Search Results (1,872)

Electric vehicles are rapidly reshaping global transportation, emerging as a central pillar of efforts to cut greenhouse gas emissions and end dependence on fossil fuels. This review provides a critical and integrative synthesis of recent advances in electric vehicle technologies, focusing on three interconnected domains: battery innovations, charging strategies, and grid integration. Progress in high-energy-density lithium-ion chemistries, emerging solid-state and sodium-ion batteries, and advanced battery management systems is examined with respect to their implications for driving range, safety, and lifecycle sustainability. Charging infrastructure developments, including fast and ultra-fast charging, wireless charging, and battery-swapping networks, are evaluated in terms of technical feasibility, grid impact, and user adoption. The evolving role of EVs in enhancing energy system flexibility is further analyzed through vehicle-to-grid (V2G) and smart grid interactions, with emphasis on control algorithms, grid stability, and renewable energy integration. By critically analyzing recent literature, this review identifies key technological, infrastructural, and system-level challenges, as well as emerging research directions that require coordinated optimization across domains. The insights presented aim to guide future research, technology development, and policy design toward the realization of a resilient, efficient, and scalable electric mobility ecosystem. Full article

Urban passenger hubs, such as airports and railway stations, generate concentrated last-mile demand from arriving passengers to spatially dispersed urban destinations. Fluctuating passenger arrivals and changing vehicle availability can create a mismatch between accepted demand and available service capacity. This paper aims to coordinate zone-level pricing and vehicle dispatching, so that fare-responsive accepted demand can be better aligned with available vehicle resources, while balancing operator financial performance and service reliability. Under the zonal pricing scheme, the transit operator determines a quoted zone-level fare for each service zone at every decision epoch. Newly arriving service requests accept the service when the quoted fare does not exceed their maximum acceptable per-passenger fare, after which the fare is committed. A rolling-horizon optimization model jointly determines zone-level fares and dispatching plans as request states and vehicle states evolve over time. The fare discretization property reduces the continuous pricing decision to a finite candidate zone-level fare selection problem, and a customized Rolling-Horizon Adaptive Large Neighborhood Search (RH-ALNS) algorithm is developed to solve the resulting problem efficiently. Case studies based on Nanjingnan Railway Station in Nanjing, China, demonstrate the operational value of coordinating pricing and dispatching decisions. In the baseline case, the proposed method achieves a passenger service rate of 76.75%, an accepted-passenger fulfillment rate of 96.07%, and an operating surplus of 1.145 CNY per passenger-kilometer. Holding the RH-ALNS dispatching method fixed, dynamic zonal pricing increases the objective value by 4.39%, the operating surplus per passenger-kilometer by 5.46%, and accepted-passenger fulfillment by 3.63 percentage points relative to fixed zonal fares. The findings indicate that coordinating dynamic zonal pricing with vehicle dispatching can better align accepted demand with available vehicle resources and provide practical guidance for designing reliable, resource-efficient, and financially balanced hub-based demand-responsive last-mile transit services. Full article

Continual learning has made significant progress toward enabling adaptive machine learning under evolving environments, yet real-world deployment increasingly exposes systems to persistent harsh conditions, including distributional shifts, feature evolution, delayed or scarce supervision, imbalance, noise, and recurring or novel classes. While prior research has largely addressed these challenges in isolation, growing environmental complexity motivates a broader rethinking of continual adaptation as a self-regulating process rather than solely a parameter update problem. Building upon the emerging framework of Self-Adaptive Learning (SAL), this perspective explores how learning systems may progress beyond reactive adaptation toward autonomous recognition, policy selection, and context-sensitive regulation of learning behavior under persistent uncertainty. Rather than proposing a specific algorithmic solution, we position SAL as a conceptual systems framework for organizing future research on resilient, long-lived machine learning systems. We discuss key implications for deployment robustness, evaluation, safety, and adaptive governance, while outlining major open challenges in developing practical self-regulating learners. By strengthening SAL as a forward-looking framework, this work aims to advance the broader conversation on machine learning systems capable of sustained autonomy in dynamic real-world environments. Full article

