Age Estimation: A Dual Pipeline For Teeth Detection And Apex-Height Measurement Using The Cameriere’s Method

Abstract

The present work proposes a new way of determining the actual age of the child with the help of Cameriere’s method and combined with techniques of radiographic examination. Pivotal to this method is training one ‘model network’, YOLOv8, alongside another separate ‘model network’, YOLOv8. The first model needs to detect and label all required teeth for analysis, at the same time the second model is aimed at emphasizing some crucial anatomical landmarks such as Apices of the detected teeth and the total height of these formations. This integration of the two models enables a full understanding of the CBCT images required when using the Cameriere’s method to obtain accurate apex distances and tooth height measurements. These two YOLOv8 pipelines lead to an effective end-to-end system that detects the age of patients from X-ray images with outstanding precision. In contrast to traditional age estimation methods based on approximate manual measurements by observers, this method minimizes inter-observer differences and improves accuracy. The large ACC obtained in this study and specifically in children suggest high potential for this method in different practical fields like forensic anthropology and pediatric dentistry. The proposed method, which enhances the logistical efficiency of age estimation, is expected to enhance decision-making in more clinical and legal situations that involve children, thus likely to result in enhanced outcomes in children healthcare and justice-related issues.