Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 12 appraisals
Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology
Handwritten signatures in neurological and psychiatric disorders: a systematic review with clinical and forensic implications.
BACKGROUND: Signature production represents a complex motor and cognitive behavior, highly automatized and repeated countless times throughout life. It ranks among the most durable graphic forms, often preserved even when writing processes are impaired. Studies examining signature characteristics in clinical cohorts are limited. The available evidence raises questions about its potential as a behavioral marker of neurological and psychiatric disorders. METHODS: We conducted a PRISMA-compliant systematic review up to 4th December 2025 to summarize evidence on how neurological and psychiatric conditions may affect signature features. Eligible articles on this topic were searched in Scopus, Web of Science, and PubMed databases. Sixteen papers were included, and their results were systematically described and qualitatively analyzed. RESULTS: The available literature suggests that signature production appears relatively independent of global cognitive decline, although it is sensitive to specific pathological conditions (e.g., micrographia in Parkinson's disease, increased tortuosity in cerebral small vessel disease, and alterations observed in mania, delirium, obsessive-compulsive disorder, and major depressive disorder). Alterations commonly affect kinematic parameters, as well as visuospatial and constructional features. However, no disorder-specific signature pattern emerges from current evidence, largely due to the methodological heterogeneity across studies. CONCLUSIONS: The handwritten signature holds promise as a behavioral marker of brain disorders, serving as a potentially useful, complementary - rather than a stand-alone - indicator of neurological and psychiatric conditions, which could be potentially useful in clinical settings. The implications for forensic applications remain largely theoretical and have not yet been standardized.
30 July 2026
Read appraisal →PloS one
New integrative methodology for the excavation and study of mass graves and partially commingled remains
The resolution of commingled human skeletal remains is a phenomenon increasingly encountered by archaeologists and anthropologists. Its difficulty is due to the complexity of the task itself, but also to the limited methodological resources available for researchers. Parallelly, its importance derives from the fact that individualized skeletons are a prerequisite for any anthropological analysis that aims to provide results. In this paper, we put forward a new methodological approach that aims to integrate different tools and techniques already available. The basis of this methodology is the analysis of the spatial relationships of the human skeletal remains. To facilitate its use and maximize its effectiveness, we have developed a set of specific tools and protocols that include a list of information to be directly recorded in Geographical Information System (GIS) software, as well as a new coding system for skeletal remains showing different levels of completeness. Results show that the application of such a methodology helped in the resolution of a large, commingled assemblage exhumed from a 16th-century-CE burial located in southwestern Hungary. The original number of complete skeletons (those individualized in their entirety during the fieldwork) was substantially increased, and for several other skeletal elements, potential matches were found during the laboratory work. We found this methodology to be a good filter prior to the utilization of the more standardized techniques such as osteometric sorting, articulation, visual pair-matching or process of elimination, as it would increase their effectiveness. Additionally, the novel coding system has proven to be useful in inventorying and storing human skeletal remains. This project shows that the addition of GIS technology and spatial analysis does increase the success rate in the reassociation of commingled human skeletal remains.
20 July 2026
Read appraisal →Proceedings of the National Academy of Sciences of the United States of America
Training humans to detect AI-generated faces
As AI-generated faces become indistinguishable from real ones, deepfake technology poses escalating threats to information integrity and security. While algorithms can detect deepfakes, they suffer from opacity and critical vulnerabilities-and training humans to identify specific visual artifacts has proven largely ineffective. Here, we introduce a fundamentally different approach that harnesses people's global impressions of faces. Building on findings that AI and human faces evoke systematically different perceptual impressions (Miller et al., 2023), we trained participants to attend to these distinguishing qualities without explicit instruction on how to use them. Using a rigorous pre-post design with untrained test faces, we demonstrate that all participants (N = 45) improved, with mean accuracy nearly doubling. High performers achieved near-perfect detection, and participants developed metacognitive insight, showing appropriate confidence calibration only after training. A test-retest control study ruled out practice effects as an explanation for training gains, and an online replication demonstrates the scalability of our approach. As the technology advances, targeting systematic biases that are inherent to generative AI, and which manifest in global features, may offer a more durable defense than approaches reliant on image artifacts alone.
