Search Results (606)

Against the backdrop of global climate change, wildfires have emerged as key disturbance factors accelerating permafrost degradation. However, how wildfires affect ground-surface deformation, including spatial patterns and potential driving mechanisms, remains unclear. Therefore, in this study, the permafrost region in the northern Da Xing’an Mountains affected by the catastrophic Great Black Dragon Fire (1987) is taken as a case study. On the basis of Sentinel-1 SAR imagery acquired from 2016 to 2021, the small baseline subset interferometric synthetic aperture radar (SBAS-InSAR) technique was employed to derive surface deformation rates. These rates were combined with historical fire severity (dNBR) and topographic factors. Random forest and spatial autocorrelation analyses were used to evaluate the long-term association between wildfire disturbance and surface deformation and its potential controls. The results indicate that (1) thirty-five years after the wildfire, vegetation in the permafrost region had not fully recovered to prefire levels; (2) surface deformation from 2016 to 2021 was dominated by subsidence overall. When unburned patches within the same region were used as controls for climate-driven background subsidence, the proportion of areas experiencing severe subsidence (annual rate ≤ −50 mm yr⁻¹) reached 12.86% in high-severity fire zones, compared with 10.21% in unburned areas, suggesting that high-severity fires may amplify regional background subsidence; and (3) the random forest model had low explanatory power (R² = 0.03) and was therefore used for exploratory comparison of the selected predictors rather than for accurate prediction of surface deformation. Among the selected variables, dNBR had the greatest relative importance, followed by terrain ruggedness and slope, whereas the remaining spatial variability may reflect unmeasured hydrological and subsurface controls. This study provides a quantitative basis for understanding the wildfire-induced “abrupt degradation” of permafrost, defined here as disturbance-driven acceleration of thaw and subsidence beyond gradual climate-driven degradation, and contributes to understanding carbon–climate feedback mechanisms in permafrost regions. Full article

Cardiovascular diseases (CVDs) remain the leading cause of mortality worldwide, driving the need for reliable tools for early risk prediction. Artificial intelligence (AI) applied to electrocardiograms (ECGs) has shown strong predictive performance for future cardiac events, yet its clinical adoption remains limited by the lack of transparency and trust associated with black-box models. This systematic review examines recent advances in AI-based cardiovascular prediction, focusing on the combined challenges of multi-modal data fusion and clinically actionable explainability. Following PRISMA 2020 guidelines, we analyzed 65 peer-reviewed studies published between 2018 and 2025, identified through a systematic search of PubMed, IEEE Xplore, Web of Science, Scopus, ACM Digital Library, and Google Scholar. The reviewed literature reveals that while most AI-ECG models achieve high predictive accuracy, typically AUC 0.85–0.95, the majority rely on post hoc explainability techniques that offer limited clinical insight, and 61.5% of included studies implement no explainability method at all. External validation remains critically underutilized, performed by only 12.3% of studies, and multi-modal approaches integrating ECG data with electronic health records, biomarkers, or genomics represent only 27.7% of the reviewed literature. While these multi-modal models demonstrate improved contextualization and predictive performance, they remain insufficiently validated and inconsistently interpretable. Among studies employing XAI techniques, attention mechanisms were the most prevalent approach (28% of XAI studies), followed by saliency maps (20%), SHAP (16%), and LIME (8%). Only 9.2% of studies were prospective or clinical trials, underscoring the gap between algorithmic development and real-world clinical deployment. Applying a pre-specified four-level clinical actionability scoring framework (Level 0–3), we found that the majority of studies (61.5%) scored at Level 0 (no actionability), with only 9.2% reaching Level 3 (demonstrated clinical impact), confirming that the clinical translation gap extends beyond trial design to encompass the broader absence of clinically contextualised evaluation of AI-ECG systems. This review highlights a persistent and critical gap between predictive performance and clinical usability, and outlines four key directions for developing AI-ECG systems that can better support trustworthy clinical decision-making: (1) developing inherently interpretable architectures, (2) advancing unified multi-modal fusion and explanation frameworks, (3) establishing standardized benchmarks for explainability evaluation, and (4) conducting robust prospective validation measuring real-world patient outcomes. Full article