The digital transformation of financial services has increased the need for intelligent information systems capable of supporting credit risk assessment in dynamic and data-intensive environments. Traditional credit scoring approaches often face limitations when evaluating customers with limited financial histories, heterogeneous data sources, and rapidly evolving behavioral patterns. In response to these challenges, this study proposes an adaptive credit risk assessment framework that integrates machine learning, an Adaptive Champion–Challenger strategy, and MLOps practices within a unified information systems architecture. The proposed framework was evaluated using real operational data obtained from a Mexican FinTech company specializing in mobile phone financing. Three machine learning algorithms—Logistic Regression, XGBoost, and TabNet—were implemented and continuously evaluated through a rolling Champion–Challenger process supported by out-of-time validation and statistically validated model promotion criteria. Experimental results indicate that different algorithms became optimal during different evaluation periods, indicating that model effectiveness varied over time as customer behavior and portfolio characteristics evolved. While XGBoost served as the initial static baseline model, TabNet and Logistic Regression achieved superior performance during several evaluation periods, illustrating the potential benefits of adaptive model selection under changing data conditions. The proposed Adaptive Champion–Challenger Framework achieved a mean AUC of 0.817, compared with 0.798 obtained by the static baseline model. Statistical validation using the DeLong test for correlated ROC curves confirmed that the observed performance improvement was significant (p = 0.0021), providing evidence that the performance gains achieved by the adaptive strategy were unlikely to be attributable to random variation. From a Digital Transformation and Information Systems perspective, the findings suggest that maintaining predictive effectiveness in dynamic FinTech environments requires not only high-performing machine learning algorithms but also governance mechanisms that support continuous model evaluation, monitoring, traceability, and adaptive model selection. The results indicate that periodic model replacement based on statistically validated out-of-time performance can help maintain predictive effectiveness under changing data conditions while supporting model governance and operational reliability. Overall, the proposed framework provides a practical and scalable approach for implementing adaptive credit risk assessment systems that support continuous model governance, data-driven decision-making, and the management of machine learning models in alternative financing environments. Full article

Background/Objective: The management of acute ischemic stroke (AIS) has evolved from rigid time-based treatment paradigms toward tissue-based selection guided by advanced neuroimaging, thereby expanding eligibility for reperfusion therapies. To provide an updated narrative review of acute ischemic stroke (AIS) classification and management, with emphasis on extended therapeutic windows for intravenous thrombolysis (IVT) and endovascular therapy (EVT), bridging strategies, posterior circulation stroke, and implementation challenges in Latin America and other resource-constrained settings. Methods: A structured narrative review was conducted using the PubMed/MEDLINE, Embase, Scopus, and LILACS databases, covering the period from 2013 to 2025. Results: Sixty-three studies were selected from 412 records and categorized into etiologic classification, extended-window thrombolysis, EVT and bridging therapy, and posterior circulation stroke. Current evidence supports imaging-guided IVT beyond 4.5 h and EVT up to 24 h in selected patients with salvageable brain tissue, including some individuals with large infarct cores. Recent trials also support EVT for basilar artery occlusion. Tenecteplase offers practical workflow advantages in many centers, particularly where transfer delays and limited access to advanced imaging constrain timely reperfusion decisions. Conclusions: Contemporary AIS management is increasingly guided by pathophysiology and imaging rather than strict time thresholds. However, improving outcomes in middle- and low-income settings requires the implementation of adapted clinical algorithms, strengthening of stroke care networks, and optimization of referral pathways. Full article

