Table 3.

Characteristics of studies.

	Author	Year	Study Type	Algorithm	Objective	Outcome	Author’s Observation
1	Liu et al. [12]	2022	Cross-sectional	CNNs	Develop a semi-automatic model to detect ectopic eruption of maxillary first molars in 4–9-year-olds’ radiographs	High sensitivity and specificity in automated screening	The algorithm may enhance clinical diagnosis and management of ectopic eruption
2	Wu et al. [16]	2021	Cross-sectional	ANNs	Create an ML model to identify caries-related oral microbes in mother–child dyads	Desirable results for both mothers and children	Further refinement needed by considering more variables
3	Park et al. [13]	2021	Cross-sectional	ANNs	Predict early childhood caries using ML-based AI models (XGBoost, random forest, and Light GBM algorithms)	Favorable performance in dental caries prediction with satisfactory AUC values	Helpful in identifying high-risk groups and applying preventive measures
4	Pang et al. [19]	2021	Cross-sectional	ANNs	Develop a caries risk prediction model for teenagers by considering environmental and genetic factors	Accurate identification of individuals at high and very high risk of developing caries	Potential as a powerful tool for performing community-level high caries risk identification
5	Karhade et al. [14]	2021	Cross-sectional	ANNs	Evaluate the accuracy of an automated ML algorithm for early childhood caries (ECC) classification	Comparable performance to that of the reference model (AUC: 0.74, sensitivity: 0.67, PPV: 0.64)	Valuable tool for ECC screening
6	Ramos-Gomez et al. [15]	2021	Cross-sectional	ANNs	Identify survey items to predict dental caries in children using a machine learning algorithm	Algorithm toolkits can help dental professionals to assess children’s oral health	Demonstrates potential for dental caries screening in children