The Role of Digital Technology in the Early Diagnosis and Prediction of Early Childhood Caries: A Literature Review

Early Childhood Caries (ECC) is a common multifactorial disease in children under six years old and is associated with considerable adverse effects on oral health, growth, development, and quality of life. Early detection of ECC is challenging because initial lesions are often subclinical and difficult to identify using conventional diagnostic methods. This study aimed to review and synthesize current evidence on the role of digital technology, particularly artificial intelligence (AI), in supporting the early diagnosis and prevention of ECC. A literature review was conducted using international databases, including Pubmed, Scopus, ScienceDirect and ResearchGate focusing on peer-reviewed English-language articles published within the last five years. Studies involving children under six years of age that applied AI-based technologies for ECC diagnosis or risk prediction were included. The analysis of ten selected articles demonstrated that machine learning and deep learning models, such as convolutional neural networks, vision transformers, and ensemble learning methods, achieved high accuracy, sensitivity, and specificity in detecting advanced carious lesions and predicting ECC risk based on clinical, behavioral, socioeconomic, salivary, and genetic data. However, limitations were consistently observed in the detection of early non-cavitated lesions. Overall, the findings indicate that AI-based digital technologies serve as effective clinical decision support tools that enhance diagnostic accuracy, risk stratification, and preventive planning for ECC. Although AI cannot replace conventional clinical examination, its integration into pediatric dental care holds strong potential to support earlier, more targeted, and personalized prevention strategies for treating dental caries in young children.