This study reports the prevalence of organic and inorganic gunshot residue (OGSR & IGSR) in various populations using emerging analytical methods.
The survey includes over 3,200 samples from six subpopulations. The known shooter samples used various ammunition types: leaded (1), lead-free (2), and a mixture (3) collected as soon as the firearm was discharged or after conducting various post-shooting tasks (4). Background samples originated from individuals who had not fired a weapon in the past 24-hours, separated by low-risk (5) and those with higher exposure to sources that mimic GSR, high-risk (6). Samples were analyzed by LIBS followed by electrochemistry in under 10 min per sample. Lastly, a subset from each population was cross-validated by SEM-EDS for morphological and elemental information, demonstrating LIBS and electrochemistry benefits for fast learning about the composition of large populations. An exploratory method was used to identify the various subgroups' population characteristics and estimate performance rates, with an accuracy of 89.6% when LIBS and EC were combined. In addition, Neural Networks were used to recognize underlying patterns in the data for group classifications, with accuracy>93%. Probabilistic outputs allowed for likelihood ratios (LR) calculations with low rates of misleading evidence (RoME less than 0.8%). The log10 LRs typically ranged from [-5, 0] for non-shooters and [2,9] for shooters, demonstrating good discrimination between the overall populations and a sensible model metric for reporting the weight of the evidence. This study addresses forensic community needs regarding the analysis of GSR evidence and provides a path forward. (Publisher abstract provided)
Popular TopicsForensic sciences Gunshot residue Evidence analysis
- High-throughput quantitative binding analysis of DNA aptamers using exonucleases
- Evaluating the Validity and Reliability of Textile and Paper Fracture Characteristics in Forensic Comparative Analysis
- Research to Develop Validated Methods for THC Quantification in Complex Matrices by High-resolution DART-MS-Focus on Edibles and Plant Materials