Search Results (33,843)

High-resolution satellite imagery has substantially improved the monitoring of vegetation dynamics; however, the influence of spatial resolution and pixel aggregation on NDVI time series consistency remains insufficiently quantified, particularly across multiple native resolutions within a single sensor platform. This study evaluates how Sentinel-2 spatial resolutions (10 m, 20 m, and 60 m) and two pixel aggregation methods (pure-pixel and centroid) affect NDVI time series in 14,031 grassland plots across mainland Spain over the period 2018–2023. High-quality NDVI time series were selected using the Interpolation Efficiency Indicator (IEI), and discrepancies relative to a 10 m pure-pixel baseline were quantified through the Time Series Angle Distance (TSAD) and Root Mean Square Error (RMSE). A sensitivity check confirmed that the radiometric differences between Band 8 (10 m) and Band 8A (20/60 m) introduce negligible bias compared with genuine spatial-resolution effects. Formal non-parametric statistical testing—omnibus Kruskal–Wallis with epsilon-squared (${\epsilon }^2$) effect sizes and pairwise Cliff’s Delta comparisons—was applied to assess the magnitude and practical significance of the observed differences across plot area categories and Köppen climate groups (B, Cs, Cf). Results show that coarser resolutions (60 m) substantially reduce NDVI reliability, excluding more than half of the plots under the pure-pixel criterion and smoothing temporal variability, whereas 10 m and 20 m resolutions preserve most spectral and temporal information. The 20 m resolution introduces moderate but non-severe phenological distortion (median TSAD ≈ 0.05 rad, RMSE ≈ 0.026) with a 78% reduction in data volume and 72% reduction in processing time. The choice between pure-pixel and centroid sampling has negligible impact at 10–20 m but becomes relevant at 60 m, where pure-pixel selection reduces errors from spectral mixing at the cost of severe sample attrition. Parcel area strongly conditions the error metrics, with large effect sizes (${\epsilon }^2=0.273$) in the smallest plots, while Köppen climate classification decisively shapes TSAD (up to ${\epsilon }^2=0.447$), indicating that spatial degradation distorts phenological patterns differently across climate classes. These findings support a multi-scale monitoring strategy: 10 m for fragmented, heterogeneous grasslands (<3 ha), 20 m as a computationally efficient alternative for homogeneous areas (>10 ha), and outline potential implications for policy frameworks such as the Common Agricultural Policy (CAP). Full article

Satellite-Derived Bathymetry (SDB) based on the Stumpf log-ratio method routinely uses a fixed parameter n = 1000, a convention that has rarely been evaluated systematically. This study assesses empirically how SDB error varies with n across 31 bathymetric scenarios at 18 coastal sites with contrasting morphological and water quality conditions using Sentinel-2 imagery, ACOLITE atmospheric correction and in situ reference depths in the −0.5 to −6 m range. For each scenario, an optimal n was retrieved by minimizing RMSE against the reference bathymetry, and the response of the error to n was characterized. Two stable regions (a left and a right plateau) are identified in the error-versus-n curve, with most scenarios (26 of 31) reaching their optimum on the left plateau at small n values. Replacing n = 1000 with the site-specific optimum, the mean R² increases from 0.70 to 0.84 and reduces RMSE by 21.9% on average (~18 cm). A single generalized value n_median(LP) = 2.2, transferable across sites, recovers most of this gain; the unscaled, parameter-free case n = 1 alone already reduces RMSE by 15.9% on average relative to n = 1000 and outperforms it in 25 of the 31 cases. The margin of improvement obtainable by tuning n is strongly correlated (Spearman) with hue angle, chlorophyll-a and Trophic State Index, while turbidity and Secchi depth show no correlation, consistent with the loss of reliability of these last two indicators in optically shallow waters. The benefit of using a small n is greatest in greener, higher-chlorophyll waters, whereas in very clear, blue waters (hue > 160°, Chl-a < 2 µg/L) n = 1000 remains a defensible choice. These results support replacing n = 1000 by n ≈ 2–3 as the practical default for Stumpf-based SDB in the studied depth range and provide a water-quality-based criterion to anticipate the expected benefit of parameter tuning. These improvements make it possible to use imagery from less clear waters for SDB extraction, broadening the range of images suitable for beach monitoring. Full article

