Tri-dimensional pattern analysis of foodstuff bitemarks - A pilot study of tomographic database.

The analysis of bitemarks has been mostly done through analogical methods. The advent of current technologies brings new and more reliable tools for analyzing evidence. Bitemarks can now be turned into more consistent evidence by using a digital tridimensional (3D) analysis. The aim of this study is to propose a reliable, accurate and monitorized approach of 3D bitemark analysis. 12 cranium Cone Beam Computed Tomographic (CBCT) files were randomly selected from a clinical database (Coimbra Hospital and University Center/Faculty of Medicine of the University of Coimbra). The individuals were recalled to bite an apple (golden delicious 65/80) immediately subjected to a CBCT scan. The 3D rendering of every bitemark was compared with the 3D upper dental arches obtained from the CBCT cranium scans of the simulated suspects. The research team was composed by 5 elements. The matching process consists of corresponding landmark points, in both the bitemark and the suspects' dentition (upper dental arch). 144 comparisons were obtained. The Kappa statistics analysis was performed. Cohen's Kappa values were obtained between 0,690 and 0,910. Fleiss Kappa obtained a value of 0,767. The Friedman's test was performed and the normality assumption was not verified (p>0,05). A non-destructive protocol of bitemark study by 3D analysis of tomographic volumes was presented. The statistic analysis supports its accuracy and reliability. This experimental study opens doors to a future where bitemark analysis could be done through software that searches for matching dentitions in a CBCT database.