30 June 2026
Read appraisal →Molecular biology reports
Update of Mexican population databases for complex kinship interpretation with twelve and seven X-STRs
BACKGROUND: X-chromosome short tandem repeats (X-STRs) are valuable for resolving complex kinship cases. Most population databases are based on the ArgusX12 kit (Qiagen), using PCR and capillary electrophoresis (CE) to identify length-based (LB) alleles, and are formatted for FamlinkX software. However, haplotype diversity is often underrepresented by population studies, compelling database updates ( https://famlink.se/fx_databases.html ). The ForenSeq DNA Signature Prep kit (Verogen) uses massive parallel sequencing (MPS) to analyze 7 X-STRs and report sequence-based (SB) alleles, for which population databases are currently unavailable. Because these 7 markers are a subset of the ArgusX12 panel, existing LB-based databases can support kinship interpretation of MPS-derived data. METHODS AND RESULTS: We therefore updated a Mexican population database for 12 X-STRs and generated a corresponding 7-marker subset with FamlinkX-compatible files. Using 500 haplotypes from the Investigator Argus X-12 kit QS (Qiagen), we estimated forensic parameters and evaluated the impact of the updated database by calculating likelihood ratios (LRs) in four representative kinship scenarios. The updated database (n = 1433) increased LRs by an average of 55% compared to the previous version (n = 933). CONCLUSIONS: These results support improved forensic interpretation in Mexican and Latin American populations lacking comprehensive national X-STR databases based on ArgusX12 and ForenSeq DNA Signature kits.
29 June 2026
Read appraisal →Croatian medical journal
Artificial intelligence enables scale, consistency, and rigor in forensic identity inference.
Advances in forensic genomics, which include massively parallel sequencing, dense single nucleotide polymorphism testing, and forensic genetic genealogy (FGG), have greatly expanded the range of cases in which DNA evidence can generate investigative leads. As a result, the primary limitations in modern forensic DNA analysis are no longer analytical sensitivity or marker availability, but the ability to reason consistently, transparently, and at scale over increasingly complex genetic, genealogical, and contextual information. Current forensic workflows remain predominantly human-centered, relying on manual reasoning that is difficult to standardize, reproduce, document, or scale across growing case inventories. The potential of artificial intelligence (AI) is described as an enabling layer for forensic identity inference, defined here as computational decision-support systems that structure, prioritize, and document reasoning over genetic associations, genealogical structures, and investigative context during identity hypothesis development. While formal statistical inference quantifies evidentiary weight, identity inference governs how evidence is explored, combined, and acted upon during investigations. AI-assisted systems can augment, but not replace, expert judgment by supporting scalable prioritization, relational reasoning, and systematic documentation of analytical decisions, while reducing bias. Properly designed AI-enabled systems offer a path to sustainably scaling FGG while supporting scientific rigor, accountability, and public trust.
15 June 2026
Read appraisal →Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy
Benchmarking spectroscopic-chemometric models for human bloodstain time since deposition in tropical outdoor microenvironments
Reliable estimation of the time since deposition (TSD) of bloodstains remains a key challenge in forensic reconstruction, particularly under realistic outdoor conditions. This study benchmarks visible (Vis) reflectance, attenuated total reflectance-Fourier transform infrared (ATR-FTIR) and micro-Raman spectroscopy, each coupled with partial least squares (PLS) regression and artificial neural networks (ANN), for TSD estimation of human bloodstains aged up to 100 days under naturally fluctuating tropical outdoor microenvironments. Blood from multiple donors was deposited on glass slides and aged under two conditions: open-air exposure and enclosed storage in sealed plastic bags. Spectra were acquired from the outer ring of each stain, preprocessed by standard normal variate transformation and partitioned by donor into training and independent test sets. Vis reflectance primarily tracked haemoglobin chromophore changes via Soret and α/β bands, whereas ATR-FTIR highlighted protein secondary structure and hydration through amide I-III and O-H/N-H stretching features. Micro-Raman spectra emphasised porphyrin breathing and CC stretching modes, together with aromatic amino-acid and amide bands that remained informative even at late ageing stages. Across all techniques, PLS models provided robust quantitative performance (R2P up to 0.98, RMSEP typically ∼5-10 days, RPD > 3), with micro-Raman-PLS giving the highest predictive accuracy and enclosed exposure consistently improving model stability relative to open air. ANN models did not provide systematic advantages over PLS and were more sensitive to condition-specific variance. In the context of previous outdoor ATR-FTIR studies, these results show that established Vis-, ATR-FTIR- and Raman-based chemometric models can deliver competitive TSD estimates when calibrated within clearly defined environmental domains, underscoring the importance of donor-independent and environmental-aware validation for forensic interpretation.