Structural health monitoring (SHM) plays a critical role in the early identification and assessment of structural damage, thereby enhancing the safety and reliability of civil infrastructure. Structural damage identification generally encompasses three key tasks: damage detection, localization, and quantification. While extensive research has been conducted on each of these tasks individually, relatively few studies have integrated all three components into a unified framework for comprehensive structural condition assessment. Physics-based approaches require an accurate finite element model (FEM), which is often difficult to calibrate to accurately represent the behavior of the actual structure. In contrast, data-driven approaches rely on sufficient labeled data collected from the actual structure, which is likewise challenging to acquire in practice. To address these limitations, this work proposes an integrated hierarchical physics-informed domain adaptation (I-HierPhyDA) framework that performs damage detection, localization, and severity classification in a hierarchical manner. The proposed framework bridges the gap between the reduced-order FEM and the higher-fidelity FEM by generating vibration signatures that are consistent across both domains. Furthermore, the proposed framework enables structural damage localization without requiring damage-state data from the target domain during training, while damage severity classification is performed using transductive domain adaptation with unlabeled damaged-state data from the target domain. The proposed framework is systematically evaluated using the numerical ASCE benchmark models under structural uncertainties and measurement noise. The results demonstrate that the proposed approach achieves accurate structural damage detection and localization. For structural damage severity classification, it achieves the highest mean accuracy and Macro-F1 score while exhibiting the lowest standard deviations for both metrics among the baseline and ablation methods, demonstrating its effectiveness for comprehensive structural condition assessment. Future work will focus on experimentally validating the proposed approach using measured data from laboratory or field structures. Full article

Environmental comparisons of blockchain systems are often reduced to electricity per transaction, although operational services also depend on validators, cloud gateways, storage, monitoring, key management, recovery, and hardware replacement. This study develops a lifecycle assessment framework with non-compensatory security and utility gates and applies its operational-energy module to Ethereum’s transition from proof of work (PoW) to proof of stake (PoS). Three units are separated: 24 h of observed network operation (FU-O), 24 h of fully security- and utility-qualified service (FU-Q), and one million included layer-1 transactions (FU-B, an attributional diagnostic). FU-Q is not evaluated because several mandatory gates remain UNRESOLVED. Matched 28-day activity windows are combined with dated network-energy estimates, not continuous metering over those windows. Using the independent Cambridge baseline, daily operational electricity decreased from 58,617.39 to 5.376 MWh, a factor of 10,903.5 and a reduction of 99.99083%. The CCRI replication factor was 8804.9, while an adverse bounded pairing still yielded a factor of 3424.7. Across 100,000 Monte Carlo realizations generated by the supplied executable workflow, the median FU-O reduction was 99.98698%, with a central 95% interval of 99.97492–99.99441%. Jansen sensitivity analysis identified post-Merge annual energy as the dominant input to the FU-O factor. The additional post-Merge cloud and annualized embodied burden required to eliminate FU-O parity was 21,408 GWh/year. The result is a bounded operational-energy application and does not establish the complete lifecycle, cloud, cybersecurity, or functional-equivalence framework. Full article

Background: Quantitative assessment of dynamic balance remains challenging in clinical practice. Wearable inertial measurement unit (IMU)-based technologies offer an objective and accessible approach for monitoring postural control. This study aimed to develop and preliminarily validate a wearable IMU-based trunk sway monitoring system (SwayTracker) and investigate trunk sway characteristics in stroke patients walking with and without arm sling support. Methods: A sternum-mounted IMU system was developed to quantify dynamic trunk sway during walking. Fifteen healthy adults and fourteen stroke patients participated. The healthy participants established normative reference values, whereas the stroke patients completed walking trials with and without arm sling support. Trunk sway was quantified using anteroposterior (AP) and mediolateral (ML) deviations and a polar-coordinate-based sway model. Results: The healthy reference cohort exhibited a mean trunk sway magnitude (radius) of 8.20 ± 3.16°. The stroke patients demonstrated greater sway during unsupported (15.36 ± 5.83°) and sling-assisted walking (15.35 ± 4.59°). Although overall sway magnitude remained unchanged, the mean sway direction shifted from 69.49° to 110.20°, indicating a redistribution of trunk sway from the anterior-right toward the anterior-left quadrant. Forward sway remained the dominant AP component, whereas ML sway shifted from predominantly rightward to leftward with sling use. Conclusions: SwayTracker provides a feasible method for objective assessment of dynamic trunk sway during walking. The stroke patients exhibited increased sway magnitude and altered directional organization compared with healthy individuals. Arm sling use primarily modified ML postural compensation patterns rather than reducing overall trunk sway, highlighting the potential of wearable trunk sway monitoring for gait and balance assessment in neurological rehabilitation. Full article

