A novel method for pair-matching using three-dimensional digital models of bone: mesh-to-mesh value comparison

Area of Science:

• Forensic Anthropology
• Bioarchaeology
• Digital Forensics

Background:

• Commingled human remains present significant challenges in forensic identification.
• Current methods for sorting skeletal elements are limited, hindering accurate individualization.

Purpose of the Study:

• To introduce and evaluate a novel digital method, mesh-to-mesh value comparison (MVC), for pair-matching commingled human skeletal remains.
• To assess the accuracy of both manual and automated versions of the MVC method.

Main Methods:

• Development of two versions (manual and automated) of the mesh-to-mesh value comparison (MVC) technique.
• Utilizing digital three-dimensional bone models for geometric comparison.
• Testing the methods on humeri and assessing performance using sensitivity and specificity.

Main Results:

• The manual MVC method achieved 100% sensitivity and 100% specificity in pair-matching humeri.
• The automated MVC method demonstrated 95% sensitivity and 60% specificity.
• The MVC method shows significant potential for accurately matching commingled skeletal elements.

Conclusions:

• Mesh-to-mesh value comparison (MVC) is a powerful new tool for the accurate pair-matching of commingled skeletal elements.
• The manual MVC method is highly accurate, while the automated version requires further refinement for optimal performance.