Related Experiment Video
Updated: Jun 19, 2025
07:53
Cutting Procedures, Tensile Testing, and Ageing of Flexible Unidirectional Composite Laminates
Published on: April 27, 2019
8.2K
A proposal for cut marks classification using machine learning: Serrated vs. non-serrated, single vs. double-beveled
Giada Sciâdi Steiger1, Matteo Borrini1
1School of Biological and Environmental Sciences, Liverpool John Moores University, Liverpool, UK.
Journal of Forensic Sciences
|July 25, 2024
Summary
This study introduces a standardized method and terminology for analyzing sharp force trauma on bones. Machine learning models accurately classify knife characteristics from cut marks, aiding forensic analysis.
Area of Science:
- Forensic Anthropology
- Forensic Science
- Bioarchaeology
Background:
- Standardization is lacking in analyzing sharp force trauma on skeletal remains.
- Existing methodologies for cut mark analysis lack consistent terminology.
- Accurate identification of tool marks is crucial in forensic investigations.
Purpose of the Study:
- To present a standardized classification method for cut marks on human bones.
- To establish applicable methodology and terminology for sharp force trauma examination.
- To develop machine learning models for objective analysis of stab wounds.
Main Methods:
- 350 cut marks were produced on pig ribs using seven different knives.
- Analysis involved stereomicroscopy with tangential lighting.
- Eleven key traits of cut marks were identified and used to train binary logistic regression models.
Main Results:
- Eleven traits significantly associated with knife type were identified.
- Machine learning models achieved 63%-85% accuracy for serration classification.
- Models achieved 63%-89% accuracy for blade bevel classification.
Conclusions:
- A novel set of traits for cut mark analysis has been proposed.
- Machine learning models can standardize and facilitate the analysis of stab wounds.
- This methodology enhances the objective examination of sharp force trauma in forensic cases.
