Search Results (79)

The rapid adoption of large language models (LLMs) in financial services has generated a growing literature on “responsible AI” in domains such as investment analysis, credit assessment, risk management, compliance, and financial advisory systems. Unlike earlier AI systems, LLMs introduce responsibility challenges that differ in important ways from those addressed by earlier responsible AI frameworks, including hallucinations, prompt injection and manipulation, generative opacity, instruction-following failures, context sensitivity, output inconsistency, and emergent capabilities that arise only at larger model scales. While responsible AI concerns extend across AI systems more broadly, this review focuses specifically on LLMs, which since 2022 have become a major technological force reshaping financial services and have generated a distinctive body of responsibility discourse that existing frameworks are only beginning to address. However, the meaning of responsibility in this literature remains heterogeneous and inconsistently operationalized across studies. This paper provides a structured review of research on responsible LLMs in finance using a broad-to-narrow screening process across Web of Science and Scopus, reported with reference to the applicable PRISMA 2020 items. We combine a descriptive bibliometric overview with qualitative content analysis to examine the corpus. The descriptive overview covers publication trends, publication venues, disciplinary orientations, geographic distributions, and institutional affiliations, while the qualitative analysis codes the corpus on two dimensions: responsibility depth—whether responsibility is constitutive of the paper’s contribution, operative within its design, or peripheral to its framing—and responsibility mode—whether that engagement is normative, technical, evaluative, or mixed. The findings reveal a field concentrated in the Operative–Technical mode. Constitutive contributions cluster in Normative and Mixed modes, while a purely Evaluative orientation remains comparatively uncommon. Addressing these limitations requires moving beyond principle-level discourse toward institution-specific accountability mechanisms, empirical evaluation across clearly documented research and operational settings, and systematic alignment with relevant financial regulatory requirements. These efforts must address the distinctive challenges posed by LLMs, rather than only concerns inherited from the broader responsible AI literature. Full article

The rapid deployment of Internet of Things (IoT)-enabled video surveillance systems has expanded the capabilities of real-time monitoring in smart cities, healthcare facilities, industrial environments and critical infrastructure. However, integrating resource-constrained cameras, heterogeneous communication protocols, edge/cloud analytics, and sensitive video data creates a complex cybersecurity landscape. This systematic review synthesizes recent evidence on security challenges and mitigation strategies in IoT-enabled video surveillance systems. Following the PRISMA 2020 guidelines, four bibliographic databases (Scopus, IEEE Xplore, Web of Science, and Google Scholar) were searched for peer-reviewed journal articles and conference papers published between January 2021 and July 2025. After duplicate removal, title/abstract screening, full-text assessment, and quality appraisal, 21 studies were included for qualitative synthesis. The findings show that vulnerabilities occur across three interdependent architectural layers: device/perception, network/communication, and application/cloud. The frequently reported weaknesses were default credentials, insecure firmware, unencrypted video streams, weak protocol configuration, metadata leakage, and inadequate cloud access control. Existing mitigation strategies, including multi-factor authentication, role-based access control, TLS/DTLS, lightweight encryption, intrusion detection systems, and secure boot, provide partial protection but remain constrained by latency, computational overhead, energy consumption, scalability, cost and legacy device compatibility. This review further identifies a persistent research–practice gap: only a small subset of studies provides evidence of real-world deployments, while most solutions remain evaluated in simulations, testbeds, or conceptual frameworks. This review contributes a domain-specific taxonomy of IoT video surveillance security, a comparative evaluation of mitigation strategies using technical, operational, and economic criteria, and deployment-oriented recommendations for smart city, industrial, healthcare, residential, and critical infrastructure settings. The study highlights the need for cross-layer security architectures, lightweight and post-quantum-ready cryptography, privacy preservation, edge AI, federated learning, zero-trust access control, and standardized security baselines. Full article