Quantifying the impacts of climate change and extreme climatic events on crop yields is essential for safeguarding global food security. The rapid growth of data availability and advances in computational capacity have established machine learning (ML) as a critical tool for unraveling the complex, nonlinear relationships between climatic factors and agricultural productivity. This systematic review synthesizes evidence from 137 peer-reviewed studies published between 2015 and 2025 that applied ML models to assess the effects of both long-term climate trends and discrete extreme events on crop yields worldwide. Bibliometric and thematic analyses reveal a rapidly evolving field, with over 85% of studies published since 2020, and a strong concentration on staple cereals—wheat, maize, and rice—in major agricultural regions including China, the United States, and India. Random Forest (RF) was the most commonly used algorithm; ensemble and deep-learning models achieved high predictive accuracy within well-resourced study contexts. Temperature and precipitation extremes emerged as the most frequently examined stressors, with distinct methodological patterns: studies focusing on climate change trends predominantly employed RF and LSTM models, whereas those investigating extreme events increasingly adopted hybrid approaches that integrate ML with process-based crop models. This review highlights the transformative potential of ML while identifying persistent challenges, such as geographical imbalances in research coverage, the need for enhanced interpretability in extreme event attribution, and the critical importance of modeling compound extremes. Future research should prioritize the development of explainable, causally informed, and transferable ML frameworks to support equitable climate adaptation strategies in global agriculture. Full article

The transition towards smart megacities has deeply integrated Artificial Intelligence (AI) with power–transport networks. While AI empowers complex operations like multi-network coordinated dispatch and emergency rescue, current algorithm-centric perspectives largely ignore its massive physical energy costs. Accordingly, this Perspective examines the dual role of AI, considering it not only as an intelligent decision-support tool but also as a potential source of additional stress on physical infrastructure. First, through a structured synthesis of the representative literature, we deconstruct the functional dependencies between algorithms and physical infrastructures, identifying how AI reshapes the operational paradigms of power, ground transport, and aerial networks under routine and emergency scenarios. We then introduce the concept of the “Computation–Energy Paradox.” Integrating conceptual analysis with a quantitative case study of a typical community, we illustrate a plausible failure mechanism: during extreme disasters, intensified AI invocation for emergency management generates surging computational loads, which paradoxically exacerbate power shortages and reduce the operating margin of already weakened systems. In addition, we analyze core engineering bottlenecks, including spatiotemporal computation–energy mismatches and physical constraints in extreme edge environments. To address these challenges, we outline a prospective roadmap encompassing lightweight emergency AI and computation–power-coordinated offloading mechanisms. Finally, the sustainable development of such systems suggests a paradigm shift: AI must evolve from a purely virtual algorithm into a physical component of an integrated compute–power–transport system. Full article

With the acceleration of global urbanization and the large-scale development of underground spaces, underground engineering faces extremely complex and variable geological environments and high-risk construction disturbances. The widespread application of the Internet of Things and novel sensing technologies has given rise to structural health monitoring data increasingly characterized by massive volume, high dimensionality, multi-source heterogeneity, and strong spatiotemporal coupling. Traditional data processing methods based on mechanical analysis, empirical formulas, or numerical simulation have increasingly exposed limitations of insufficient accuracy, lengthy computation times, and weak generalization capability when confronted with such engineering big data. Machine learning and deep learning technologies, by virtue of their superior nonlinear mapping capability, advantages in feature extraction from massive data, and flexible architectural design, provide solutions for efficient knowledge extraction and intelligent assessment of underground engineering monitoring data. This paper reviews the current application status and frontier advances of machine learning technologies in the field of underground engineering monitoring data processing in recent years. First, the development trajectory of analytical algorithms evolving from classical shallow machine learning, through temporal and spatial deep learning, to physics-data dual-driven approaches is delineated. Second, targeting the critical challenges of missing field data and sparse sensor deployment, spatiotemporal fusion imputation techniques and spatial reconstruction methods incorporating mechanical prior knowledge are thoroughly evaluated, elucidating the paradigm shift in monitoring philosophy from discrete point-based alarming to inference-augmented sparse sensing that approximates full-field state awareness through model-dependent estimation rather than direct measurement. Third, the applications of machine learning in underground structural deformation mechanism interpretation, key influencing factor identification based on explainable artificial intelligence (AI), and rapid back-analysis of geomechanical parameters are summarized. Finally, composite network architectures and physics-constrained guidance strategies for non-stationary deformation time series prediction under complex and variable working conditions are discussed. A methodological audit of the 73 included studies—of which 33 enter the quantitative comparison tables—reveals that 26 of the 33 audited studies (78.8%) validate exclusively on single-project data, only 1 study conducts rigorous out-of-distribution generalization testing, and none of the 33 studies (0%) provides uncertainty quantification. These findings highlight cross-project generalization and probabilistic prediction as important methodological challenges. This paper aims to provide theoretical references and methodological guidance for safety early warning, intelligent construction, and full life-cycle health management of underground engineering. Full article

