Related Experiment Video
Updated: Jan 17, 2026
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
1.4K
Artificial intelligence based techniques for caries risk prediction and assessment: A scoping review
Sonal Bhatia1, Vinay Kumar Gupta1, Sumit Kumar1
1Department of Public Health Dentistry, King George's Medical University, Lucknow, Uttar Pradesh, India.
Journal of Oral Biology and Craniofacial Research
|September 25, 2025
Summary
Artificial intelligence (AI) shows promise in dental caries risk assessment (CRA) and prediction (CRP). Machine learning algorithms, particularly logistic regression and random forest, demonstrate superior performance compared to traditional methods for identifying at-risk individuals.
Area of Science:
- Dentistry
- Artificial Intelligence
- Machine Learning
- Public Health
Background:
- Dental caries remains a significant public health concern, necessitating accurate risk assessment and prediction.
- Traditional methods for caries risk assessment (CRA) may have limitations in accuracy and efficiency.
- The integration of artificial intelligence (AI) offers potential advancements in dental diagnostics and personalized risk evaluation.
Purpose of the Study:
- To systematically review the existing evidence on the application of AI for caries risk assessment (CRA) and prediction (CRP).
- To identify the scope of methodologies employed in AI-based CRA/CRP.
- To summarize the performance metrics, limitations, and challenges of these AI applications.
Main Methods:
- A comprehensive search of MEDLINE, EMBASE, and Google Scholar databases was conducted for studies published between 2013 and 2023.
- Study selection involved title, abstract, and full-text screening based on predefined criteria.
- Data extraction utilized a custom checklist comprising eight dimensions to chart study characteristics and findings.
Main Results:
- The review included 13 articles from an initial pool of 3059 retrieved studies.
- Logistic regression and random forest were the most frequently employed AI methods.
- Performance metrics were heterogeneous, with sensitivity ranging from 0.59 to 0.996 and specificity from 0.531 to 0.943. Common predictors included socio-demographic, oral hygiene, and dietary factors.
Conclusions:
- Machine learning algorithms are predominantly used in AI-based CRA models.
- AI methods exhibit potentially superior specificity and overall performance compared to traditional approaches.
- AI applications in CRA hold significant implications for preventing avoidable chronic diseases like dental caries.
