Dental Age Estimation in Forensic Medicine: A Comprehensive Review of Methods and Recent AI Developments

Abstract

Age estimation remains a cornerstone in forensic science, paediatric dentistry, and legal investigations, with dental development serving as a reliable biological indicator. Conventional approaches—such as Demirjian’s, Willems’, and Cameriere’s methods—rely on visual assessment of dental radiographs but are often time-consuming and subject to observer variability. Recent advances in artificial intelligence, particularly deep learning using convolutional neural networks (CNNs), have automated and enhanced age estimation by analysing panoramic dental images with high precision. These AI-driven models demonstrate remarkable accuracy, in some cases achieving mean absolute errors of less than one year, and can identify developmental stages of dental germs at a level comparable to expert clinicians. Advanced frameworks, such as 3D CNNs applied to cone-beam computed tomography (CBCT), provide volumetric insights into root and pulp development, thereby improving predictive performance. Innovations such as multi-task learning, temporal modelling, and advanced feature extraction have strengthened reproducibility and scalability while explainability techniques like Grad-CAM enhance transparency and support forensic defensibility. This review underscores AI-driven dental age estimation as a promising alternative to traditional methods, offering automation, enhanced accuracy, interpretability, and potential for regulatory acceptance. A wide range of articles and books were reviewed to assess current practices and emerging trends in dental age estimation.