Land use and land cover (LULC) dynamics are critical indicators of environmental transformation and anthropogenic pressure on regional landscapes. Mirzapur, located in the transitional zone between the Indo-Gangetic Plain and the Vindhyan uplands in Northern India, represents a region characterized by ecological sensitivity, mineral-based industries, agricultural dependency, and rapid infrastructural growth. In recent decades, Northern India has experienced accelerated urbanization, population pressure, land fragmentation, and environmental stress, thus making systematic LULC monitoring crucial for sustainable resource management and policy planning. The present study examines the spatio-temporal changes in land use and land cover in Mirzapur for the years 2004, 2014, and 2024. The study employed a cloud-based platform and the Random Forest algorithm for supervised classification of multi-temporal satellite imagery. LULC maps were generated and post classification comparison was used to assess changes across the selected years. Accuracy assessment was conducted using standard validation metrics, including the Kappa coefficient, to evaluate classification. From 2004 to 2024, urban areas expanded by a relative increase of 169.36%, largely through the conversion of cropland, although the overall cropland area showed a slight increase due to agricultural expansion in other parts of the study area. A slight increase in forest cover was also observed during this period. Water bodies and barren lands declined, indicating ecological stress in the region. These changes reflect rapid urbanization, demographic pressure, and evolving socio-economic activities within the district. The LULC classification achieved overall accuracies of 96.50% (2004), 97.52% (2014), and 96.08% (2024), showing the reliability of the generated maps. The study demonstrates the effectiveness of cloud-based geospatial analysis combined with a machine learning algorithms for long-term LULC monitoring and provides valuable insights for sustainable land management and regional planning. Full article

Traditional image encryption methods based on static permutation and diffusion are vulnerable to structural cryptanalysis and often exhibit limited robustness under imperfect communication conditions. To address these issues, this paper proposes a robust topology-evolving image encryption algorithm driven by complex hyperchaotic dynamics for secure visual data transmission. First, a two-dimensional Rosenbrock–Schwefel hyperchaotic map is constructed to generate high-quality pseudorandom sequences for both permutation and diffusion. Based on this map, a bidirectional oscillatory spatial permutation mechanism governed by a dynamic linked-list topology is developed. Unlike fixed-path permutation strategies, the proposed topology continuously evolves with the system state during image traversal, thereby increasing nonlinear path complexity and improving resistance to structural attacks. Furthermore, a plaintext-dependent adaptive diffusion mechanism is designed to enhance sensitivity to plaintext variations and produce a strong global avalanche effect. Experimental results demonstrate that the proposed algorithm achieves favorable encryption performance, with an information entropy of up to 7.9994, a Number of Pixels Change Rate (NPCR) of 99.6076%, and a Unified Average Changing Intensity (UACI) of 33.4683%. In addition, the algorithm maintains good recovery performance under cropping attacks and noise interference, indicating its robustness and applicability for secure image transmission in complex communication environments. Full article