The PPR Global Research and Expertise Network (GREN) plays a central role in identifying research priorities supporting the Global Peste des Petits Ruminants (PPR) Eradication Programme (GEP). A scoping review of PPR epidemiological research published between 2015 and 2025 was conducted to assess progress and remaining gaps across research domains relevant to GREN priorities. A total of 670 PPR-related publications were retrieved. Epidemiological research constituted 46.9% of the PPR literature, pathogenesis (19.1%), and diagnostics (15.6%), followed by vaccinology (9.5%) and immunology (7.9%). From the 273 epidemiological studies, most originated from Africa (53.1%) and Asia (37.7%). The epidemiological studies were categorized into seven domains. Molecular epidemiology and virus strain characterization represented the largest proportion of published studies (40.0%), followed by surveillance and disease monitoring (23.8%), transmission dynamics and risk modelling (12.5%), vaccination epidemiology (10.6%), analytical and risk factor epidemiology (8.4%), control strategy evaluation (2.9%), and socioeconomic research (1.8%). Progress towards GREN research priorities was variable. Molecular epidemiology and surveillance demonstrated the greatest methodological advancement, supported by widespread application of phylogenetics, serological surveillance, and expanding sequencing capacity across endemic settings. Transmission dynamics and risk modelling also showed increasing analytical sophistication through spatial modelling, network analysis, ecological suitability modelling, and dynamic simulation approaches, although applications remained geographically concentrated and only partially integrated into operational decision-support systems. In contrast, vaccination epidemiology, comparative evaluation of control strategies, and socioeconomic research remained limited in volume and operational integration. Across domains, important gaps persisted in implementation-focused evaluation, integrated surveillance systems, and translation of research evidence into adaptive eradication planning and policy development. Overall, the findings demonstrate substantial growth in the global PPR epidemiological evidence base over the past decade but also reveal a persistent imbalance across research domains and geographies. While major advances have been achieved in molecular characterization and descriptive surveillance, greater integration of epidemiological, operational, modelling, and socioeconomic evidence is needed to strengthen adaptive eradication planning and improve alignment between research priorities and policy implementation. Research and policy need to be better integrated, with policy needs driving research focus and research findings then informing policy. Experience from rinderpest eradication demonstrates that strong research–policy interfaces are essential to ensure that scientific evidence is translated into effective eradication strategies and decision-making. Full article

Potato quality is determined by multiple physicochemical indicators, including dry matter content (DC), starch content (SC), and color parameters (lightness L*, redness a*, yellowness b*, and browning index (BI)). Conventional spectrometers are costly, non-portable and lack wireless in-situ monitoring, restricting efficient postharvest quality assessment. Chemical methods are destructive and inefficient for field inspection and high-throughput detection. The primary objective of this study was to develop and validate a low-cost wireless 12-channel visible/near-infrared (Vis/NIR) spectral sensing system, comprising 6 Vis channels and 6 NIR channels, for the real-time non-destructive prediction of six potato quality indicators. After preprocessing the spectral data with mean normalization, a multiple linear regression (MLR) model was established to optimize the prediction performance of quality parameters. The six indicators evaluated were DC, SC, L*, a*, b*, and BI. Statistical analysis and cross-validation were further conducted to quantitatively evaluate the stability and credibility of the prediction model. Among these, the b* parameter demonstrated the most robust predictive performance, achieving a cross-validated coefficient of determination (R²_CV) of 0.881. The MLR model was integrated into the sensing hardware to realize synchronous data collection and prediction. This study provides a validated, low-cost, wireless solution for rapid potato quality assessment under controlled conditions, offering a potential alternative to conventional spectrometers and destructive chemical methods. Full article

