Search Results (6,218)

Historical paper collections held in archives and libraries represent an irreplaceable component of cultural heritage; yet, their systematic chemical assessment is often constrained by limited measurement datasets, heterogeneous material composition, and restricted access for sampling. This paper presents an extended version of the ARTEMISA framework (Intelligent Surface Assessment), a web-based decision-support application designed to assist conservators and archivists in assessing the chemical condition and degradation dynamics of paper-based materials under real-world constraints. The framework characterises the chemical state of paper through surface pH measurement—a non-destructive, practically accessible proxy for acidity-driven cellulose degradation. Rather than aiming at deterministic lifetime prediction, ARTEMISA emphasises chemically consistent segmentation of samples into homogeneous pH regimes, followed by regime-specific statistical modelling. For long-term trend exploration, the system integrates three regression approaches: log-transformed linear, polynomial, and exponential models. For short-term time-series analysis within restricted chemical regimes, an ARIMA model is applied. Model performance was evaluated using standard statistical metrics: Mean squared error (MSE), Root mean squared error (RMSE), Mean absolute error (MAE) and the coefficient of determination (R²). The framework was applied to a case study comprising twelve book titles from the Public Library in Trenčín (Slovak Republic), published between 1959 and 2016. Results indicate that log-transformed linear and exponential regression models provide physically consistent representations of long-term pH trends, while ARIMA modelling captures short-term dynamics directly actionable for conservation planning. Exponential regression achieved R² = 0.783, linear 0.826, polynomial 0.911; however, the polynomial model produced physically implausible extrapolations and is retained for comparative purposes only. The Arrhenius principle is incorporated as an interpretative component, contextualising pH-derived degradation trends within temperature-dependent reaction kinetics. ARTEMISA offers an accessible, open-architecture tool for the prioritisation of conversation interventions under conditions of data scarity. This is applicable to broader material degradation studies with limited measurements. Full article

Magnetic resonance imaging (MRI)-based prediction of multiple sclerosis (MS) progression depends on how imaging data are prepared, represented, modelled, and evaluated. This systematic review synthesized artificial intelligence (AI) methods across the MRI-to-prediction pipeline. PubMed, Scopus, Web of Science, and Google Scholar were searched for studies published from 2010 to March 2026. Eligible studies included patients with MS or clinically isolated syndrome, used brain MRI directly or as the source of predictors, evaluated future progression-related outcomes, and reported quantitative predictive performance. Thirty-seven studies were included and synthesized narratively because of methodological and outcome heterogeneity. Prediction targets included CIS-to-MS/CDMS conversion, future MRI disease activity, PIRA/PIRMA, disability worsening or confirmed progression, future EDSS or disability status, and RRMS-to-SPMS conversion. Lesion morphology and location, radiomics, tissue atrophy, longitudinal imaging changes, and multimodal variables provided useful prognostic information. Classical machine learning (ML) remained effective for structured biomarkers, while deep learning (DL) enabled direct image-based modelling; neither approach was uniformly superior. PROBAST + AI assessment rated model Development as High concern in 35 of 37 studies and Unclear in two, Evaluation as High risk of bias in all 37 studies, and Applicability as Low concern in 32, Unclear in two, and High in three. Overall, MRI-based AI shows promising prognostic potential, while stronger data separation, calibration, transparent reporting, and independent validation are needed to support reproducible clinical translation. Full article