Management of acute coronary syndromes (ACS) in the cardiac intensive care unit (CICU) requires rapid diagnosis, timely reperfusion or invasive assessment, appropriate antithrombotic therapy, and early recognition of haemodynamic, electrical, and mechanical complications. This narrative review examines contemporary evidence across ST-segment elevation and non-ST-segment elevation presentations, with emphasis on decisions during hospitalisation and early secondary prevention. High-sensitivity cardiac troponin algorithms support accelerated assessment of suspected non-ST-segment elevation ACS, whereas next-generation assays require further implementation validation. Twelve-month dual antiplatelet therapy remains the default after ACS in patients without high bleeding risk; abbreviated regimens, de-escalation, cangrelor, and combined antiplatelet–anticoagulant treatment are selective strategies. Contemporary care also includes risk-based invasive timing, complete revascularisation in suitable haemodynamically stable patients, culprit-lesion-only initial PCI in cardiogenic shock, and individualised management of frailty and renal impairment. High-intensity statin therapy, with early ezetimibe when needed, is guideline-supported. Early PCSK9 inhibition and low-dose colchicine are selective strategies, whereas SGLT2 inhibitors and GLP-1RAs are established for specific comorbid indications. hs-cTnT Gen 6 and AI-assisted tools remain evidence-evolving, while multiomics and targeted anti-inflammatory therapies remain investigational. Clear separation of guideline-supported, selective, evidence-evolving, and investigational approaches is essential for clinically appropriate ACS care. Full article

Artificial Intelligence-driven detection mechanisms present both opportunities and challenges in modern systems, particularly within smart cities that rely on complex operational technologies. In the context of Saudi urban infrastructure, rapidly evolving and multidimensional cyber threats require advanced, energy-efficient security solutions. This research proposes an Artificial Intelligence-based Threat Detection Mechanism designed to proactively identify and mitigate cyber threats while maximizing energy efficiency and minimizing cost and system complexity. Purpose: The proposed theoretical framework focuses on securing sustainable smart cities by integrating Artificial Intelligence-based anomaly detection with quantum-enhanced algorithms to address high-dimensional and emerging cyber threats across interconnected urban systems. The Artificial Intelligence-based Threat Detection Mechanism enables early and proactive threat detection across sustainable smart city networks, including connections to external and global infrastructures, ensuring continuous monitoring, resilience, and service continuity. Methods: The methodology emphasizes the development of energy-efficient Artificial Intelligence models and quantum protocols, incorporating intelligent risk assessment, adaptive calibration, and automated response mechanisms. In addition, the framework introduces distributed security hubs to enhance cybersecurity robustness and scalability. Anticipated Results and Conclusions: Anticipated outcomes include improved security management policies, automated threat detection and response, and adaptive protection against evolving cyber risks. The proposed framework provides a scalable and cost-effective solution aligned with sustainability objectives. Ultimately, this research contributes a proactive and intelligent framework for securing smart city ecosystems, supporting long-term development goals and aligning with Saudi Vision 2030. Full article

The use of artificial intelligence chatbots that mirror human interaction and emotional closeness has given rise to new forms of crime. Traditional online grooming is generally conceptualized as an offence in which a human perpetrator plans, initiates, and executes criminal conduct. However, the increasing involvement of artificial intelligence has introduced novel and complex scenarios. AI systems may either autonomously engage in conduct that facilitates the sexual exploitation of children or serve as tools that enhance, automate, or scale offenders’ activities. These developments challenge the traditional understanding of the offence and expose significant gaps in existing legal frameworks. Consequently, current regulatory approaches may prove inadequate to address the evolving nature of AI-assisted online grooming and associated forms of child sexual exploitation. This study investigates the case of grooming via social media using AI chatbots and discusses whether the current criminal legislation is sufficient to address this offence. Through a legal comparative method, this study examines the legal rules in the European Union, the United Kingdom, the United States, and China, focusing on the elements of criminal acts and criminal intent and the consideration of the liability of platform operators, developers, and deployers of AI systems. The study also discusses the problem of intermediary liability rules and less mature AI governance policies to tackle the fragmented and hidden nature of algorithmic actions. The study concludes that existing criminal law frameworks face significant challenges in addressing AI-assisted grooming, particularly regarding criminal intent, foreseeability, and liability allocation. The fragmentation of responsibility among offenders, platforms, and AI developers creates regulatory and enforcement gaps in the law. Accordingly, this study advocates for a risk-based liability framework, enhanced platform accountability, greater algorithmic transparency, and stronger child-centered safeguards. Full article

