000 to 2023, utilizing databases such as PubMed, Google Scholar, and Scopus. Key terms included ‘dental caries,’ ‘artificial intelligence,’ ‘machine learning,’ and ‘diagnosis.’ Studies were included if they reported on AI-based models for detecting, diagnosing, or predicting DC.

Results: The review identified numerous AI models demonstrating superior accuracy in diagnosing DC compared to traditional methods. For instance, deep learning algorithms achieved accuracies ranging from 73.3% to 98.8%, with high sensitivity and specificity across various datasets. AI applications were noted to significantly reduce false negatives and enhance early detection of caries, thereby improving patient outcomes and reducing healthcare costs.

Conclusion: AI models represent a transformative approach to the detection and diagnosis of dental caries, offering enhanced precision and efficiency. Integrating these technologies into clinical practice can aid dental professionals in making more informed decisions, thereby improving treatment quality. However, the variability in dataset sizes and characteristics necessitates further research to optimize AI performance and generalizability across diverse populations.