A Comprehensive Hybrid Approach for Predicting Oral Health Status Based on Different Factors Through New Artificial Intelligence Techniques | ||
| Egyptian Dental Journal | ||
| Volume 70, Issue 4 - Serial Number 1, October 2024, Pages 3063-3073 PDF (778.46 K) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/edj.2024.300530.3096 | ||
| Authors | ||
| Marwa mohamed sabry* 1; Asser Mohamed Gad2 | ||
| 1Lecturer of Pediatric Dentistry & Dental Public Health, Kafr El Sheikh University | ||
| 2Lecturer of orthodontics, Faculty of Dentistry, KafrelSheikh University, KafrelSheikh, Egypt | ||
| Abstract | ||
| ABSTRACT Purpose: This study investigated the feasibility of using Artificial Intelligence (AI) techniques such as Logistic Regression, Decision Tree, Support Vector Machine, and Random Forest to predict oral health status based on factors like health habits, orthodontic treatment, Body Mass Index (BMI), cholesterol, smoking, dental sealants, tooth decay, fluoride, and oral hygiene. Material and Methods: The study used two datasets - an online open-access dataset from Kaggle for model training and testing, as well as data from Kafr El-Sheikh University Hospital. Exploratory Data Analysis (EDA) techniques had been used to examine the data, followed by the AI algorithms. The predictive models were evaluated using cross-validation to assess their accuracy and generalizability. Results: The study achieved high prediction accuracy, around 90%, for both the online dataset and the Kafr El-Sheikh University Hospital data. The AI-based models had outperformed traditional regression methods in predicting oral health status. Conclusion: This study demonstrated the potential of AI-powered predictive models to accurately identify individuals at high risk for poor oral health outcomes. The integration of AI into oral healthcare had the potential to improve preventative care strategies and address oral health disparities within communities. | ||
| Keywords | ||
| Oral health; Public health factors; dental habits Predictive modeling; Artificial intelligence | ||
|
Statistics Article View: 119 PDF Download: 95 |
||