Free-surface sloshing in partially filled tanks and open reservoirs can produce spatially complex surface deformation under horizontal excitation. This study presents an image-based experimental and computational workflow for reconstructing measured free-surface profiles in a laboratory-scale rectangular reservoir mounted on a Haas VF-3/VF-3SS-family CNC machining center. The tank underwent a programmed horizontal back-and-forth motion. The retained CNC controller record shows commanded x positions spanning approximately $-19.976$ mm to $+19.851$ mm, corresponding to a programmed peak-to-peak stroke of approximately 39.8 mm. The realized forcing frequency was not independently measured; therefore, no experimental near-resonance condition is claimed. Videos were recorded approximately normal to the transparent tank wall, converted to still frames, calibrated directly in physical length units, and manually digitized with WebPlotDigitizer. A truncated spatial Fourier series was fitted independently to each measured time slice. The accompanying Python applications dynamically evaluate the experimentally fitted representation: the GUI tools animate reconstructed profiles over the recorded time sequence, while the final FourierN3 implementation builds time-dependent coefficient models, linearly interpolates coefficient histories, evaluates $y(x,t)$, and generates a three-dimensional space–time reconstructed surface. These outputs are described as dynamic Fourier reconstructions because they remain conditioned on measured coefficient histories and do not constitute independent CFD or potential-flow predictions. The Fourier harmonics are mathematical reconstruction functions and are not identified one-to-one with physical sloshing eigenmodes. For a 0.45 m tank length, linear theory gives first-mode reference frequencies of 0.687 Hz at a 4 cm water depth and 0.763 Hz at a 5 cm water depth. The workflow is demonstrated on three selected datasets that provided sufficiently clear free-surface records for consistent frame-by-frame digitization. Using a common three-harmonic benchmark, mean RMSE values were 0.8815 cm for Case A (4 cm), 1.419 cm for Case B (5 cm), and 0.5022 cm for Case C (4 cm), corresponding to mean NRMSE values of 22.0%, 28.4%, and 12.6%, respectively. For Case C, adding a fourth harmonic reduced the mean RMSE to 0.4427 cm (11.1% of water depth), an 11.8% reduction relative to the three-harmonic fit. All RMSE and NRMSE values are reconstruction errors relative to the manually digitized coordinates, not estimates of absolute experimental measurement accuracy. The results therefore demonstrate the reconstruction workflow for the three selected records rather than establish a general water-depth effect. Full article

The rapid evolution of Large Language Models (LLMs) has opened new frontiers in cybersecurity automation, enabling intelligent agents capable of multi-step reasoning, tool invocation, and autonomous decision-making across complex security tasks. While individual applications have emerged across threat intelligence, vulnerability assessment, penetration testing, and security operations center (SOC) automation, a systematic understanding of the LLM-based agent paradigm in cybersecurity—encompassing both single-agent and multi-agent architectures—remains lacking. This paper presents a systematic literature review following PRISMA guidelines, identifying records through 59 structured web-search queries whose results resolve predominantly to arXiv, Semantic Scholar, the ACM Digital Library, IEEE Xplore, USENIX, MDPI, SpringerLink, and Elsevier ScienceDirect, supplemented by citation chaining, for works published between January 2022 and April 2026; the full query record is published with the paper. We applied structured inclusion and exclusion criteria and classified 59 primary studies along five dimensions: security function, agent architecture pattern, knowledge augmentation strategy, human-in-the-loop posture, and evaluation rigor. Our analysis reveals that penetration testing and threat intelligence are the most extensively studied domains, while incident response and compliance verification remain critically underrepresented. Penetration testing alone accounts for over half the corpus (50.8%). Single-agent tool-calling remains the most prevalent architecture (30.5% of studies), whereas centralized multi-agent orchestration—present in 18.6%—yields the strongest reported performance gains, up to 4.3× on zero-day exploitation; prevalence and performance therefore point in opposite directions. No included study achieves production-grade (E4) evaluation: the entire field currently rests on controlled laboratory assessments. An independent search of six bibliographic databases recovers 86.3% of the studies the primary search had surfaced (79.7% of the full corpus) while indicating a total eligible literature of roughly 400 studies, so the corpus is reported as a documented subset rather than an exhaustive census. We propose a unifying taxonomy, identify cross-cutting challenges including hallucination, prompt injection, and benchmark fragmentation, and outline open research directions with particular emphasis on multi-agent orchestration design. Financial sector applicability under DORA and the EU AI Act is treated as a documented evidence gap rather than a synthesis: the corpus’s only compliance and risk assessment study is also its only banking-specific system. Full article

