Enhancing Age Determination From Panoramic Dental Radiographs Through Machine Learning

In anthropological research, forensic odontology, and clinical dentistry, it is essential to accurately determine the age from dental radiographs. In order to identify age from panoramic dental X-rays, this study suggests a hybrid machine learning method that combines the Synthetic Minority Over-sampling Technique (SMOTE) to handle class imbalance, Linear Discriminant Analysis (LDA) for feature extraction, and XGBoost for classification. There are 947 samples in the dataset, which are divided into nine different age groups. By creating synthetic samples for minority classes, SMOTE improves model performance and ensures balanced training. LDA facilitates effective data representation by lowering dimensionality while maintaining the most discriminative features, which enhances classification even further. After that, XGBoost is used to classify the retrieved features, utilizing gradient boosting to enhance classification accuracy and optimize decision-making. The suggested hybrid model outperforms traditional standalone techniques with a remarkable accuracy of 94.89%. These results demonstrate how data balancing, feature extraction, and machine learning classifiers work together to estimate age, providing a reliable and automated solution for use in forensic science, pediatric dentistry, and legal investigations.