Revisiting Age Estimation on Panoramic Dental Images

Forensic dentistry has traditionally relied on bone or dental indicators, primarily utilizing dental radiographs, for age estimation. However, limited research has been conducted on automatic age estimation on panoramic images, needing a reeval-uation of the existing methodologies to assess the performance of computer-based methods. This study proposes to revisit the analysis of age estimation methods using panoramic dental radio-graphs. We have curated the largest publicly available dataset of panoramic dental images, encompassing diverse dental conditions and age ranges. Specifically, our study focuses on evaluating three distinct classes of deep-learning architectures: ViT, ConvNeXt-V2, and EfficientNets, employing a comprehensive to assess their performances that better favor reproducibility. By comparing our approach with existing studies in the literature, we offer valuable insights for forensic investigations in the field of age estimation.

I. Introduction

Forensic science uses age estimation to identify unaccom-panied refugees who may be minors, individuals undergoing adoption processes, or bodies in an advanced state of de-composition with no proven age [1]. Particularly in dentistry, teeth and jaws are key indicators of age due to their physical durability and ability to withstand forces longer than other tissues in the body. Consequently, they have become a primary focus of research in determining an individual's age, alongside various other methods and structures available for examination [2]. These indicators follow a defined progression during youth development, thus allowing manual inspection methods to ac-curately estimate the age. Estimating age accurately becomes more challenging when dealing with adults or seniors. In these cases, achieving the same level of precision becomes more difficult.