A novel identification method using perceptual degree of concordance of occlusal surfaces calculated by a Python...

Area of Science:

• Forensic Dentistry
• Biometrics
• Computational Biology

Background:

• Victim identification is a critical challenge in mass disaster response.
• Dental records, specifically molar morphology, offer a unique identifier.
• Existing methods may lack efficiency or simplicity for rapid identification.

Purpose of the Study:

• To develop a simple, accurate method for individual identification using molar occlusal morphology.
• To leverage computational techniques for matching antemortem and postmortem dental data.
• To assess the efficacy of a new identification system in a simulated disaster scenario.

Main Methods:

• Development of a Python-based program utilizing the perceptual Hash (pHash) function.
• Calculation of Hamming distance (HD) to quantify differences between antemortem data (AMD) and postmortem data (PMD).
• Analysis of 2,215 dental models (AMD) and 17 postmortem models (PMD) with temporal variations.

Main Results:

• Over 90% of postmortem models (16 out of 17) were ranked within the top 5% similarity.
• The system demonstrated high accuracy in matching dental records.
• Identification using a single molar proved challenging, but multiple teeth improved accuracy.

Conclusions:

• The developed system offers a simple and effective method for individual identification from molar data.
• The approach shows promise for improving victim identification efficiency in mass disaster scenarios.
• Further verification using multiple teeth can significantly enhance identification accuracy.