An Explainable Machine Learning Approach for Age and Gender Estimation in Living Individuals Using Dental Biometrics

Machine learning can estimate age and gender from dental measurements with moderate accuracy: The study found that the XGB model achieved an F1 score of 73.26% for age estimation using dental radiographs, suggesting the technology works but has meaningful limitations that should be considered before real-world deployment.

Explainability is essential for responsible use in forensic and medical contexts: The researchers specifically developed an explainable AI model using SHAP technology so that "dental experts can make judicious decisions based on comprehensible insight," highlighting that accuracy alone is insufficient—decision-makers need to understand why the system makes its predictions.

Ensemble approaches combining multiple specialized models may improve reliability: Rather than relying on a single model, the researchers developed a novel technique using multiple models each tailored to distinct dental metrics, suggesting that combining different approaches could reduce errors and increase trustworthiness in high-stakes applications like forensic investigations.

Responsible deployment requires human expertise in the loop: The emphasis on enabling "dental experts" to make decisions—rather than automating decisions entirely—reflects a responsible AI principle: these systems should augment professional judgment rather than replace it, particularly in forensic and legal contexts where accuracy has serious consequences.