Increasing soil contamination by potentially toxic elements (PTEs) compromises environmental quality and ecosystem health, reinforcing the need for effective remediation strategies. Phytoremediation uses plants to remove, immobilize, or neutralize contaminants and represents a sustainable approach for restoring contaminated soils. Although nickel (Ni) is an essential plant micronutrient, elevated concentrations can cause phytotoxicity. This integrative review aimed to synthesize and critically analyze the scientific literature on the phytoremediation of Ni-contaminated soils published between 2000 and 2026, focusing on the plant species evaluated, their phytoremediation potential, and the main research trends and knowledge gaps. The Web of Science Core Collection search identified 230 records. Original research articles addressing the phytoremediation of Ni-contaminated soils and providing sufficient information to characterize phytoremediation potential were considered eligible, resulting in 91 included studies. Plant responses varied according to species and Ni concentration. Among 449 plant records, defined as the occurrence of a plant taxon within a study, more than 86% corresponded to herbaceous plants; Poaceae accounted for 27.2% of the records, followed by Brassicaceae (15.1%). Brassica juncea was consistently identified as a promising species for Ni phytoremediation, showing Ni tolerance, phytoextraction, and hyperaccumulation across different studies. Alyssum murale stood out among the reported hyperaccumulators for its high Ni accumulation and phytoextraction potential. Overall, the reviewed evidence supports phytoremediation as a sustainable strategy for Ni-contaminated soils. However, important knowledge gaps remain, particularly the scarcity of long-term field studies, methodological heterogeneity, limited understanding of plant–soil–microbiota interactions, underrepresentation of major Ni-producing regions, and insufficient evidence on technical and economic feasibility. Future research should address these gaps through standardized approaches and integrated field assessments under contrasting edaphoclimatic conditions to advance the effective and sustainable application of Ni phytoremediation. Full article

Background/Objectives: Molecular classification has reshaped risk assessment in endometrial cancer (EC), but the strength and clinical applicability of the evidence vary across molecular subgroups, assay platforms, patient populations, and treatment settings. This systematic review critically evaluated the prognostic and predictive value of TCGA-derived classifiers, additional biomarkers, multiple-classifier tumors, and molecularly informed therapeutic strategies. Methods: A protocol was developed a priori but was not prospectively registered. PubMed/MEDLINE, Embase, Scopus, Web of Science, and the Cochrane Library were searched for primary studies published from January 2013 through June 2026. Clinical guidelines were used only to provide clinical context and were not included in the formal evidence set. Because of substantial methodological heterogeneity, findings were synthesized narratively by study design and clinical question. Risk of bias was assessed using QUADAS-2, the Newcastle–Ottawa Scale, and RoB 2. Results: Seventy-four studies involving approximately 42,000 patients were included. POLE-mutated tumors showed the most favorable outcomes in most cohorts, whereas p53-abnormal tumors, defined by abnormal immunohistochemical patterns, generally had the poorest prognosis. Mismatch repair deficiency identified by immunohistochemistry and microsatellite instability-high status identified by PCR- or NGS-based testing were highly related but analytically distinct biomarkers and were not treated as exact synonyms. The NSMP group remained heterogeneous; integrated assessment of estrogen/progesterone receptor expression, CTNNB1, L1CAM, ARID1A, and chromosome 1q alterations may improve risk discrimination, although most of these markers remain investigational. In multiple-classifier tumors, available data support a practical hierarchy in which a pathogenic POLE mutation generally takes precedence, followed by MMR deficiency and then p53 abnormality. Across studies, molecular classification usually added prognostic information to clinicopathologic assessment, but the magnitude of benefit varied, and no overall certainty rating was assigned. Conclusions: Molecular classification is incorporated into contemporary guidelines and can support prognostic assessment and treatment selection when interpreted together with stage, histology, and other clinicopathologic factors. The strongest current clinical utility relates to POLE, MMR/MSI, p53 immunohistochemistry, and HER2 in selected settings; additional NSMP and multi-omics biomarkers require prospective validation, assay standardization, and demonstration of clinical utility before routine adoption. Full article