Background: Autopsy pathology remains the reference standard for determining the cause of death, reconstructing disease and injury mechanisms, ensuring diagnostic quality, and supporting medical education and forensic investigations. However, declining autopsy rates, workforce shortages, biosafety concerns, increasing diagnostic complexity, and the rapid evolution of digital technologies have stimulated the development of complementary investigative approaches. This review critically examines whether artificial intelligence (AI) and emerging technologies are driving a genuine paradigm shift toward digitally integrated death investigation. Methods: A structured narrative review informed by a systematic literature search was conducted in PubMed/MEDLINE, Embase, Scopus, and Web of Science, covering publications from January 2000 through June 2026. Evidence addressing postmortem imaging, virtopsy, digital pathology, computational pathology, molecular autopsy, robotics, artificial intelligence, machine learning, and emerging omics technologies was critically appraised. Owing to the methodological heterogeneity of the available literature, findings were synthesized qualitatively according to technological maturity, forensic applicability, validation status, and implementation readiness. Results: The reviewed evidence demonstrates substantial progress in postmortem computed tomography, postmortem CT angiography, postmortem magnetic resonance imaging, whole-slide imaging, molecular autopsy, robotic-assisted postmortem procedures, three-dimensional reconstruction, and AI-assisted forensic analysis. These technologies enhance trauma evaluation, vascular imaging, ballistic reconstruction, digital documentation, remote consultation, diagnostic reproducibility, and multimodal integration of forensic evidence. Nevertheless, the level of evidence varies considerably across technological domains. Postmortem imaging represents the most mature and extensively validated technology, whereas most AI applications remain supported predominantly by retrospective proof-of-concept studies with limited multicenter external validation. Current systematic evidence further indicates that AI should presently be regarded as an assistive technology that augments expert forensic interpretation rather than replacing conventional autopsy or autonomous medicolegal decision-making. Conclusions: Contemporary autopsy pathology is evolving toward a hybrid model of digitally integrated death investigation in which conventional autopsy, imaging, digital pathology, molecular diagnostics, robotics, and AI function as complementary components of a unified forensic workflow. Current evidence supports a conceptual paradigm shift characterized by transformation of evidence acquisition, preservation, interpretation, and integration, while reaffirming that conventional autopsy remains the indispensable biological reference standard for the development, validation, and medicolegal interpretation of all emerging technologies. Future implementation should prioritize multicenter validation, standardized forensic datasets, explainable AI, digital chain-of-custody procedures, and robust regulatory governance to ensure safe and scientifically reliable integration into forensic practice. Full article

Firefighters operate in high-risk, rapidly evolving environments where exposure to extreme heat, toxic gases, and physiological stress significantly increases the likelihood of injury and fatality. This study systematically maps the emerging research landscape of real-time artificial intelligence (AI)-driven digital twins for environmental and physiological risk prediction in firefighting contexts. A combined bibliometric and qualitative content analysis was conducted using peer-reviewed literature retrieved from the Web of Science database (2010–2025). Bibliometric techniques were used to identify publication trends and thematic clusters, while content analysis examined the integration of sensing technologies, AI models, and digital twin architectures. The results reveal four dominant technological domains shaping the field: AI-enabled fire risk modeling, sensor data acquisition systems, IoT-based digital infrastructures, and predictive analytics for disaster simulation. Sensing technologies such as temperature, gas, particulate matter, thermal imaging, heart rate, and blood oxygen monitoring form the foundational data layer, while machine learning and deep learning models enable real-time hazard prediction and situational awareness. Digital twin architectures serve as the integration layer, fusing multi-source data and supporting simulation-based decision-making. Despite rapid advancements, key gaps persist, including limited integration of environmental and physiological data, insufficient predictive capabilities, a lack of standardized architectures, and minimal development of human-centered decision-support systems. This study provides a structured synthesis of current technologies and identifies future research directions toward integrated, explainable, and real-time digital twin systems to enhance firefighter safety and operational resilience. Full article

