Predicting Dental Anxiety and Cooperative Behavior in Children Using Machine Learning: A Cross-Sectional Predictive...

Area of Science:

• Pediatric Dentistry
• Artificial Intelligence in Healthcare
• Behavioral Science

Background:

• Dental anxiety and uncooperative behavior are significant challenges in pediatric dental care.
• These behaviors can negatively impact treatment outcomes and children's oral health.
• Early identification and management are crucial for effective pediatric dental treatment.

Purpose of the Study:

• To evaluate machine learning models for predicting dental anxiety and uncooperative behavior in children aged 6-11.
• To estimate continuous dental anxiety scores and identify key predictors.
• To explore behavioral subgroups using unsupervised learning techniques.

Main Methods:

• An analytical cross-sectional study involving 952 children.
• Utilized Logistic Regression for classification and Random Forest for regression and feature importance.
• Applied PCA and K-Means clustering for behavioral subgroup analysis.

Main Results:

• Machine learning models achieved high accuracy: 0.92 for dental anxiety and 0.91 for cooperative behavior.
• Random Forest Regressor predicted anxiety scores with R² = 0.97.
• Sensory and cognitive responses were identified as primary predictors; two distinct behavioral profiles emerged.

Conclusions:

• Machine learning models show strong predictive performance for dental anxiety and cooperation in pediatric patients.
• These models offer potential for early detection and personalized management strategies.
• Further validation is recommended prior to widespread clinical implementation.