Self-supervised knowledge distillation is widely used to compress reusable encoders, but an untrusted distillation implementation can itself become an attack surface. We study an algorithm-level threat in which the teacher encoder and user-visible distillation dataset remain unchanged, while malicious code internally generates trigger-bearing views and optimizes an additional backdoor objective. To instantiate this threat, we propose soft gradient collaboration (SGC), which combines distribution-alignment-based distillation, target-representation-based backdoor design, and conflict-avoidance gradient collaboration to reduce interference with benign representation transfer while embedding a trigger-to-target association in the student encoder. Experiments on CIFAR-10 and STL-10 show that SGC maintains competitive downstream accuracy and effective non-target attack success. Quantitative CKA, feature-distribution, and class-structure analyses further indicate that SGC retains clean representations closer to benign distillation than fixed scalarization or removal of distribution alignment. Its no-defense attack success is not the highest among the compared attacks; instead, its main empirical advantage is stronger residual attack persistence after MIMIC, MKD, and SSLDefender. Under SSLDefender, SGC retains 9.12% non-target ASR on CIFAR-10 and 9.06% on STL-10, the highest residual values among the compared attacks. Additional experiments with a compact ResNet-18 student, multiple target classes and trigger configurations, and a supplemental CIFAR-100 setting broaden the empirical evaluation across student capacity, target semantics, trigger configurations, and label-space complexity. These results show that security assessment of self-supervised distillation should include executable training logic in addition to model weights and visible data. The concealment considered here is limited to dataset-only inspection and clean-output validation; SGC is not claimed to evade source-code auditing, runtime data-flow monitoring, or training-log inspection. Full article

This study presents a standalone yarn-level assessment of the mechanical and functional durability of polymer-encapsulated electronic yarns (E-yarns) for wearable electronic textile applications. The investigated E-yarns consisted of miniaturised electronic components soldered onto fine conductive wires, protected by polymer encapsulation, and enclosed within braided textile yarn structures. This heterogeneous architecture enables textile-compatible functionality but creates local regions that may be susceptible to damage under various deformation modes. E-yarns incorporating light-emitting diode, photodiode, and resistor components were evaluated under cyclic bending fatigue, torsional fatigue, quasi-static tensile loading, and wash durability conditions. Electrical measurements were used to monitor functional degradation and failure, while X-ray imaging, scanning electron microscopy and finite element analysis were used to examine structural damage, failure localisation, fracture surface morphology, and local stress and strain distribution. The results show that E-yarn durability depended on the imposed loading condition, with distinct mechanical and functional responses observed across the different test modes. The polymer-encapsulated region emerged as a mechanically important feature of the E-yarn architecture, particularly at transitions between encapsulated and non-encapsulated regions. By addressing multiple deformation and loading conditions at the standalone yarn level, this work provides a systematic reliability assessment of polymer-encapsulated E-yarns, enabling intrinsic failure mechanisms to be distinguished from textile integration effects and supporting the development of more reliable yarn-based electronic textiles. Full article

Background/Objectives: To evaluate the combined effect of systemic and ocular comorbidities on postoperative recovery after phacoemulsification by assessing corneal status, visual acuity outcomes, and visual recovery dynamics using a real-world data study approach. Methods: This retrospective analysis was conducted on patients undergoing small-incision phacoemulsification. Patients were categorised based on the presence of ocular and systemic comorbidities into four groups: those with only ocular diseases, those with only systemic diseases, those with neither ocular nor systemic diseases, and those with both ocular and systemic comorbidities. For evaluation of postoperative recovery and the impact of comorbidities, we analysed postoperative visual acuity and the recovery of visual acuity after the surgery. Results: Our results indicate that ocular comorbidities are associated with slower early postoperative visual recovery, while systemic comorbidities alone did not have a significant impact on visual acuity values and recovery. However, the differences between comorbidity groups decreased over time following surgery and postoperative visual acuity outcomes became comparable across groups. This suggests that regardless of the initial disease burden, the long-term effectiveness of cataract surgery has benefit and leads to improvement in visual function. Conclusions: Patients with ocular comorbidities, because of the slower postoperative recovery based on visual acuity, should receive closer monitoring in the early postoperative period. These findings can help in the optimisation of the postoperative follow-up protocols and reduce the healthcare burden associated with cataract management. Full article