Benign prostatic hyperplasia (BPH) affects more than 95% of intact male dogs, yet most remain asymptomatic until prostatic enlargement causes clinical signs. Standardised assessment of BPH severity is therefore needed to improve clinical decision-making and enable comparable research outcomes. This study aimed to investigate current diagnostic and therapeutic approaches to canine BPH and to develop and assess a web-based application implementing the Zambelli Symptom Index (ZSI), an established symptom-based scoring system for BPH severity. A voluntary questionnaire was completed by veterinary practitioners to assess current clinical practice, and agreement between the web-based and reference paper-based ZSI classifications was evaluated. The survey revealed considerable variability in the recognition of clinical signs, assessment of BPH severity, and selection of treatment strategies, despite broad agreement on the diagnostic value of ultrasonography and the need for further investigation of prostatic enlargement. The web-based ZSI showed very high agreement with the reference classification. These findings demonstrate that the ZSI can be implemented digitally without materially affecting severity classification. The web-based tool may improve the standardisation and reproducibility of BPH assessment in clinical practice and research. Further studies are warranted to evaluate interobserver variability, clinical validity, and responsiveness to treatment. Full article

This scoping review maps the available literature aimed at identifying the most appropriate technique for suturing pelvic flexure enterotomies in horses. A comprehensive search of Web of Science, Google Scholar, and the National Center for Biotechnology Information was conducted for studies published between January 1996 and February 2026, using terms related to pelvic flexure enterotomy and equine patients. From 22,146 records, only five ex vivo studies on healthy equine colon specimens met the inclusion criteria. The studies compared variables such as suturing time, suture pattern, material, bursting pressure, luminal diameter reduction, and cost. Double-layer hand-sewn closure generally achieved the highest bursting pressures, although it required longer operative time, while TA90 stapling was the most expensive option. Because ex vivo models cannot reproduce tissue healing or host–construct interactions, the clinical applicability of the available evidence remains limited. Nonetheless, a full-thickness simple continuous pattern oversewn with a continuous Cushing pattern appears to offer the most dependable mechanical security until stronger in vivo evidence becomes available. Full article

Traditional fermented cereals are affordable, embedded foods that may provide viable microorganisms, fermentable substrates, and bioactive metabolites that can influence the gut microbiota and human physiology. However, evidence linking indigenous cereal fermentations in Sub-Saharan Africa (SSA) to microbiome-mediated health outcomes and precision nutrition applications remains limited. This scoping review mapped the types of traditional fermented cereal foods studied in SSA; their microbial communities and probiotic attributes; the reported effects on gut microbiota composition and function; the documented health outcomes; and the geographic gaps. The review followed the Arksey and O’Malley framework, subsequent methodological refinements, and the PRISMA-ScR reporting guideline. Searches were conducted in Scopus, PubMed, Web of Science, ScienceDirect, and Google Scholar from 2000 to mid-2026, complemented by citation searching. Eligible human, animals, and in vitro studies examined SSA-fermented cereal products and reported microbial, probiotic, gut microbiota, metabolite, or health-related outcomes. After deduplication and screening, 53 studies were included. The evidence was concentrated in SSA, particularly in cereal-based foods. Lactic acid bacteria and yeasts dominated reported fermentation communities, but strain-level validation, dose definition, and controlled human studies were uncommon. Evidence for direct gut–microbiota modulation and clinically important health effects was substantially weaker than evidence for food-level microbial diversity, in vitro probiotic properties, and improved nutrient bioaccessibility. Overall, traditional fermented cereals in sub-Saharan Africa demonstrate promising microbial, nutritional, and functional properties, but current evidence remains heterogeneous and insufficient to establish consistent effects on gut microbiota and human health. Further well-designed human studies integrating dietary, microbiome, and metabolic outcomes are needed to clarify their potential role in precision nutrition. Full article

