A probabilistic model for murder weapon identification using stab-marks in human ribs.

Area of Science:

• Forensic Science
• Biomechanical Engineering
• Statistical Modeling

Background:

• Stab-mark analysis is crucial in forensic investigations.
• Deterministic models have limitations in identifying stabbing weapons.
• Geometric characteristics of stab-marks offer potential for weapon identification.

Purpose of the Study:

• To establish a scientific and statistical basis for identifying stabbing weapons using stab-mark geometry on human ribs.
• To develop a quantitative predictive model for stab-mark analysis.
• To utilize Bayesian inference for probabilistic weapon identification.

Main Methods:

• Developing a quantitative predictive model based on geometric measurements of stab-marks and their rib location.
• Employing a general method based on Bayesian inference and probabilities.
• Collecting and analyzing stab-mark information from human ribs.

Main Results:

• The developed probabilistic model achieved high accuracy in identifying the murder weapon.
• The model successfully distinguished between two macroscopically identical blades with microscopic differences.
• Over 90% correct identification was achieved in distinguishing between similar blades.

Conclusions:

• Geometric stab-mark analysis provides a reliable method for murder weapon identification.
• Probabilistic models offer superior accuracy compared to deterministic approaches in forensic stab-mark analysis.
• This research enhances the scientific basis for forensic investigations involving stabbing incidents.