The rapid expansion of wind and solar generation has significantly increased the share of variable renewable energy in power systems worldwide, introducing new operational challenges. Among these, the simultaneous growth of renewable energy curtailment and persistent blackout risk reveals structural limitations in energy planning and system flexibility. This study conducts a Systematic Literature Review (SLR) following the PRISMA protocol to examine how the scientific literature has addressed the relationship between curtailment, energy security, and optimization strategies in high-renewable power systems. A total of 53 Q1-indexed articles published between 2021 and 2025 were analyzed using bibliometric and qualitative content analysis techniques. The results indicate that curtailment should not be interpreted solely as an operational inefficiency but rather as a potential flexibility asset when integrated with energy storage systems, power-to-X technologies, demand-side management, and stochastic optimization frameworks. The findings also highlight a shift from deterministic planning approaches toward robust and distributionally aware models capable of managing renewable uncertainty. Despite significant advances, geographic imbalances in case studies and limited integration between regulatory mechanisms and technical optimization remain key research gaps. This review contributes by synthesizing mitigation strategies into a structured flexibility framework and by outlining research directions for enhancing reliability in renewable-dominated systems. Full article

The construction sector is undergoing a rapid transition toward more resilient, sustainable, and digitally connected systems, creating increasing demand for materials capable of providing functions beyond conventional structural performance. In this context, smart materials have emerged as promising solutions due to their ability to respond to mechanical, thermal, chemical, or electromagnetic stimuli through adaptive behaviors such as self-healing, structural sensing, energy regulation, vibration control, and reversible deformation. Despite growing scientific interest, available knowledge remains fragmented across specific material families and isolated application domains. Therefore, this study presents a PRISMA-based systematic review of smart materials in construction using peer-reviewed journal literature indexed in Scopus during the 2021–2026 period. The review examines the principal smart material families currently applied in construction, including self-healing concretes, self-sensing cementitious systems, Shape Memory Alloys (SMA), piezoelectric materials, phase change materials, adaptive coatings, conductive nanocomposites, and multifunctional geopolymers. Their engineering functions, structural and architectural applications, reported performance characteristics, sustainability contributions, digital integration potential, and implementation barriers are comparatively discussed and qualitatively synthesized based on the reviewed literature. The findings indicate that smart materials can improve durability, structural health monitoring, seismic resilience, thermal efficiency, lifecycle performance, and carbon reduction when properly integrated into buildings and infrastructure. However, large-scale adoption remains constrained by high initial costs, manufacturing scalability, regulatory uncertainty, long-term durability validation, and limited market confidence. The review further shows that the greatest future potential lies in combining material intelligence with IoT platforms, artificial intelligence, BIM environments, and digital twins. Overall, smart materials are positioned as strategic enablers of next-generation low-carbon, adaptive, and intelligent construction systems. Full article

Background: Diabetic neuropathy (DN) is a common complication of diabetes mellitus that remains frequently undetected by conventional diagnostic methods. Sudomotor dysfunction, reflecting small-fiber impairment, has emerged as a potential early marker. SUDOSCAN, a rapid and non-invasive device measuring electrochemical skin conductance (ESC), has been proposed as a screening tool for early DN. The objective of this study was to systematically evaluate the diagnostic performance and clinical utility of SUDOSCAN in the early detection of DN. Methods: A systematic review was conducted in accordance with the PRISMA 2020 guidelines. Studies assessing SUDOSCAN-derived ESC in adults with diabetes were included. Data on diagnostic accuracy, correlations with established neuropathy measures, and clinical applicability were extracted. Where feasible, pooled sensitivity and specificity were estimated using a random-effects model. Results: Fifteen studies (n = 7343 participants) were included in the qualitative synthesis, with five of them contributing to the quantitative analysis. Reduced ESC values were consistently associated with DN, including early and asymptomatic cases. Pooled sensitivity and specificity for detecting DN were 0.81 (95% CI 0.73–0.87) and 0.73 (95% CI 0.57–0.85), respectively. ESC values correlated with neuropathy severity scores and autonomic dysfunction measures. However, substantial heterogeneity was observed due to variability in diagnostic criteria, ESC thresholds, and study populations. Conclusions: SUDOSCAN is a feasible, rapid, and non-invasive tool for detecting DN, particularly in the early-stage or small-fiber disease. It shows promise as a screening and adjunctive diagnostic modality, especially when combined with established clinical tools. Nevertheless, the lack of standardized thresholds limits its standalone use. Full article

