3D Dental Biometrics: Transformer-based Dental Arch Extraction and Matching

Abstract

The dental arch is a significant anatomical feature that is crucial in assessing tooth arrangement and configuration and has a potential for human identification in biometrics and digital forensic dentistry. In a previous study, we proposed an auto pose-invariant arch feature extraction Radial Ray Algorithm (RRA) and a matching framework [1] based solely on 3D dental geometry. To enhance the identification accuracy and speed of our previous work, we propose in this study a transformer architecture that can extract dental keypoints by encoding both local and global features. The dental arch is then constructed through robust interpolation of the dental keypoints using B-Spline and is compared using the same identification framework. To evaluate the effectiveness of our proposed approach, we conducted experiments by matching the same 11 post-mortems (PM) samples against 200 antemortem (AM) samples. Our results show that our approach achieves higher accuracy and faster speed compared to our previous work. Specifically, 11 samples (100%) achieved a top 6.5% (13/200) accuracy out of the 200-rank list, compared to the top 15.5% (31/200) accuracy previously [1]. Additionally, the time required to identify a single subject from 200 subjects has been reduced from 5 minutes to 3 minutes. The dental arch can be used as a powerful filter feature. Our findings make a significant contribution to the existing literature on dental identification and demonstrate the potential practical applications of our approach in diverse fields such as biometrics, forensic dentistry, orthodontics, and anthropology.