Dental age estimation in forensic odontology: a systematic review comparing traditional methods and artificial intelligence approaches

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Dental age estimation is a fundamental tool in forensic odontology, with important applications in human identification and legal contexts. This systematic review evaluated the accuracy, objectivity, and reproducibility of traditional dental age estimation methods compared with artificial intelligence–based approaches. A search was conducted in PubMed, Scopus, and Web of Science including studies published between 2015 and 2025, selecting 16 studies that analysed traditional methods, machine learning, and deep learning models. Artificial intelligence–based approaches, particularly machine learning and deep learning, demonstrated higher accuracy with lower mean absolute error (MAE) values compared with traditional methods such as Demirjian, Cameriere, and Kvaal. In addition, these models reduced intra- and inter-observer variability, improving objectivity and reproducibility. Despite these promising results, methodological heterogeneity and the need for large, well-structured datasets remain important limitations. Artificial intelligence represents a valuable and promising tool for improving dental age estimation in forensic practice.