Journal Article

From images to detection: Machine learning for blood pattern classification

Li, Yilin, Shen, Weining

2025

Overview

Bloodstain Pattern Analysis (BPA) helps us understand how bloodstains form, with a focus on their size, shape, and distributions. This aids in crime scene reconstruction and provides insight into victim positions and crime investigation. One challenge in BPA is distinguishing between different types of bloodstains, such as those from firearms, impacts, or other mechanisms. Our study focuses on differentiating impact spatter bloodstain patterns from gunshot backward spatter bloodstain patterns. We distinguish patterns by extracting well-designed individual stain features, applying effective data consolidation methods, and selecting boosting classifiers. As a result, our model exhibits competitive accuracy and efficiency on the tested dataset, suggesting its potential in similar scenarios. •A novel method distinguishes gunshot and impact bloodstain patterns using images.•Ellipse- and shade-based features improve interpretability and classification accuracy.•XGBoost achieves 92.89% accuracy, outperforming previous BPA models.•A new Stability Importance Score offers consistent feature ranking across model runs.•The method shows strong potential for practical forensic bloodstain analysis.

• Publisher: Elsevier B.V,Elsevier Limited
• Subject: Accuracy / Blood Stains / Bloodstain pattern analysis / Boosting Machine Learning Algorithms / Classification / Crime / Crime Victims / Criminal statistics / Discriminant analysis / DNA methylation / Feature extraction / Firearms / Forensic Medicine - methods / Forensic sciences / Forensic statistics / Humans / Image processing / Image Processing, Computer-Assisted - methods / Machine learning / Pattern analysis / Pattern classification / Pattern Recognition, Automated / Random forest / XGBoost