Concrete-filled steel tube (CFST) structures are widely used in modern infrastructure due to their superior strength, ductility, and composite action. However, long-term corrosion of the steel tube, particularly under aggressive environmental conditions, poses significant challenges to their durability and structural performance. This study presents a comprehensive review of corrosion mechanisms and mitigation strategies for CFST structures. The primary corrosion processes, including general corrosion, localized (pitting) corrosion, and circumferential corrosion, are critically examined with emphasis on the influence of chloride ingress, carbonation, marine exposure, and combined environmental actions such as freeze–thaw cycles and sustained loading. The effects of corrosion on structural behavior are analyzed in terms of load-carrying capacity, ductility, buckling resistance, and failure modes. A systematic evaluation of existing mitigation strategies is conducted, encompassing material-based approaches, protective coatings, cathodic protection systems, and structural strengthening techniques such as fiber-reinforced polymer (FRP), fabric-reinforced cementitious matrix (FRCM), and steel jacketing. The comparative performance of these methods is assessed based on effectiveness, cost–benefit considerations, service life extension, and practical implement ability. The review highlights that no single mitigation strategy is universally optimal; instead, integrated approaches combining multiple techniques provide the most effective long-term protection. Key research gaps are identified in the areas of long-term performance monitoring, internal corrosion detection, and durability modeling under combined environmental actions. The findings of this study provide valuable insights for the design, maintenance, and rehabilitation of CFST structures, contributing to the development of more durable and sustainable infrastructure systems. Full article

Low-aspect-ratio flying-wing unmanned aerial vehicles (UAVs) are attractive drone platforms for civilian remote sensing, environmental monitoring, infrastructure inspection, disaster assessment, and persistent public-service monitoring because their integrated tailless layout offers high aerodynamic efficiency and payload volume. A trajectory-command-based three-loop nonlinear dynamic inversion (NDI) control architecture enhanced by a nonlinear disturbance observer (NDO) is designed to address the critical challenges of rapid time variation, strong nonlinearity, strong coupling, and restricted yaw authority in low-aspect-ratio flying-wing UAVs. The core innovation lies in the development of a trajectory-command-to-attitude kinematic mapping mechanism, integrated with the NDO for active torque compensation of lumped uncertainties and time-varying external disturbances. Leveraging a mathematical model of a low-aspect-ratio flying-wing UAV standard model, a three-loop NDI controller comprising angular rate, attitude, and trajectory command loops was designed based on the time-scale separation principle. The NDO was further designed to estimate lumped disturbances and provide feedforward compensation, thereby establishing an NDI-DO system that mitigates the high sensitivity of conventional NDI to modeling inaccuracies. Simulation and robustness tests involving typical high-angle-of-attack maneuvers (e.g., Cobra and Split-S maneuvers) demonstrated that the NDI-DO system achieved a reduction in angular-rate tracking error by over 77.2% compared to the baseline NDI. Furthermore, the permissible range of aerodynamic parameter perturbations was improved by 23%, significantly enhancing tracking fidelity and disturbance rejection. In a 3-DOF wind tunnel free-flight test, the NDI-DO system achieved a substantial expansion of the controllable angle-of-attack (attitude-stability) envelope from 72.9° to 99.19°, substantiating the high reliability and engineering utility of the control framework in post-stall nonlinear regimes. These results indicate that the proposed NDI-DO framework can support safer envelope expansion, autonomous upset recovery, and robust flight control for civilian flying-wing drones operating under uncertain aerodynamic and environmental conditions. Full article

Distributed fiber optic sensing (DFOS) has emerged as a transformative technology for structural health monitoring (SHM) of railway infrastructure, offering continuous, high-resolution measurements along extended optical fiber lengths, capabilities that conventional point sensors such as strain gauges and accelerometers cannot match. This review critically examines DFOS technology and its railway SHM applications, covering system components, interrogator units, optical fiber cables, and data acquisition systems, alongside the three principal scattering mechanisms: Rayleigh, Brillouin, and Raman, each offering distinct trade-offs in spatial resolution, sensing range, and measurand sensitivity. Field applications across track and sleeper monitoring, bridge health evaluation, tunnel lining assessment, and embankment stability are reviewed and critically compared. The integration of artificial intelligence (AI) and machine learning (ML) with DFOS data streams is discussed, demonstrating detection accuracy exceeding 97% in recent studies. Its main application rail embankment monitoring is discussed. Key challenges are identified, including high interrogator costs, large data volumes, installation complexity in retrofit scenarios, and environmental noise under operational train speeds. Future research priorities include lower-cost interrogation hardware, automated signal processing pipelines, digital twin integration, and standardized performance frameworks to accelerate large-scale adoption across railway networks worldwide. Full article