Hydrogen (H₂) is expected to play a critical role in modern industry, particularly in ammonia synthesis, petroleum refining, and low-carbon transportation. The safe storage of H₂ remains a major challenge due to its low volumetric density under ambient conditions. Metal–Organic Frameworks (MOFs), highly porous crystalline materials, have emerged as promising H₂ storage candidates owing to their high surface areas and tuneable pore structures. Molecular simulations such as grand canonical Monte Carlo or density functional theory are costly and limited in exploring large material spaces, motivating efficient predictive tools to accelerate discovery. Here, Machine Learning (ML) techniques are compared to an explainable artificial intelligence (XAI) approach using symbolic regression (SR), trained on 10,123 experimentally measured H₂ adsorption datapoints from real-world MOFs. The best performing model achieved a goodness of fit of 0.9986 with lower computational demand, but reduced interpretability, addressed using XAI analysis and clustering. SR achieves a lower goodness of fit of 0.914 but produces a physically meaningful equation highlighting structural features driving high gravimetric efficiencies. These results demonstrate strong ML capability for predicting how MOF properties and environmental conditions affect H₂ uptake. This offers engineers and researchers a practical means of screening potential MOFs for H₂ storage applications, with the XAI analyses providing additional confidence in the predictions. They allow researchers to understand the physical reasoning behind each output, assess the reliability of individual predictions, and make fully informed decisions, enabling predictive models to be acted upon with confidence in real-world contexts. Full article

There is an urgent and persistent need to design efficient and sustainable methods to manufacture chemicals on a large scale. Heterogeneous photocatalysts use light to drive organic reactions and offer high recyclability and improved efficiencies for chemical synthesis. However, a detailed study of these photocatalysts using standard laboratory analytical techniques is challenging due to their poor solubility. Successful application of heterogeneous photocatalysts in the chemical industry requires the development of a robust analytical technique that can be used as a predictive and scalable tool for their photocatalytic performance. Herein, we report a simple approach that uses the color of cotton-bound heterogeneous photocatalysts as a potential indicator of their performance. These photocatalysts were synthesized by covalently attaching perylene-based molecular photocatalysts to the surface of cotton using amino-substituted triethoxysilane as the linker. Colorimetry coupled with NMR analysis revealed two important findings: (i) cotton-bound photocatalysts catalyzed sulfide oxidation to sulfoxide under blue-light illumination, and (ii) a general relationship was observed between color intensity and catalytic performance, with darker samples generally exhibiting faster reaction rates. These findings suggest that color may serve as a simple and rapid tool for assessing photocatalyst performance. Future studies will focus on enhancing the reproducibility of photocatalyst binding procedures and validating the color–performance relationships in a wider range of color intensities of the cotton-bound photocatalysts. Full article

Background/Objectives: Obstructivesleep apnea (OSA) is one of the most widespread forms of sleep disease, affecting over 936 million adults globally. The health consequences of obstructive sleep apnea (OSA) are well documented; however, it remains largely underdiagnosed because the current gold-standard diagnostic method, polysomnography (PSG), is often costly, time-consuming, and unavailable in many healthcare settings. To address these challenges, this study presents a novel explainable deep learning (DL) framework for automated multi-class OSA severity classification using single-lead electrocardiogram (ECG) signals. Methods: The proposed framework integrates a hybrid CNN–BiLSTM architecture with explainable artificial intelligence (XAI) techniques to generate clinically meaningful predictions and explanations across four OSA severity classes: Normal, Mild, Moderate, and Severe. The framework was evaluated using the publicly available PhysioNet Apnea-ECG dataset (70 recordings) together with an institutional ECG dataset (150 recordings), resulting in a combined cohort of 220 recordings. Results: The proposed framework achieved an overall classification accuracy of 94.7%, with sensitivity and specificity values of 92.3% and 96.1%, respectively. Furthermore, the proposed model consistently outperformed conventional machine learning algorithms, including Support Vector Machine (SVM), Random Forest, and XGBoost, by 5.5%, 4.2%, and 2.9%, respectively. To enhance transparency and clinical trust, SHAP (SHapley Additive exPlanations) was employed to identify the most influential physiological predictors driving model decisions. Heart rate variability features, particularly RMSSD and pNN50, emerged as the strongest indicators of OSA severity. Moreover, computational efficiency analysis revealed that the model required only 0.23 s to process a 60 s ECG epoch on a standard computing platform, supporting its suitability for real-time deployment. Conclusions: The findings demonstrate that explainable deep learning applied to ECG signals can provide accurate, interpretable, and computationally efficient assessment of OSA severity. The proposed framework may support OSA screening, clinical triage, and early intervention, particularly in resource-constrained healthcare environments. Full article