7 June 2026
Read appraisal →Analytica chimica acta
Optimized HSSE-GC-MS and machine learning for forensic gender classification of human scent from clothing
BACKGROUND: Clothes are considered a valuable piece of evidence in forensic investigations because of the underlying information that can be extracted from them. Human scent is one kind of trace evidence that can be extracted from clothes, which can link a specific person or location to the crime. This study optimized a headspace sorptive extraction (HSSE) method coupled with gas chromatography-mass spectrometry in combination with machine learning for objective discrimination of gender from human scent traces on clothing for forensic applications. RESULTS: The HSSE was optimized using the Box-Behnken design and Response Surface Methodology to maximize the Euclidean distance between the total ion sum spectrum (TIS) of two independent samples. The optimized extraction parameters were determined at 100 °C extraction temperature, 180 min extraction time, and 5 × 5 cm cloth dimension at m/z range 101 - 250 which produced an R2 of 0.9874. Method validation demonstrated an acceptable absolute error of 6.60%. Intra- and inter-day precisions were calculated for each m/z from 101 to 250 to remove outliers from the dataset for the multivariate analysis. Using the optimized extraction method, 62 samples were analyzed and the pre-processed TIS from the samples were subjected to various supervised machine learning algorithms. Support vector machine (SVM) and partial least squares-discriminant analysis (PLS-DA) models achieved perfect 100% accuracy for gender classification in both training and test sets. Random forest (RF) and linear discriminant analysis (LDA) also showed high test set accuracies of 93% and 86%, respectively. SIGNIFICANCE: This integrated approach, combining optimized stir bar HSSE-GC-MS with different machine learning algorithms using the TIS untargeted approach, provides a fast and automatic gender classification from human scent traces on clothes that can be used as a rapid screening technique in forensic investigations to narrow down suspect or victim profiles.
22 May 2026
Read appraisal →PloS one
A collaborative approach to applying Natural Language Processing (NLP) to Domestic Homicide Reviews (DHRs): A study protocol
Since 2011, there has been a statutory requirement in England and Wales to conduct a Domestic Homicide Review (DHR) into any domestic abuse-related death: a multi-agency review into the death of a person aged 16 or over that appears resulting from violence, abuse or neglect from an intimate partner, family member or household member. However, analyses of large numbers of DHRs are rare. One of the core challenges is the time and effort required to analyse narrative text within reports. Doing so manually is both time-consuming and resource-intensive and is a primary reason why researchers typically focus on only a portion of the available data. Natural Language Processing (NLP)-a sub-branch of artificial intelligence that enables computers to interpret and process natural language-provides a viable and scalable alternative by offsetting much of the heavy data processing to a computer. This protocol outlines a study to assess the feasibility of applying NLP to DHRs. The study will take a collaborative approach which balances the speed and scale of automation with the embedded knowledge and expertise of practitioners. This approach helps to ensure that outputs of NLP are sensitive and transparent about the biases common within datasets on violence and abuse. Based on initial consultations, we identify the priority research questions for investigation. In addition, we outline details of an ongoing collaboration with one partner, Imkaan. The protocol describes the data access, and retrieval and analysis stages before summarising how feasibility will be evaluated. We anticipate that several challenges will emerge within this study and identify strategies for mitigation. We suggest that working with practitioners who hold deep contextual knowledge about the social realities of violence and abuse, including language, risks, and experiences, mean that tools can be developed that are accountable to communities and appropriately applied to real-world problems.
22 May 2026
Read appraisal →mSystems
Insects shape the cadaver decomposition microbiome and postmortem interval estimation accuracy
The breakdown and recycling of carrion is a crucial ecological process that largely relies on a community of necrophagous insects and microbes. Recent work has shown that a specialized microbial network, likely dispersed throughout the environment by insects, assembles during cadaver decomposition to break down flesh regardless of climate and geography. Because of their broad distribution and successional nature, decomposer microbes have been used in machine learning models to predict the postmortem interval (PMI) of human remains, an important contribution to the field of forensics. How factors such as an indoor environment, which alters insect activity, impact the composition of microbial communities decomposing human remains is unclear. Here, we investigate the effects of enclosed shelter on microbial community assembly and successional patterns during human decomposition and provide important considerations for PMI modeling. Compared to outdoor cadavers, we show that indoor cadavers experienced delayed colonization of key decomposer microbes over the course of decomposition due to restricted insect access. Consequently, machine learning models trained on outdoor cadavers frequently underestimated the PMI of cadavers decomposing indoors. Delayed colonization by blow fly maggots (Diptera: Calliphoridae) was correlated with higher PMI prediction errors, suggesting that insects are an important source of microbial decomposers that drive PMI model predictions. Incorporating microbial data from indoor cadavers and insect activity into PMI models significantly improved prediction capabilities for both indoor and outdoor decomposition environments. Ultimately, we demonstrate the important role insects play in the maintenance and distribution of microbes that help to recycle vertebrate remains.IMPORTANCEMicrobes are critical for the decomposition and recycling of organic matter. Recently, microbiome-based models have shown promising performance in estimating the postmortem interval (PMI). However, many deaths occur indoors, yet no studies have investigated the impact of enclosed shelter on the cadaver microbiome in a controlled setting. Here, cadavers were decomposed indoors, and we found that blow fly maggots serve as an important source of decomposer taxa that significantly alter the cadaver microbiome following infestation. Notably, PMI estimation models trained on outdoor data sets failed to accurately predict the PMI when insect colonization was delayed. We show that incorporating 16S rRNA amplicon data from cadavers decomposing indoors, along with environmental variables, significantly improves PMI estimates, suggesting a microbiome-based forensic tool may be feasible across decomposition environments. Importantly, this research demonstrates the critical ecological role insects play in the dispersal of specialized microbes that are involved in the breakdown and recycling of vertebrate remains.