Background/Objectives: Perinatal asphyxia (PA) is a major cause of neonatal mortality and morbidity both in the short and in the long term. The identification of novel reliable biomarkers is essential in order to improve early diagnosis and allow for accurate prognostication of short- and long-term outcomes. The aim of the current study was to identify serum metabolites substantially affected by PA and resuscitation, using an experimental model in neonate piglets. Methods: A prospective, randomized experimental pilot animal study was conducted in 33 neonate Landrace/Large White female piglets, 1–4 days old. Following initial preparation and stabilization, the animals were allocated to three groups. Group A served as the control group while Group B and Group C piglets underwent asphyxia until severe bradycardia or hypotension occurred. Group C animals were subsequently resuscitated, and after return of spontaneous circulation (ROSC), they were stabilized and remained under further monitoring for 30 min. Blood samples for metabolic profiling were obtained at predefined timepoints as defined below. “Baseline” samples were taken from all animals after the initial stabilization; “asphyxia” sampling was performed at the time of hemodynamic compromise, while “final” sampling was performed 1 h after baseline in Group A animals and 30 min post-ROSC in Group C animals. The serum samples obtained were further analyzed using nuclear magnetic resonance (NMR) spectroscopy. Results: Distinct metabolic phenotypes were observed between the “baseline” state and asphyxia. Post-resuscitation and post-ROSC, the metabolic phenotype appeared to partially shift back to the “baseline” cluster but remained distinct from both of the other groups. The results were further processed using a structured biomarker discovery pipeline. Key metabolites that were found to significantly differentiate “baseline” and “asphyxia” states were lactate, succinate, lysine, fumarate, hypoxanthine and isoleucine (decrease) (p < 0.001). As far as the “baseline” against stabilization post-ROSC comparison is concerned, lactate, lysine, fumarate, hypoxanthine, succinate, acetate, alanine, glutamine, glutamate and choline differed significantly (p < 0.001). No metabolite survived False Discovery Rate correction and reached statistical significance in the direct “Asphyxia” versus “Resuscitation” comparison. Conclusions: This pilot study demonstrates that severe asphyxia in neonatal piglets is associated with a distinct serum metabolic signature, and several abnormalities remain detectable 30 min after ROSC, suggesting incomplete early metabolic recovery. The findings support the candidacy of lactate, succinate, fumarate, hypoxanthine and related metabolites for further assessment and validation as markers of acute hypoxic injury. Further investigation focused on these metabolites could also contribute to the elucidation of the involved pathophysiological mechanisms of PA and the development of novel therapeutic approaches. Full article

Background: Artificial intelligence (AI) is increasingly embedded in critical care nursing through monitoring, decision support, and documentation systems, yet nurses’ readiness to use it remains uncertain, particularly in the Middle East region. Critical care nurses are central to AI implementation at the bedside, and their AI-related literacy and fears can influence safe and ethical integration into clinical practice. Aim: To assess AI-related literacy and fears among critical care nurses in Oman and the associated factors. Methods: A nationwide cross-sectional survey was conducted among critical care nurses (N = 477) working in tertiary hospitals in Oman. The Multidimensional Artificial Intelligence Literacy Scale and the Fear towards AI Scale were used to measure AI literacy and fears, respectively. Results: The participants had low overall AI literacy (146.62 ± 84.03), and low AI self-efficacy and AI self-competency. The lowest level of literacy was related to creating AI (2.43 ± 2.63). On the other hand, participants reported moderate overall fear towards AI and moderate levels of fear related to job issues and humanity and ethics. Age, marital status, levels of education, receipt of prior computer or information technology, AI-related training, and work experience, were significant predictors of AI literacy. The predictors of AI self-efficacy and AI competence are presented. Conclusions: Nurses working in critical care settings in Oman reported low levels of AI literacy, but moderate fears related to AI, and this provides an opportunity to build AI competencies and capacity. There is need for deliberate and structured continuing education programs to build capacity for AI utilization, competence, and self-efficacy among critical care nurses. Structured, hands-on AI training that integrates ethical reflection for older nurses with more experience but limited professional education is needed and essential to support safe and equitable AI utilization by critical care nurses in Oman. Full article