Trajectory prediction is a critical component of autonomous driving systems, enabling vehicles to anticipate future motion behaviors and perform safe decision-making in dynamic traffic environments. While recent trajectory forecasting methods achieve state-of-the-art prediction accuracy, many operate as black-box systems and are evaluated primarily on high-end computing platforms, limiting their interpretability and practical deployment feasibility in resource-constrained autonomous driving systems. To address these limitations, this work proposes an interpretable and edge-deployable spatio-temporal trajectory prediction framework for autonomous driving. The proposed architecture integrates a Temporal Convolutional Network with Multi-Head Self-Attention (TCN–MHSA) in ActorNet for selective temporal modeling, a Lane Graph Attention Network (LaneGAT) for structured spatial reasoning, and a multi-stage FusionNet for actor–lane interaction. To improve model interpretability, a comprehensive Explainable AI (XAI) evaluation framework is introduced, including temporal sensitivity analysis, interaction-aware perturbation studies, spatial influence analysis, and gradient-based feature attribution methods. These analyses provide insights into how the model captures temporal motion dependencies, neighboring vehicle interactions, and environmental context during trajectory prediction. To improve the robustness of the interpretability analysis, temporal sensitivity was additionally evaluated over 100 validation scenes, demonstrating that recent observations consistently exert the greatest influence on trajectory prediction, while neighboring interaction effects gradually diminish with increasing spatial separation. Furthermore, practical real-world deployment feasibility is investigated on the NVIDIA Jetson Xavier NX platform using edge-aware optimization strategies, including mixed-precision inference and graph-complexity reduction techniques for efficient resource-constrained inference, achieving 125.74 ms latency at 12.86 W. Additional edge deployment comparisons with HiVT and SIMPL approaches under identical hardware conditions demonstrate that the proposed framework provides a more favorable balance between computational efficiency and embedded deployment performance. Experimental evaluation on the Argoverse 1 dataset demonstrates a minimum Average Displacement Error (minADE) of 0.90 m, a minimum Final Displacement Error (minFDE) of 1.50 m, a Miss Rate (MR) of 0.19, and DAC = 0.95, while establishing an accuracy–deployability operating point under embedded hardware constraints with low power consumption and practical inference throughput. Full article

This paper presents a comprehensive review of control techniques, simulation mechanisms, and validation methods applied to DC-DC converters, with a focus on high-step-up topologies used in renewable energy systems such as photovoltaic and wind power applications. Control strategies including classical PID, fuzzy logic, sliding mode, and model predictive control (MPC) are analyzed in terms of performance, robustness, and implementation complexity. Simulation platforms and hardware-in-the-loop (HIL) validation frameworks are also discussed as key enablers for rapid prototyping. The findings reveal a clear trend toward intelligent and hybrid control schemes that combine nonlinear techniques with artificial intelligence to address the inherent nonlinearities and parametric uncertainties of DC-DC converters. However, challenges remain in real-time implementation due to computational demands, which drives the need for future developments focused on the (i) integration of AI-based controllers with low-cost embedded platforms, (ii) standardization of HIL-based validation workflows, and (iii) optimization of converter topologies for specific applications such as electric vehicle charging and photovoltaic grid integration. Looking forward, the convergence of advanced control algorithms, real-time validation platforms, and application-specific converter design is expected to define the next generation of power electronics systems, enabling more efficient, reliable, and scalable renewable energy integration. Full article

