Objectifying evidence evaluation for gunshot residue comparisons using machine learning on criminal case data

An overview of log likelihood ratio cost in forensic science - Where is it used and what values can we expect?

There is increasing support for reporting evidential strength as a likelihood ratio (LR) and increasing interest in (semi-)automated LR systems. The log-likelihood ratio cost (Cllr) is a popular metric for such systems, penalizing misleading LRs further from 1 more. Cllr = 0 indicates perfection while Cllr = 1 indicates an uninformative system. However, beyond this, what constitutes a "good" Cllr is unclear. Aiming to provide handles on when a Cllr is "good", we studied 136 publications on (semi-)automated LR systems. Results show Cllr use heavily depends on the field, e.g., being absent in DNA analysis. Despite more publications on automated LR systems over time, the proportion reporting Cllr remains stable. Noticeably, Cllr values lack clear patterns and depend on the area, analysis and dataset. As LR systems become more prevalent, comparing them becomes crucial. This is hampered by different studies using different datasets. We advocate using public benchmark datasets to advance the field.

Elemental Characterization of Leaded and Lead-Free Inorganic Primer Gunshot Residue Standards Using Single Particle Inductively Coupled Plasma Time-of-Flight Mass Spectrometry

This study describes the use of single particle inductively coupled plasma time-of-flight mass spectrometry (spICP-TOFMS) for the detection and classification of inorganic gunshot residue (IGSR) particles. To establish reliable multi-element criteria to classify IGSR particles, leaded and lead-free IGSR reference materials were analyzed, and the elemental compositions of the individual particles were quantified. The results suggest that expanded element compositions may be used to classify IGSR particles via spICP-TOFMS compared to those used in conventional IGSR analysis using scanning electron microscopy energy dispersive X-ray spectroscopy (SEM-EDS). For spICP-TOFMS analysis of leaded IGSR particles, classification may be based on the presence of lead (Pb), antimony (Sb), and barium (Ba) just as in SEM-EDS; however, additional particle types, such as lead-copper (Pb-Cu) particles, contribute significantly (∼30%) to the leaded IGSR particle population. In lead-free IGSR particles, the dominate multi-metal particle composition found is titanium-zinc (Ti-Zn) with a conserved Zn:Ti ratio of 1.4:1, but other elements, such as copper (Cu), are also characteristic. In mixtures of the two IGSR reference materials, we were able to classify over 80% of the multi-metal particles detected with no false-positive particle-type assignments. With spICP-TOFMS, particles smaller than those typically measured by SEM-EDS are detected, with estimated median diameters for leaded and lead-free IGSR of 180 and 320 nm, respectively. Through measuring these smaller particles, up to ∼two times more particles per mL are recorded by spICP-TOFMS compared to that found by SEM-EDS. Overall, high-sensitivity and high-throughput analysis using spICP-TOFMS enables quantitative, rapid multi-elemental characterization, and classification of individual IGSR particles.

A strawman with machine learning for a brain: A response to Biedermann (2022) the strange persistence of (source) “identification” claims in forensic literature

We agree wholeheartedly with Biedermann (2022) FSI Synergy article 100222 in its criticism of research publications that treat forensic inference in source attribution as an “identification” or “individualization” task. We disagree, however, with its criticism of the use of machine learning for forensic inference. The argument it makes is a strawman argument. There is a growing body of literature on the calculation of well-calibrated likelihood ratios using machine-learning methods and relevant data, and on the validation under casework conditions of such machine-learning-based systems.