22 May 2026
Read appraisal →Analytical methods : advancing methods and applications
Beyond identification: inferring physical activity from fingerprint lipids using machine learning.
Fingerprints are a widely recognized form of forensic evidence, valued for their ability to link individuals with specific locations. Traditional fingerprint analysis relies on optical imaging to identify a match in a fingerprint database; however, where no match is found, the evidential value of a latent print is limited. Here, we present the first study to integrate matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) with supervised machine learning to infer physical activity from fingerprint chemistry, expanding the utility of fingerprints beyond identification alone. Physical activity labels were derived from a validated questionnaire and converted into binary classes. Supervised machine learning algorithms were trained on the lipid features and evaluated against the survey-derived labels. The top-performing models were an ensemble algorithm based on multiple decision trees and a neural network, which classified physical activity with accuracies of 75 ± 8% and 73 ± 7%, respectively. These results demonstrate that fingerprint lipid chemistry encodes biologically meaningful information related to physical activity and establish a new approach for extracting lifestyle and behavioral indicators from trace evidence, with potential applications in forensic investigations and noninvasive fingerprint-based assessments in medicine.
4 May 2026
Read appraisal →BMJ paediatrics open
Does osteogenesis imperfecta predispose infants to metaphyseal fractures? A systematic review.
OBJECTIVE: The notion that osteogenesis imperfecta (OI) predisposes children to metaphyseal fractures is controversial. This review aimed to ascertain the prevalence of metaphyseal fractures in children with OI less than 2 years of age. MATERIALS AND METHODS: Medline, CINAHL and Web of Science were searched for studies reporting metaphyseal fractures in children with OI less than 2 years of age. Articles not retrievable, not published in English, or where 'metaphyseal' fracture was not defined/illustrated, were excluded. Article quality was independently assessed using the Critical Appraisal Skills Programme and Case Reports tools. RESULTS: Of 298 retrieved abstracts, 82 were duplicates: 216 articles were screened by title and abstract, excluding 186. Hand search of the remaining 30 articles identified two further articles. Only four articles were included: (1) 4/41 children with OI aged between 1 day and 3 years with 'metaphyseal' fractures; (2) a case series including a 16-month-old boy with three 'metaphyseal' fractures. Abuse concerns remained despite a type IV OI diagnosis (not genetically confirmed); (3) a 7-month-old girl with two 'corner' fractures suspected to have OI and to have been abused; and (4) a pair of siblings with 'metaphyseal' fractures and type XI OI (FKBP10 mutation). CONCLUSIONS: There is limited evidence that OI predisposes infants and young children to metaphyseal fractures. The two diagnoses are not mutually exclusive and until further large studies are conducted, physical abuse should always be considered in infants and young children with metaphyseal fractures, even in those with confirmed OI.
9 Apr 2026
Read appraisal →Journal of proteome research
Protein Biomarkers as Molecular Tools for Forensic Biofluid Identification
Forensic proteomics has rapidly established a significant role in forensics, particularly when DNA analysis is insufficient. Proteomics has been frequently recognized for helping reveal useful information about the origin, state and context of forensic samples. Proteins are considered crucial biomarkers in biological samples because of their ability to elucidate cellular functions and post-mortem alterations. Combining advanced mass spectrometry and bioinformatics has increased the importance of proteomics in forensic sciences. The current study examined the use of proteomic technologies, mass spectrometry, and sophisticated software across the key domains of forensic research. The procedure for Post-Mortem Interval (PMI) estimation and body fluid classification proved to be more precise because of the use of machine learning. Peptide biomarkers have helped identify various species by samples of blood, saliva, and semen, and recognize brain, muscles, and skin tissues. Despite significant advancements, the wide acceptance of forensic proteomics remains problematic due to intricate sample stability, high equipment costs, and strict legal standards. Recent advancements in analytical sensitivity, data interpretation tools, and collaborative efforts toward robust protocols position forensic proteomics as an indispensable component of the forensic toolkit. This review indicates the increasing relevance of proteomics in forensic applications, especially relating to PMI estimation, body fluid differentiation, and disease profiling. It promises to significantly enhance the depth of evidentiary interpretation and contribute to more precise and equitable outcomes in the criminal justice system. Proteomic biomarkers need further validation across a range of environments, and standardized protocols should be developed and tested to ensure that proteomics is suitable for forensic use in the courts.
6 Apr 2026
Read appraisal →