Deep visual recognition has achieved remarkable success across various domains, including medical imaging, autonomous vehicles, and security systems. However, the black-box nature of deep learning models poses challenges in terms of transparency and trust, especially in critical applications where human understanding is essential. Explainable AI (XAI) seeks to address these concerns by providing human-interpretable explanations for model predictions. This review explores the key techniques for explainability in deep visual recognition, including model-agnostic methods such as LIME and SHAP, model-specific approaches like saliency maps and feature visualization, and intrinsically interpretable models like decision trees and rule-based systems. We also discuss the evaluation of explainability through metrics like fidelity, consistency, and stability, and explore the challenges of balancing model performance with interpretability. Furthermore, we examine applications of XAI in medical imaging, autonomous driving, security and surveillance, agriculture, satellite imagery and remote sensing, industrial inspection, and visual forensics, highlighting how domain-specific data and operational constraints affect the required form and validation of explanations. Finally, we address current research gaps and propose future directions for enhancing the robustness and human–AI interaction in explainable visual recognition systems. As AI continues to be integrated into safety-critical domains, the development of explainable, transparent, and trustworthy AI systems will be crucial for their widespread adoption and ethical use. Full article

Ultracold neutrons (UCNs) are free neutrons with kinetic energies so low that their equivalent thermal temperature lies below 3.5 mK (below 3 × 10⁻⁷ eV). At these extreme energies, neutrons exhibit de Broglie wavelengths on the order of hundreds of angstroms and move slowly enough to be confined in material, magnetic, and gravitational traps through total internal reflection. For context, this is about three orders of magnitude colder than the 1 K regime used in superfluid helium UCN sources, which underscores why these neutrons are called “ultracold”: their equivalent thermal energy is comparable to millikelvin physics, even though UCN sources themselves typically operate at 0.8–5 K and produce UCN through superthermal downscattering rather than thermal equilibrium. Over the past several decades, substantial progress in ultracold-neutron source technology has been achieved through the transition from mechanical neutron turbines to superthermal converters based on solid deuterium and superfluid helium. This review provides a comprehensive analysis of modern reactor-based (ILL, PNPI, TRIGA) and spallation-driven (PSI, TRIUMF, SNS, ESS) UCN sources, together with next-generation facilities targeting UCN densities of 10³–10⁴ cm⁻³. Particular attention is devoted to anomalous neutron losses during storage. It is shown that hydrogen-containing surface contaminants, inelastic scattering processes, and wall-induced depolarization contribute significantly to losses beyond those predicted for ideal materials. Current approaches for loss reduction are discussed, including diamond-like carbon coatings, magnetron sputtering techniques, optimization of the ortho–para ratio in neutron converters, and purification of superfluid ⁴He from trace concentrations of ³He impurities. The review further examines key precision experiments that drive advances in UCN technology, including investigations of the neutron lifetime discrepancy and searches for the neutron electric dipole moment at sensitivities approaching 10⁻²⁷–10⁻²⁸ e·cm as probes of CP violation and baryon asymmetry of the Universe. Finally, future directions for increasing UCN density, extending storage times, and enhancing the sensitivity of fundamental physics experiments are discussed. Full article

This study analyses the evolution of the financial literature on climate risk, examining the integration of artificial intelligence techniques into its measurement and management. To this end, a bibliometric approach is employed based on 221 articles indexed in the Web of Science Core Collection, using the Bibliometrix package. Moving beyond existing descriptive bibliometric reviews on ESG and green finance, the novelty of this paper lies in its analytical focus on how financial science operationalises quantitative AI mechanisms to price and integrate climate transition risk into asset and portfolio valuation. The structural analysis reveals that natural language processing (NLP) and digital transformation acting as driving motor themes, suggesting that the reviewed literature associates AI innovation policies with the mitigation of corporate greenwashing and enhance information transparency. Furthermore, while machine learning algorithms establish the cross-cutting predictive foundation for risk assessment, empirical evidence unveils a critical academic shift of traditional ‘financial performance’ towards a declining quadrant, indicating that empirical studies frequently find that that multi-phase investments in risk technologies do not yield immediate financial returns. Finally, the study maps a persistent geographical gap where emerging markets lack the data infrastructure of advanced economies, alongside isolated high-dimensional causal econometric niches like double machine learning. This analytical mapping provides key implications for global risk management and future quantitative research avenues. Full article

