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

Area of Science:

• Forensic Odontology
• Digital Imaging
• Evidence Analysis

Background:

• Traditional bitemark analysis relies on analog methods, which can be subjective and less reliable.
• Advancements in digital technology offer new tools for more accurate forensic evidence examination.
• Digital tridimensional (3D) analysis provides a more consistent method for evaluating bitemark evidence.

Purpose of the Study:

• To propose a reliable, accurate, and monitorized approach for 3D bitemark analysis.
• To establish a non-destructive protocol for bitemark examination using tomographic volumes.
• To assess the accuracy and reliability of 3D bitemark analysis in forensic investigations.

Main Methods:

• Randomly selected 12 cranium Cone Beam Computed Tomographic (CBCT) files from a clinical database.
• Recalled individuals to bite an apple, immediately followed by a CBCT scan to capture the bitemark.
• Compared 3D renderings of bitemarks with 3D upper dental arches from simulated suspects' CBCT scans.
• Utilized landmark point correspondence for matching bitemarks with dentition.

Main Results:

• Performed 144 comparisons between bitemarks and dental arches.
• Achieved Cohen's Kappa values ranging from 0.690 to 0.910, indicating substantial to almost perfect agreement.
• Obtained a Fleiss Kappa value of 0.767, signifying good inter-rater reliability.
• Statistical analysis supported the accuracy and reliability of the 3D bitemark analysis protocol.

Conclusions:

• A non-destructive protocol for 3D bitemark analysis using tomographic volumes has been successfully presented.
• The statistical analysis validates the accuracy and reliability of this digital 3D approach.
• This study paves the way for future software-driven bitemark analysis using CBCT databases for matching dentitions.