Validity of dental findings for identification by postmortem computed tomography

Abstract

Postmortem computed tomography (PMCT) images help identify individuals and extract information from corpses. PMCT may substitute for a standard examination when bodies are severely damaged or when resources are limited in a mass fatality incident. In such situations, the dental information revealed by PMCT has the potential to narrow down candidates for identification further. However, the validity of the dental findings obtained from PMCT images remains unclear. In this study, we investigated the validity of dental findings on PMCT images compared to regular dental examinations as the reference standard. We routinely collected PMCT images of 148 unidentified corpses and compared the dental findings of each tooth obtained from PMCT with those of a regular dental examination. The validity of dental findings of PMCT was measured by sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). We also evaluated the accuracy of dental findings per corpse using 32 teeth as the denominator. The dental findings with high sensitivity and high specificity were a present tooth (0.96 and 0.97), a missing tooth (0.97 and 0.97), pontic (0.92 and 1.00), a dental implant (1.00 and 1.00), and a root filling (0.94 and 0.99). The mean accuracy of the dental findings per corpse in PMCT was 95.6% (standard deviation: 6.9, minimum ≤ median ≤ maximum: 65.6 ≤ 100 ≤ 100). The number of corpses with 100% accuracy in the dental findings was 81 (54.7%). The information obtained in this study highlights the potential use of PMCT during human identification in several settings with limited resources, such as the number of specialists present and the condition of the corpses.