Metabolic dysfunction–associated steatotic liver disease (MASLD) has become the leading global cause of chronic liver disease, driving cirrhosis, hepatic decompensation, hepatocellular carcinoma (HCC), and a disproportionate burden of major adverse cardiovascular events (MACEs). Fibrosis stage remains the strongest prognostic hepatic determinant, yet liver biopsy and Hepatic Venous Pressure Gradient (HVPG)—despite their diagnostic value—are invasive and unsuitable for population-level risk assessment. Over the past two decades, non-invasive tools (NITs), including serological scores, elastography-based techniques, and composite models, have transformed fibrosis and portal hypertension (PH) evaluation, though their accuracy declines in obese or comorbid patients and their ability to predict long-term hepatic and extrahepatic outcomes remains limited. These gaps have catalyzed the emergence of AI-driven, multimodal predictive systems capable of integrating biochemical, imaging, clinical, and histopathological data into individualized and dynamically updated risk profiles. This narrative review synthesizes current evidence on traditional and next-generation NITs, highlighting how AI-enhanced approaches may overcome long-standing diagnostic constraints and enable a unified assessment of hepatic and cardiometabolic risk. Collectively, these innovations outline a path toward precision hepatology, with the potential to reshape surveillance strategies, refine prognostic stratification, and improve outcomes across the full spectrum of MASLD. Full article

Accurate state-of-charge (SOC) estimation is essential for preventing battery degradation, improving energy management, and providing reliable driving-range predictions in electric vehicles (EVs). The extended Kalman filter (EKF) is a widely adopted model-based estimation technique and remains an industry-standard approach in EV battery management systems (BMS). However, its performance can be degraded by model nonlinearities, parameter uncertainties, measurement noise, and interference bursts commonly encountered in real-world operating environments. To overcome these limitations, this paper proposes an event-triggered adaptive SOC estimation framework that integrates a recursive least squares (RLS) filter with the EKF. In the proposed approach, the RLS filter recursively updates its weighting coefficients in real time to compensate for model uncertainties and measurement disturbances, thereby generating an alternative residual signal for SOC estimation. An event-triggered mechanism dynamically selects the most reliable innovation sequence for updating the EKF state estimate, enhancing estimation robustness under adverse operating conditions. A second-order RC equivalent circuit model (ECM) is employed as the nominal battery model, and a Hybrid Pulse Power Characterization (HPPC)-based current profile is used to evaluate performance over the entire SOC operating range. Extensive simulations are conducted to assess the effectiveness of the proposed event-triggered RLS-EKF algorithm under various noise levels and interference-burst scenarios. The estimation accuracy is compared with that of the conventional EKF, cubature Kalman filter (CKF), and square root cubature Kalman filter (SR-CKF) using root mean square error (RMSE) and mean absolute error (MAE) as performance metrics. Simulation results demonstrate that, under regular noise conditions and short-term interference bursts, the proposed event-triggered RLS-EKF achieves estimation performance comparable to that of the SR-CKF while consistently outperforming the EKF and CKF in both RMSE and MAE. Under long-term interference-burst conditions, the proposed method further surpasses the SR-CKF, achieving approximately 10% improvement in overall estimation accuracy as measured by RMSE and MAE. These results confirm the effectiveness and robustness of the proposed framework, highlighting its potential for practical implementation in advanced EV battery management systems. Full article

