Journal of Forensic Research

Anomaly detection in digital forensics data is critical for identifying suspicious activities and potential security breaches. This mini-review investigates the efficacy of a novel Support Vector Neural Network (SVNN) for anomaly detection in digital forensics datasets. By examining recent literature, this article elucidates the principles of SVNN, its advantages over traditional methods, and its application in detecting anomalous behavior in various forensic scenarios. Furthermore, it discusses challenges, opportunities, and future directions for enhancing anomaly detection using SVNN in digital forensics investigations.

Teeth remains are often the only means of positive identification in an unidentified body after being subjected to high temperature injury. The routine use of dental implants made them as an active contributing element to the identification of unidentified cadavers. Their resistance to prolonged high temperature might make them a substantial contributor to the identification of high temperature burned bodies. The aim of the study was to observe the effects of high temperature on three dental implants of distinct elemental composition: C1-MIS®, made of Ti grade 23; BL- Straumann®, made of Ti grade 4 and of a Titanium-Zirconium alloy; Roxolid-Straumann®, by detecting the changes of their microstructural and elemental composition. Scanning Electron Microscopy and Electron Dispersive Spectroscopy were used to characterize the surface structure and elemental composition of the implants before and after implants being subjected to a high temperature protocol of 1375 °C for 30 minutes. Macroscopic and microscopic changes in the samples after exposure to the high temperature. Dental implants demonstrated specific macroscopic changes and microstructural deteriorations, after exposure to high temperature. After exposure to high temperature, dental implants demonstrated specific macroscopic changes and microstructural deteriorations. Although several changes occurred in the elemental content of the materials, the original elemental composition was preserved. The ability to discriminate between dental implants by elemental analyses can have a determinant impact on the identification process of burned bodies.

Insulin overdose is a critical concern in forensic investigations due to its potential lethality and widespread availability. This mini-review explores the biochemical and toxicological effects of insulin overdose in rats, offering insights from a forensic perspective. By examining relevant literature, this article elucidates the mechanisms of insulin toxicity, including hypoglycemia, neurologic impairment, and cardiovascular complications. Additionally, it discusses forensic considerations such as postmortem findings, analytical techniques for insulin detection, and interpretation challenges in forensic pathology. Understanding the intricacies of insulin overdose in animal models provides valuable insights for forensic scientists and medical examiners in cases involving suspected insulin-related fatalities.

Public-related cybercrime is a new type of crime that relies on science and technology, and has the characteristics of public-interested, cross-regional, and virtuality compared with traditional crimes. Punishment by the law is inseparable from evidence, and public-related cybercrime mainly occurs in the virtual space, which makes it significantly more challenging to investigate and collect evidence. Gathering evidence for public-related cybercrime, many problems have emerged, mainly including the lack of professionalism of the forensic subject, the non-standard forensic procedures, and the impact on the evidentiary ability and probative power of digital evidence. At present, these problems have seriously hindered the comprehensive, objective, and timely collection of evidence by the judicial authorities to crack down on public-related cybercrime. The primary evidence for determining the public-related cybercrime is digital evidence, so the discussion of the evidence collection and issues of the public-related cybercrime revolves around digital evidence. In response to problems in the process of evidence collection for public-related cybercrime, it shall be based on the characteristics of the cybercrime itself, the evidence collection system, and combined with the characteristics of digital evidence, such as the volatility, virtuality, and easy and accurate reproducibility of digital evidence, to find targeted and effective countermeasures. The targeted measures mainly include that increasing the professionalism of the forensic subject, promoting the censorship of forensic procedures, clarifying the legal principles of collecting evidence in cybercrimes, improving the mechanisms for the assessment and preservation of evidence and improving the application of relevant rules on primary evidence.

Objective: Identifying the origin of human biological traces detected at crime scenes by comparing DNA profiles to national or international forensic databases is often key to provide new orientations to police investigations. However, when unknown profiles are established, investigators can benefit from forensic genetics to propose new leads, for example by predicting the physical appearance of individuals. Since blood traces are of primary interest for forensic investigators and often lead to the extraction of usable genetic material, in this study, we developed a methodology to predict the biological age from blood samples based on the analysis of DNA methylation of human genomic regions.

Methods: We first established a cohort of blood samples obtained from 170 French donors aged from 0 to 101 years old. We analyzed the methylation status of 5 age-associated CpG sites using the SNaPshot method, a primer-extension based assay routinely used in the French forensic police laboratories. Using a training set of 136 samples, we generated an age-prediction model based on multiple regression analyses of DNA methylation data and we tested its predictive performances on a validation set.

Results: The SNaPshot assay was adapted to limiting quantities of genomic DNA relevant for forensic investigations. The DNA methylation levels were established for 5 age-related CpG sites in 170 blood samples collected from French male and female donors. We established a statistical model optimized for 5 CpG sites that can explain 97% of age variation with a Mean Absolute Error (MAE) of 3.45 years between the estimated biological and chronological age of individuals.

Conclusion: We developed an approach to predict the biological age of individuals strictly based on the methylation levels of 5 CpG sites from circulating blood samples and that is compatible with routine genetic analyses in French forensic police laboratories.