This is the Final Research Report on a project with the goal of developing a comprehensive approach to address inefficiencies and unreliable results in the screening of firearm discharge residues.
The project’s central hypothesis was that electrochemical detection (EC) techniques in combination with laser-induced breakdown spectroscopy (LIBS) will provide screening methods that are faster, more selective, and more informative than current laboratory-based and field tests. Project objectives were 1) to develop versatile and fast methodologies for firearm discharge residues (FDR) detection on hands and other target materials and 2) to apply statistical methods for the probabilistic assessment and interpretation of the evidence. These objectives were achieved through four tasks. One task was the collection of a large sample set of FDRs from target materials (fabrics), hands of shooters, and background data. A second task validated LIBS and electrochemical sensors for estimation of shooting distance and FDR detection on clothing. The third task was to validate screening methods (LIBS and ECD) and confirmatory methods (scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) for inorganic gunshot residue (IGSR) and organic gunshot residue (OGSR) detection on hands of potential shooters. The fourth task was the validation of statistical data pre-processing, multivariate analysis, and machine learning algorithms to generate decision thresholds and probabilistic approaches for interpretation of the significance of the evidence. The results show that the use of fast emerging methods and the OGSR/OGSE data derive from this large population study, combined with probabilistic interpretation, can provide more comprehensive tools for assessing GSR evidence. This opens new venues to respond to the court-relevant questions regarding whether a person of interest fired a gun. 37 figures, 10 tables, and 46 references
Downloads
Similar Publications
- A Feature Mapping Technique for Complex Data Object Generation With Likelihood and Deep Generative Approaches
- Enhancing Fault Ride-Through Capacity of DFIG-Based WPs by Adaptive Backstepping Command Using Parametric Estimation in Non-Linear Forward Power Controller Design
- Direct Comparison of Body Fluid Identification Technologies