Safety culture has become a central construct in occupational safety management due to its recognized influence on organizational performance and accident prevention. However, the absence of a universally applicable assessment methodology has contributed to substantial methodological fragmentation. Therefore, this scoping review aimed to identify, classify, and critically analyze methodological approaches for safety culture assessment. Following PRISMA-ScR guidelines, studies published between January 2016 and October 2025 were retrieved from Scopus, Web of Science, and Dimensions. Study selection and methodological quality appraisal were conducted using predefined eligibility criteria and the Mixed Methods Appraisal Tool (MMAT, version 2018). Through a descriptive synthesis, the 25 included studies revealed a heterogeneous landscape encompassing psychometric surveys, multi-criteria and maturity-based models, and triangulated approaches. The findings indicate that each methodological family captures different dimensions of safety culture. Psychometric approaches provide scalability and comparability but remain dependent on self-reported perceptions, whereas multi-criteria models support structured organizational diagnosis. Triangulated approaches integrate multiple sources of evidence and facilitate the examination of discrepancies between formal organizational arrangements and operational practices. Overall, the review highlights the diversity of available assessment approaches and suggests that the evaluation’s objectives, context, and scope should guide the selection of the methodology. Full article

In the field of architectural heritage, industrial built heritage plays an important role. Traditional research methods such as literature studies and field surveys have gradually evolved into digital-twin and machine-learning-based approaches. However, the study of industrial heritage remains a challenge. This study employs the PRISMA framework to conduct a systematic review of the existing literature, aiming to elucidate the impact of GIS on industrial heritage sites and its potential limitations. This study conducts a systematic review of GIS-based research on industrial heritage, focusing on studies published between 2000 and 2025. Among the 2166 records initially retrieved from Scopus and the 1771 records initially retrieved from Web of Science, 77 papers were ultimately retained after screening and deduplication. A total of 32 studies were selected based on thematic screening criteria, all of which focused on analysing the spatial characteristics of industrial heritage sites using GIS technology. Among these, 12 core studies were further selected for in-depth qualitative and methodological analysis. These documents form the foundation for exploring research hotspots, methodological approaches, and technological evolution. This review contributes a five-dimensional analytical framework that systematically classifies GIS-based industrial heritage research and identifies eight structural research gaps. The findings indicate that GIS applications remain predominantly focused on spatial mapping and distribution analysis, while their integration with ecological assessment, social and participatory approaches, archaeology, and multi-source technologies remains limited. This review provides a roadmap for integrating advanced spatial analysis tools into future industrial heritage research and planning. Full article

Circadian–behavioral misalignment, defined as a mismatch between internal circadian timing and daily sleep activity behavior, is associated with sleep quality, daytime function, cognitive performance, and long-term health risks. However, practical approaches for integrating circadian-phase-related physiological markers with behavioral and environmental sensing under free-living conditions remain limited. This study presents a multimodal wearable and lifelog monitoring workflow that combines a heat-flux-based wearable temperature-estimation sensor, a wrist-worn actigraph, melanopic equivalent daylight illuminance (EDI) measurement, and web-based lifelog entries. Ten healthy young adults were monitored for approximately one week in daily life. The midpoint between the descending and ascending mid-level crossings of the nocturnal temperature rhythm was extracted as a circadian-phase-related temperature rhythm marker. In a practical comparison with gastrointestinal temperature measured using an ingestible capsule, the wearable-derived rhythm marker showed a mean absolute difference of 27.1 ± 16.6 min. The temporal difference between the two markers was defined as the phase angle, and coupling stability was evaluated using phase angle variability and the correlation between the markers. After excluding one participant who traveled overseas, clustering and principal component analyses were used to visualize exploratory participant-level differences in circadian–behavioral coupling. Integrated analysis with melanopic EDI, activity, and lifelog-derived features suggested that light exposure, activity timing, and behavioral timing may provide contextual information for interpreting coupling patterns in this small cohort. These findings support the feasibility of a sensing system application that integrates physiological, behavioral, environmental, and lifelog data for the exploratory assessment of circadian–behavioral coupling under free-living conditions, rather than the development of new sensor hardware. Full article