Background and objectives: Sentinel lymph node biopsy (SLNB) has become the standard approach for cervical staging in patients with early-stage oral cavity squamous cell carcinoma (OCSCC). Although it demonstrates diagnostic accuracy exceeding 90% in referral centres, the occurrence of false-negative (FN) results undermines oncological safety and adversely impacts patient prognosis. This scoping review aims to synthesise and evaluate the scientific evidence regarding the risk factors and technical errors that underpin FN incidence. Methods: A systematic search was conducted across MEDLINE (PubMed), Scopus and Web of Science for studies published between 2000 and 2025, adhering to PRISMA-ScR guidelines. Primary clinical and observational studies specifically addressing the variables and aetiology of diagnostic failure in SLNB for early-stage OCSCC (cT1-T2 N0) were included. Results: Twenty-seven studies were included in the final qualitative synthesis. Four critical domains characterising FN risk were elucidated, facilitating risk-stratified selection and surveillance strategies: (1) Clinical and tumour factors: Depth of invasion >4–5 mm, T3–T4 tumour size, and prior cervical anatomical disruption. (2) Surgical factors: Insufficient sentinel lymph node (SLN) harvest (≤2 SLNs), lack of exhaustive lymphatic basin exploration, the initial learning curve, and the exclusive use of non-isotopic tracers. (3) Anatomical factors: Floor of the mouth tumours and the radioactive shine-through effect. (4) Histopathological protocol: Suboptimal ultrastaging, insufficient frozen section biopsy, and limitations in rapid molecular techniques. Overall, studies using standardised protocols report a false-negative rate between 5% and 15%. Conclusions: FN events in SLNB are multifactorial and predictable phenomena, arising from cumulative vulnerabilities along the diagnostic continuum. The technique achieves an optimal diagnostic yield in cT1–T2 N0 cases without prior cervical treatment when applied under optimal conditions. However, methodological heterogeneity in the literature limits the interpretation of the results and partially constrains their clinical application. Full article

Monitoring inland surface water quality is essential for water security, ecosystem conservation, public health, and sustainable water resource management. Although in situ measurements remain indispensable, they are often limited by high costs, restricted spatial coverage, low temporal frequency, and discontinuous monitoring networks. This study presents a systematic literature review, guided by the PRISMA 2020 framework, of empirical studies published between 2021 and 2025 on the integration of artificial intelligence (AI) and remote sensing (RS) for inland surface water quality monitoring. Searches were conducted in the Web of Science database, resulting in a final corpus of 367 peer-reviewed articles. Preliminary bibliometric characterization and qualitative content analysis were performed to identify sensors, platforms, AI paradigms, algorithms, estimated parameters, validation strategies, limitations, challenges, trends, and research gaps. The results show rapid growth in the field, with Sentinel-2 and Landsat-8 as the most recurrent sensors and multispectral data as the dominant spectral source. Machine learning approaches, especially Random Forest, Artificial Neural Networks, XGBoost, and Support Vector Machine, predominated, while deep learning, multi-source integration, hybrid models, and Explainable AI emerged as relevant trends. AI–RS integration shows strong potential to complement conventional monitoring, but persistent challenges remain regarding in situ data dependence, limited external and temporal validation, model transferability, generalization, uncertainty reporting, validation robustness, and interpretability. Full article

