Teeth: Our personal ID

Introduction: In forensic odontology, individuality or unique features are important concepts for antemortem (AM) and post-mortem (PM) dental record comparisons. There are various studies on the individuality of soft tissue and hard tissue structures like rugoscopy, cheiloscopy, enamel rod patterns, etc. However, the data is scarce for studies specifically based on the individuality of single tooth morphology. Objective: To determine the individuality of morphology of the maxillary first and second premolars (PM1 and PM2) using AI segmented three-dimensional (3D) CBCT scans. Methodology: Sixty CBCT scans (30 male and 30 female) were selected for segmentation. Inclusion criteria were healthy fully developed PM1 and PM2. Exclusion criteria were teeth with any anomaly or trauma obscuring the tooth morphology, restored teeth, and distorted scans. Maxillary arch teeth were segmented from the remaining CBCT scan using AI segmentation tool and saved as STL files, mimicking AM data. The same CBCT scans were then segmented again to mimic PM data. Next, PM1 and PM2 were segmented from the STL files using 3-Matic software. The AM and PM sets of scans were decoded by examiner A. One hundred and twenty pairs of each PM1and PM2 were made by examiner A and superimposed by examiner B using CloudCompare software. The pairs included 60 matched and 60 non-matched pairs for each tooth type. The decision for the pairs was based on root mean square (RMS) values calculated. Results: There was statistically significant difference between the RMS of matched and non-matched pairs for both PM1 and PM2. Based on the threshold of RMS value, Examiner B gave the correct decision for all the pairs, and PM1 and PM2 showed 100% uniqueness of their morphology. Conclusions: Maxillary PM1 and PM2 exhibit the uniqueness of tooth morphology and thus may be used for human identification.