Gravity energy storage (GES) has emerged as a promising option for medium- to long-duration storage in renewable-dominated power systems. However, the literature remains fragmented across architectures, modeling approaches, and application contexts, hindering systematic comparison and consistent assessment of technological maturity. This paper presents a systematic review of GES conducted in accordance with the PRISMA 2020 guidelines; 107 studies retrieved from 4 databases (Scopus, Web of Science, IEEE Xplore, and ScienceDirect) were analyzed by integrating fundamental principles, system architectures, modeling and control strategies, and techno-economic performance. Unlike previous reviews, this work explicitly links physical principles, design variables, and system-level performance, enabling consistent cross-architecture comparison and clearer positioning of GES within the broader storage landscape. Liquid-based systems and solid gravity energy storage (SGES) architectures, including tower-, shaft-, rail-, mountain-, piston-, and modular-based systems, are compared in terms of scalability, efficiency, response, and infrastructure requirements. Attention is given to the gap between theoretical and real-world performance, highlighting low intrinsic energy density, strong dependence on large-scale infrastructure, the absence of standardized techno-economic assessment, and limited long-term operational data. The review positions GES as a complementary technology rather than a direct competitor to electrochemical or hydro-based storage, and critically examines emerging trends, including hybridization, digitalization, advanced control, and material optimization. Finally, it identifies key technological and deployment gaps and proposes a research roadmap centered on scalability, standardization, and system integration. Full article

Clinical adoption of large language models (LLMs) in public healthcare faces a structural impasse: the capital expenditure of centralised high-performance computing on one side and the privacy risk of routing patient data through third-party cloud interfaces on the other. Italian local health authorities (Aziende Sanitarie Locali, ASL) operate large fleets of clinical workstations that remain idle outside peak administrative hours. We present OmniMed Federated, a browser-native architecture using the WebGPU application programming interface (API) and the WebLLM framework to distribute LLM inference tasks across these existing workstations. The system federates task allocation rather than model training or partitioned inference: each query executes in full on one node, selected under a data residency constraint. A five-tier escalation model, coordinated by a metadata-only PHP back end, ranks tiers by data exposure rather than capability, with commercial cloud fallback disabled by default. In a pilot three-node testbed (50 queries), federated throughput reached 19.5 versus 8.2 tokens/second standalone, peak per-node memory fell 62%, and node discovery took 140 ms; query content remained within the institutional perimeter throughout. These figures establish infrastructural feasibility at pilot scale. Clinical output quality, security hardening, and scalability remain unevaluated. Full article

Reliable wearable motion tracking depends not only on nominal sensor precision but on whether calibration remains valid after re-donning, anatomical misalignment, magnetic disturbance, temperature change, and prolonged operation. This structured critical review treats calibration as a system-level engineering decision spanning intrinsic inertial measurement unit (IMU) correction, sensor-to-segment alignment, flexible-sensor calibration, drift management, and complementary or external references. A reproducible Scopus core search conducted in May 2026 identified 807 records. Three co-authors screened titles, abstracts, and keywords using explicit eligibility criteria, with full-text inspection when required; the final 110-source reference corpus combines eligible core records with supplementary and contextual sources identified through coverage checks in IEEE Xplore, Web of Science, ScienceDirect, and the MDPI database. Unlike prior reviews that mainly treat individual calibration layers or inertial-error mechanisms, this review jointly compares sensor-level, anatomical, multimodal, adaptive, and external-reference routes within a common deployment-oriented framework. Evidence is coded by failure mechanism, calibration route, sensing modality, application context, validation context, and deployment constraint. No universally superior route is supported: static and functional alignment differ in burden and anatomical sensitivity; magnetometer use depends on field reliability; vision and ultra-wideband (UWB) references can restore absolute observability but add infrastructure dependence; and direct IMU–flex calibration remains comparatively sparse. The resulting decision framework prioritizes reproducibility, robustness, calibration stability, and fitness for real-world deployment over isolated best-case accuracy. Full article