Rapid urban expansion is reshaping large metropolitan regions into polycentric systems in which multiple centers interact through transport, infrastructure, land-use, economic and ecological networks. Urban heat island (UHI) research has traditionally relied on single-city or core–periphery models; these remain useful for explaining heat contrasts within individual cities, but are insufficient for explaining how thermal loads form, propagate and accumulate across interconnected metropolitan regions. This study combines bibliometric analysis and a PRISMA-guided systematic review to synthesize research on UHI processes in polycentric cities, mega-urban regions and metropolitan systems. The bibliometric corpus comprises 468 Scopus-indexed records published in 2020–2025, while 35 full-text studies were retained for qualitative synthesis. The results show strong publication growth from 54 records in 2020 to 124 in 2025, an annual growth rate of 18.09%, and an interdisciplinary evidence base led by environmental science, social science, Earth-system science and engineering. Three spatial patterns recur across the core studies: multi-core hotspots, corridor-based heat propagation and peripheral thermal expansion. The review contributes a network-based interpretation of UHI as a nested metropolitan process in which node morphology, functional hierarchy, transport connectivity, blue–green infrastructure (BGI) and governance coordination jointly shape heat intensity, footprint and exposure. Rather than displacing single-city or core–periphery interpretations, the proposed framework extends them by positioning local heat analysis as one layer within a larger multiscale heat-governance architecture. Full article

Leucaena leucocephala is a nitrogen-fixing legume widely used in agroforestry systems, although its invasive potential poses increasing risks to wetlands and riparian ecosystems. This systematic review synthesizes current knowledge on the ecological mechanisms, environmental stressors, and management strategies associated with the invasion of L. leucocephala in humid tropical environments. Following PRISMA guidelines, 60 studies retrieved from Scopus, Web of Science, and Consensus were qualitatively analyzed. The results indicate that invasion success is strongly associated with environmental disturbances and stress conditions, particularly drought stress, altered hydrological regimes, fire occurrence, and land-use change, which reduce ecosystem resistance and facilitate species establishment. Key invasion mechanisms include high seed production, persistent soil seed banks, rapid growth, allelopathic effects, and strong resprouting capacity, leading to suppression of native vegetation and structural simplification of plant communities. Integrated management strategies combining mechanical and chemical control with active revegetation consistently showed higher effectiveness than isolated approaches. The evidence further suggests that climate-related stressors may intensify invasion dynamics and increase ecosystem vulnerability under future climate scenarios. Despite recent advances, important knowledge gaps remain regarding long-term ecosystem functioning, hydrological feedback, and adaptive management in invaded wetlands. Full article

Rapid urbanization has intensified congestion, environmental pressures, and transport inequities, thereby increasing interest in Smart Urban Mobility (SUM) as an approach that combines digital technologies, sustainable transport strategies, and data-informed decision-making to respond to these challenges. However, the evaluation of SUM remains fragmented due to the absence of harmonized assessment frameworks and the diversity of methodologies applied across smart city contexts. This study presents a systematic literature review of evaluation approaches and indicator architectures for SUM in smart city contexts. Using a PRISMA-guided screening process, 33 eligible studies were selected from 412 retrieved records. Three main methodological groups were identified: quantitative approaches, multi-criteria decision-making methods, and qualitative or participatory frameworks. A total of 273 indicators were organized into eight factor categories, confirming the multidimensional nature of smart mobility assessment while also revealing limited consistency in indicator selection and application across studies. Across the selected studies, current evaluation practices are increasingly linked to project prioritization, planning, and decision support; however, their effectiveness remains constrained by data inconsistencies, governance fragmentation, and insufficient user inclusion. These findings highlight the need for assessment frameworks that are sufficiently comparable to enable cross-city learning, yet flexible enough to reflect local contexts and institutional realities. Full article

