Evaluation of a machine learning algorithms for predicting the dental age of adolescent based on different...

Area of Science:

• Forensic odontology
• Machine learning applications
• Radiographic analysis

Background:

• Machine learning (ML) algorithms are crucial for dental age estimation.
• This study evaluates three ML models for dental age prediction using distinct preprocessing techniques.

Purpose of the Study:

• To compare the accuracy of different ML models and preprocessing methods for dental age estimation.
• To identify the optimal ML approach for accurate dental age determination.

Main Methods:

• Analyzed 748 digital panoramic radiographs (ages 5-13) from Eastern China.
• Applied Decision Tree (DT), Bayesian Ridge Regression (BRR), and K-Nearest Neighbors (KNN) models.
• Utilized Cameriere and Demirjian methods for data extraction and employed metrics like R², ME, RMSE, MSE, and MAE for evaluation.

Main Results:

• ML algorithms significantly impact dental age prediction accuracy.
• The KNN model, utilizing the Cameriere method, achieved the highest accuracy.
• Key performance metrics for KNN (Cameriere) included ME = 0.015, MAE = 0.473, MSE = 0.340, RMSE = 0.583, and R² = 0.94.

Conclusions:

• Dental age prediction accuracy is influenced by both ML algorithms and preprocessing methods.
• The KNN model combined with the Cameriere method offers a highly accurate approach for dental age estimation in clinical settings.