Trustworthy Data-driven Chronological Age Estimation from Panoramic Dental Images

This article explores the integration of artificial intelligence (AI) in dental age estimation, focusing on the development of a trustworthy and interpretable system called AgeX. The study addresses the challenges of transparency and explainability in deep learning models, particularly in high-risk healthcare applications. It introduces AgeX, a human-centric intelligent system that generates textual explanations for predictions made by a non-transparent deep learning model. The system combines a convolutional neural network (CNN) for dental age estimation with a surrogate white-box method to enhance interpretability. The article also discusses the EU AI Act's requirements for healthcare AI systems and how AgeX complies with these regulations. Additionally, it presents a statistical validation of the system's performance and a human expert validation to assess the quality of the generated explanations. The study concludes with a discussion of the limitations and future work, highlighting the potential for improving the accuracy and richness of explanations in dental age estimation.