Multimodal image fusion has emerged as a core technology for complex perception systems—such as autonomous driving, remote sensing monitoring, and medical diagnosis—by integrating complementary information from heterogeneous sensors. Given the rapid technological evolution within this field, particularly driven by the emergence of Mamba architectures, Generative Diffusion Models, and Vision Foundation Models (VFMs), traditional classification methods no longer fully encompass the ongoing paradigm shifts. Following the PRISMA guidelines to ensure the objectivity and reproducibility of the findings, this paper provides a systematic literature review and data extraction for multimodal image feature fusion. Under this standardized framework, a five-dimensional decoupling classification architecture is proposed to deconstruct models across fusion hierarchy, backbone architecture, fusion operator, supervision paradigm, and deployment constraints. Specifically, the analysis highlights the linear computational efficiency of Mamba in long-sequence modeling, the high-fidelity reconstruction capabilities of diffusion models via generative priors, and the universal semantic alignment achieved by VFMs. Furthermore, this study summarizes qualitative and quantitative evaluation metrics alongside cross-domain public datasets for performance benchmarking while discussing critical future directions, including cross-modal alignment in complex environments, parameter-efficient fine-tuning of large models, and real-time inference at the edge. Full article

Water security is crucial for human well-being and environmental sustainability. The rapid increase in urbanization, climate change, pollution, etc., leads to water scarcity in many parts of the world. Therefore, it is important to understand the concept and growing challenges of water security. Using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, 146 articles were identified for this study. The results highlight a novel aspect of the definition of water security, presenting it in a simpler, broader way with key components. The indicators for assessing water security are categorized into quantitative, qualitative, and combined types, and are further arranged across different dimensions, domains, and spatial scales. The study also examines Urban Water Security assessment methods and categorizes them into distinct methodological groups. Additionally, the studies show that only 25 articles explore artificial intelligence in the context of water security indicators. This reveals the need to address the gap between artificial intelligence and the assessment of water security. From these limited articles, artificial intelligence types and models were identified, and their applications were grouped into thematic categories. In general, this study supports improved assessment, decision-making, and sustainable water security management. Full article

Phyllodes tumours (PTs) of the breast are rare fibroepithelial neoplasms with potentially aggressive behaviour, characterised by rapid growth, a significant risk of local recurrence, and occasional metastatic spread. Optimal management remains controversial, particularly regarding surgical margins, adjuvant radiotherapy, and the relevance of molecular markers in predicting tumour behaviour. A PRISMA 2020-guided qualitative systematic review was conducted of studies published between January 2000 and December 2024 in PubMed/MEDLINE, Scopus, and Web of Science. Eligible studies included malignant PTs of the breast and addressed at least one of the following domains: molecular pathology, surgical margins and local recurrence, adjuvant radiotherapy, or predictors of recurrence and metastasis. A clinical case of malignant PT treated at our institution is presented as an illustrative study. Thirty-four studies met the inclusion criteria. Evidence suggests that margin status, stromal proliferative activity, and selected molecular markers influence recurrence risk. Several retrospective studies suggest that adjuvant radiotherapy may improve local control in selected high-risk malignant PTs, although the evidence remains heterogeneous, retrospective, and potentially affected by treatment-selection bias, and no consistent survival benefit has been demonstrated. Molecular alterations, including MED12 mutations, TERT promoter mutations, TP53 alterations, and increased Ki-67 expression, have been associated with tumour progression and aggressive behaviour. A 44-year-old woman presented with a 2.4 cm left breast mass on radiological examination. Lumpectomy revealed a malignant PT with stromal hypercellularity, nuclear atypia, and a mitotic index of 20/10 HPF with close margins. Immunohistochemistry showed positivity for CD99, Bcl-2, and CD34 with a Ki-67 proliferation index of 20%. The patient underwent wide local re-excision followed by adjuvant radiotherapy (60 Gy), and at 24-month follow-up, the patient remained disease-free. Evidence synthesis highlights the importance of complete surgical excision, multidisciplinary management, and consideration of adjuvant radiotherapy in selected malignant PTs. Emerging molecular profiling may contribute to improved biological understanding and future risk stratification of malignant PTs, although its routine clinical utility remains to be validated in prospective studies. Full article