Luminescence thermometry has evolved through decades of research focused on optimising materials and on extracting temperature information from isolated spectral features such as luminescence intensity ratios, bandwidth, line shift and excited-state lifetime. Despite extensive material development, these conventional methods remain fundamentally limited by construction: only a small subset of pre-selected spectral features is exploited, while the bulk of the temperature-relevant information encoded in the full spectrum is systematically discarded. A paradigm shift is presented here: Luminescence Thermometry 2.0 (LT 2.0), implemented through the newly developed SpectraSensML platform, in which machine learning regression operates on the entire emission spectrum to deliver temperature readout. The approach is demonstrated on a Yb³⁺-doped phosphor emitting in the near-infrared biological transparency window across 100 to 700 K. Yb³⁺ is a particularly demanding case: only the single ²F_5/2 multiplet emits, and its weakly thermally coupled Stark sub-levels yield modest sensitivity under conventional intensity-ratio thermometry. A total of 27 regression algorithms drawn from four families, namely tree ensembles, physics-aware regression models, kernel and instance methods, and neural networks, are systematically benchmarked. A sensor-fusion estimator that combines the first three principal components reaches an average root-mean-square error of 0.36 K on an unseen-temperature test set, a seven-fold improvement over the best luminescence intensity ratio variant. Standard normal variate (SNV) normalisation is identified as the most effective preprocessing strategy because it isolates the band-shape deformations that encode temperature. Single-component approaches that rely on the first principal component alone are shown to be quantitatively sub-optimal: multi-component regressors that exploit the first three principal components reduce the temperature uncertainty by close to an order of magnitude. The structural reason behind the failure of decision-tree ensembles on unseen temperatures is explained: their piecewise-constant predictions cannot interpolate beyond training set-points. The open-source SpectraSensML application used to obtain the results is released alongside the manuscript to enable reproducible community benchmarks. Full article

Background: Voice disorders have become an important healthcare issue owing to advances in artificial intelligence (AI) and digital healthcare. Although AI-related studies have rapidly increased, little is known about how research topics and knowledge structures have evolved during the COVID-19 pandemic and the post-pandemic period. This study investigated changes in Korean voice disorder research across these two periods. Methods: Bibliographic data were collected from the Research Information Sharing Service (RISS) and analyzed using TEXTOM V6.0. Frequency analysis, term frequency–inverse document frequency (TF-IDF), degree centrality, ego network analysis, convergence of iterated correlations (CONCOR), and the Louvain clustering detection algorithm were performed to identify major keywords, knowledge structures, and thematic clusters. Results: A total of 5332 records were analyzed, including 2818 from the pandemic period and 2514 from the post-pandemic period. The keywords “voice,” “disorder,” and “research” remained dominant throughout both periods. Degree centrality for “voice” increased from 47.52 to 50.07, while “research” increased from 39.76 to 42.17. AI-related keywords, including “recognition,” “application,” “foundation,” and “accessibility,” became more prominent after the pandemic. CONCOR and Louvain analyses revealed a shift from conventional clinical research toward AI-enabled digital healthcare and intelligent rehabilitation. Conclusions: Korean voice disorder research has evolved into a multidisciplinary, AI-driven field. These findings provide quantitative evidence of changing knowledge structures and may guide future AI-based voice disorder research and rehabilitation. Full article