This paper presents a high-performance sensorless control strategy for Brushless DC (BLDC) motors based on predictive current control combined with back-EMF-based speed and position estimation. The main objective is to achieve accurate speed regulation and fast dynamic response without the need for mechanical speed sensors, thereby reducing system cost, improving reliability, and simplifying hardware complexity. The proposed predictive current control algorithm ensures precise current tracking and rapid torque production under varying operating conditions, including speed reference changes and load torque disturbances. A comprehensive comparative analysis is conducted between the proposed sensorless approach and a conventional sensored control scheme. The obtained results demonstrate that the sensorless controller achieves speed, torque, and current performance that is nearly identical to the sensored system, with negligible differences in rise time, overshoot, steady-state error, and torque ripple. The dynamic response remains smooth and well-damped, confirming the effectiveness of the proposed estimation technique in maintaining accurate rotor synchronization under transient conditions. In addition, the influence of a proportional–integral speed controller with and without anti-windup compensation is investigated. The results show that the anti-windup mechanism significantly improves transient performance by reducing overshoot, eliminating startup undershoot, shortening settling time, and mitigating speed estimation errors during large reference changes and actuator saturation conditions, while preserving zero steady-state error. To validate the real-time feasibility of the proposed control strategy, a Processor-in-the-Loop (PIL) co-simulation platform is implemented using the C2000 LaunchXL-F28379D digital signal processor. The PIL results confirm that the algorithm can be executed under strict real-time constraints with acceptable computational burden, memory usage, and execution time. Overall, the proposed sensorless predictive current control strategy demonstrates strong robustness, high accuracy, and practical applicability for industrial BLDC motor drive systems operating under diverse and dynamic conditions. Full article

Model reference adaptive system (MRAS)-based techniques are widely used for sensorless speed estimation in permanent magnet synchronous motor (PMSM) drives. However, conventional MRAS approaches rely on PI/PID controllers in the adaptation mechanism, which require careful tuning and may suffer from performance degradation under varying operating conditions. This paper proposes a tuning-free finite speed-set model reference adaptive system (FSS-MRAS) for sensorless speed estimation in electric vehicle (EV) applications. The suggested design methodology substitutes the traditional adaptive controller design with a speed-set predictive selection approach, hence removing the need for controller parameter tuning and making the scheme more straightforward to implement. To validate the performance of the FSS-MRAS algorithm, a comprehensive EV model is simulated according to the FTP-75 driving cycle, which provides very dynamic and realistic test conditions. Simulation results prove that the suggested algorithm guarantees accurate speed estimation in all operational scenarios, including low-speed, high-speed, and high-dynamics modes. It is also demonstrated that the estimated value matches the actual value with negligible errors at the steady-state condition and stays bounded around the real value within ±40 rpm during transient dynamics conditions. Moreover, compared with the conventional MRAS scheme, the proposed approach eliminates the need for PI controller tuning while maintaining accurate and stable speed estimation under the highly dynamic operating conditions of the FTP-75 driving cycle. Additionally, the proposed scheme proves stable performance without any oscillation or divergence issues through the whole driving cycle. Full article

As generative artificial intelligence (GenAI) increasingly permeates educational contexts, understanding the factors driving learners’ Behavioral Intention (BI) toward GenAI-powered tools has become critical. This study integrates the Technology Acceptance Model (TAM), the Task-Technology Fit (TTF) framework, and privacy and ethical risk considerations to explore the determinants of Chinese higher education students’ Behavioral Intention to adopt these tools. Data were collected from 716 students via a structured self-reported questionnaire. A multi-stage analytical approach was employed by integrating structural equation modeling (SEM) with artificial neural networks (ANN) and support vector regression (SVR). SEM was first utilized to validate the theoretical hypotheses and the measurement model. Subsequently, ANN and SVR models were constructed to explore non-linear relationships and rank the importance of core predictors for Behavioral Intention, including Perceived Ease of Use (PEU), Privacy and Ethical Concerns (PEC), Perceived Technical Features (PTF), and TTF. The modeling performance of the two algorithms was then rigorously compared. The SEM results indicate that PTF exerts an indirect impact on Behavioral Intention via the sequential mediation of Task-Technology Fit and Perceived Usefulness (PU), while PEU positively influences both Perceived Usefulness and Behavioral Intention. Notably, PEC did not exhibit a significant negative effect on users’ Attitude (ATT) or Behavioral Intention. These findings were further elucidated by the machine learning analyses, where PTF and PEU emerged as the dominant predictors, whereas the non-linear contribution of PEC was marginal. Furthermore, SVR outperformed ANN in terms of predictive accuracy and model stability. This study demonstrates the efficacy of combining theoretical modeling with machine learning techniques to elucidate the adoption mechanisms of GenAI in higher education. In addition, preliminary teaching observations in undergraduate mathematics and logistics management courses link quantitative results with actual learning scenarios. We acknowledge that future research should validate these patterns using observed behavioral data. Full article