Proposal Description

Forensic statistics is an interesting and influential area of statistics, but relatively unknown. Dramatic depictions of how crimes rarely investigate the statistics behind police and laboratory procedures. Civil litigation, such as personal injury claims or commercial disputes also rely on statistics. The role of machine learning or AI in the prevention of crime and management of the justice system is a matter of current debate.

Presenter: Dr Joe Marsh

University of Birmingham, UK

Title: Estimating the number of modern slavery victims or illicit drug users

Estimating the size of hidden populations, such as victims of modern slavery or individuals engaging in illicit drug use, is a long-standing methodological challenge with significant implications for crime science and forensic statistics. Multiple Systems Estimation (MSE) is a widely adopted capture-recapture technique used to estimate these dark figures of crime by analysing incomplete, overlapping administrative lists. However, real-world forensic and sociological datasets are frequently imperfect, suffering from censoring and missingness due to stringent privacy concerns.

Operating at the intersection of computer science, mathematics, and social sciences, this paper proposes a simulation-based Bayesian inference framework utilising Neural Bayes Estimators (NBE) and Neural Posterior Estimators (NPE). By leveraging modern computational power, these neural networks bypass explicit likelihood calculations by learning to approximate posterior quantities from synthetic data generated by model simulators. Crucially, these neural methods are amortised: once the computationally intensive training is completed "offline," they provide instantaneous, highly efficient posterior estimates for any new dataset.

This "train once, infer many times" capability makes neural estimators ideal for secure or trusted research environments where sensitive crime data is stored but computational resources are limited. Through extensive simulation studies and real-world applications, including estimating the prevalence of modern slavery in the UK and female drug use in North East England, we demonstrate that neural estimators achieve accuracy comparable to MCMC while being orders of magnitude faster. Ultimately, this framework offers a robust, AI-driven solution to convergence failures in sparse settings, providing policymakers and justice systems with a powerful tool to reliably quantify hidden populations and inform evidence-based interventions.

Presenter: Julia Mortera,

Università Roma Tre, Italy

Title: Statistical issues in investigation of suspected medical misconduct

Justice systems are sometimes called upon to evaluate cases in which health care professionals are suspected of harming or killing their patients. These cases are difficult to evaluate since investigators need to consider whether the deaths that prompted the investigation could plausibly have occurred for reasons other than homicide, in addition to considering whether, if homicide was indeed the cause, the person under suspicion is responsible.

Suspicions about medical murder sometimes arise due to a surprising or unexpected series of events, such as an apparently unusual number of deaths among patients under the care of a particular healthcare worker. But also, a single unexpected event might trigger suspicion about a particular healthcare worker, and this might then lead to investigation of events which happened when he/she was thought to be present. We show that an apparently striking association between a healthcare’s presence and a high rate of deaths in a hospital ward can easily be completely spurious.

There is a statistical challenge of distinguishing event clusters that arise from criminal acts from those that arise coincidentally from other causes.

The talk will be based on the RSS report 2022 Health care serial killer or co-incidence? and Dotto, Gill, Mortera (2022) Statistical analyses in the case of an Italian nurse accused of murdering patients.

Presenter: Dr Rowland Seymour

University of Birmingham, UK

Title: Network models to tackle human trafficking for child sexual exploitation.

Effective police intervention strategies and national crime strategies to combat human trafficking for child sexual exploitation material production require accurate prevalence estimates. In this talk, I will describe the results of a survey method called the network scale-up method (NSUM), where survey respondents were asked how many people do you know who traffic children to produce CSAM. This means they can provide information about those in their social network without disclosing personally identify information or their own experiences. NSUM models often necessitate standalone surveys for each geographic region, escalating costs, and complexity. I will introduce a partially pooled NSUM model, using a hierarchical Bayesian framework that efficiently aggregates and utilizes data across multiple regions without increasing sample sizes. I developed this model for a novel national survey dataset from the Philippines and I will demonstrate its ability to produce detailed municipal-level prevalence estimates of trafficking for CSEM production. Our results not only underscore the model’s precision in estimating hidden populations but also highlight its potential for broader application in other areas of social science and public health research, offering significant implications for resource allocation and intervention planning.

Presenter: Professor Hutton

University of Warwick, UK

Title: Comparison of legal and statistical approaches to causation.

Law courts have to decide whether an action or exposure to an agent such as radiation or chemical substances have lead to a personal injury. The legal systems of Canada, Australia, South Africa, Ireland and the UK have similar methods for establishing responsibility for injury. A common legal approach when there is uncertainty about exposure levels for an individual claimant is to rely on confident medical opinion.

Statistical approaches, particularly those developed in health research, evaluate risks and causes at population levels. In an important UK judgement expressed the judges' dislike of the ``use of statistics alone", and tried to distinguish between "fact probability" and "belief probability" (Sienkiewicz v Greif (UK) Ltd [2011] UKSC 10). How might statisticians respond to experts who assert that there is no need for statistics, and that the use of large databases and AI are sufficient to answer questions about causes?

This talk will illustrate the debates with reference to high profile cases: exposure to asbestos, metal-on-metal hip replacements and effectiveness of breast cancer surgery.