LEVERAGING DATA SCIENCE FOR EARLY DETECTION OF DENTAL HEALTH ISSUES IN UNDERSERVED COMMUNITIES

Abstract

Early detection of dental health issues is critical for preventing disease progression and reducing healthcare disparities, particularly in underserved communities where access to routine dental care is limited. This study presents a data science–driven framework for the early identification and risk stratification of common dental health problems using community-level demographic, behavioral, and basic clinical data. The proposed methodology integrates systematic data preprocessing, feature engineering, and supervised machine learning models to classify individuals into low, moderate, and high dental risk categories. Multiple algorithms, including Logistic Regression, Decision Tree, Random Forest, and Gradient Boosting, were implemented and comparatively evaluated using standard performance metrics. Experimental results demonstrate that ensemble-based models achieve superior predictive performance, with Gradient Boosting attaining the highest accuracy and recall, highlighting its effectiveness in identifying high-risk individuals. The findings indicate that the proposed framework can serve as a scalable, low-cost decision support tool for community dental screening programs, enabling early referral, improved resource allocation, and enhanced preventive oral healthcare delivery in underserved populations