Search Results (476)

Coffee is widely consumed by physically active individuals and athletes, often with the intention of enhancing alertness, reducing fatigue and supporting exercise performance. Its performance-related effects are primarily attributed to caffeine; however, responses to coffee consumption vary substantially across individuals and type of training. This narrative review examines current evidence on the role of coffee consumption in sport, with particular emphasis on caffeine-related mechanisms, physical performance, fatigue, muscle pain and recovery, as well as factors modifying individual responses, including circadian rhythm, habitual intake and genetic variability. Available evidence indicates that coffee consumption prior to physical activity may support endurance performance, reduce perceived exertion and attenuate muscle pain, particularly during prolonged or demanding exercise and under conditions of sleep restriction or circadian misalignment. In contrast, effects on muscle strength and power performance appear less consistent and more context dependent. Interindividual differences in caffeine metabolism, sensitivity to stimulation and susceptibility to sleep disruption further influence both the effectiveness and tolerability of coffee intake. Emerging findings also suggest that coffee, when consumed in combination with carbohydrates, may enhance post-exercise muscle glycogen resynthesis, a potential benefit of particular relevance for endurance athletes facing limited recovery time. At the same time, evidence regarding non-caffeine bioactive compounds present in coffee remains limited and largely indirect, indicating a modulatory rather than ergogenic role. Overall, coffee should not be regarded as a universally effective ergogenic aid, as its effects may vary according to individual characteristics, training demands and the timing of consumption. Its effective use requires careful consideration of individual characteristics, training demands and timing of consumption, underscoring the importance of personalized strategies in both performance and recovery settings. Full article

Cardiovascular diseases remain the leading global cause of mortality, necessitating continuous monitoring solutions that extend beyond clinical settings. This paper proposes a real-time, end-to-end Edge-IoT framework for cardiovascular risk assessment that integrates biomedical signal acquisition, edge processing, and interpretable deep learning. The system includes a three-tier architecture: (i) physiological signal acquisition using AD8232 ECG, MAX30102 photoplethysmography, DS18B20 temperature, and NEO-6M GPS sensors interfaced with an ESP32 microcontroller; (ii) real-time signal preprocessing, including digital filtering, normalisation, and PQRST feature extraction performed at the edge; and (iii) cloud-based analytics using an Enhanced TabNet classifier with modified attention mechanisms for cardiovascular risk prediction. The Enhanced TabNet architecture incorporates Entmax-based sparse attention and modified Gated Linear Units to improve predictive performance and clinical interpretability. Signal quality enhancement using Kalman filtering and class imbalance correction using SMOTE further support robust model performance. The Enhanced TabNet model achieves 97.43% accuracy, 96.18% precision, and 97.24% recall on the combined Cleveland, Hungarian, Switzerland, Long Beach VA, and Statlog heart disease datasets ($n=1190$). The developed Edge-IoT prototype maintains an end-to-end communication and processing latency below 200 ms. The framework also includes automated risk alert generation via SMS when the predicted cardiovascular risk probability exceeds a predefined threshold (e.g., 0.85), including the patient’s vital information and geolocation to support emergency response. The integrated edge-cloud architecture with attention-based feature selection provides interpretable cardiovascular risk predictions while maintaining the computational efficiency required for potential continuous patient monitoring outside hospital settings. Full article

It is becoming harder to feel at home. As wildfires, floods, drought, and heat become recurrent rather than exceptional, and as everyday life fills with weather alerts, evacuation warnings, and preparedness drills, many people experience a particular kind of distress: the pain of remaining in a place that is still physically present but no longer feels safe, familiar, or trustworthy. This paper examines that experience through the concept of solastalgia—homesickness felt while still at home—and argues that it is not a private mood or a distortion of thinking, but a meaningful disturbance in the very conditions of dwelling. Bringing together phenomenology, emotional geography, and emotion-regulation theory, it shows how a pervasive “culture of emergency” reshapes the way we inhabit space, time, our bodies, and our relationships with others, narrowing the room we feel we have to live and act. Rather than treating climate-related anxiety as something to be corrected, the paper understands it as an honest response to a world whose stability has genuinely been compromised. It then asks how home might become livable again, arguing that repair must happen at several levels at once—personal, communal, and institutional—and that a workable sense of belonging can be rebuilt without denying what has been lost. Full article

Background and Objectives: Decision-making capacity is essential for informed consent, yet its assessment in emergency departments (EDs) is often subjective and inconsistently documented. Older adults are particularly vulnerable to impaired capacity during acute illness. We aimed to develop a scalable, electronic health record (EHR)-based approach to support early identification of older ED patients at risk of decisional vulnerability using the Medical Information Mart for Intensive Care (MIMIC)-IV database. Methods: We conducted a retrospective cohort study of 51,195 ED patients aged ≥65 years. Clinicians manually reviewed 2000 patients using a conservative consensus protocol to establish a consensus-derived proxy for decisional vulnerability. Such an approach yielded a definitive reference subset (capacity vs. no capacity) and an “uncertain” category when consensus was not achieved. We developed a weak-supervision expectation–maximization (EM) label model that combined multiple noisy labeling functions derived from triage vital signs, acuity measures, arrival mode, and large language model (LLM)-classified chief complaints to estimate the probabilistic risk of impaired capacity. Model discrimination and calibration were assessed on an independent holdout subset of definitive reference labels using receiver operating characteristic area under the curve (ROC-AUC), precision–recall area under the curve (PR-AUC), calibration plots, and Brier score. To support clinically conservative use, operating thresholds were a priori constrained to limit automated flagging to ≤15% of patients, with the remaining ones deferred for clinician review. Results: On the definitive holdout set, the weak-supervision model achieved an ROC-AUC of approximately 0.855 and a PR-AUC of approximately 0.837. Calibration assessment demonstrated residual miscalibration in the generative posterior, which improved after a lightweight discriminative refinement step (logistic regression trained on EM-derived probabilistic labels), reducing the Brier score to approximately 0.224 on holdout evaluation. Under the prespecified operational constraint (≤15% auto-flagged), the model functioned as a conservative, selective alerting strategy, achieving high specificity and positive predictive value while identifying only a minority of patients with decisional vulnerability. Conclusions: This study demonstrates a methodological proof of concept for using weak supervision to model a retrospectively defined proxy for decisional vulnerability from routinely collected ED EHR data. The framework is intended to support conservative, triage-oriented prioritization. Further prospective validation, external testing, and workflow governance are needed before clinical implementation. Full article

The increasing deployment of lithium-ion batteries in electromobility, industrial logistics, and stationary energy storage systems has introduced new operational safety challenges associated with thermal runaway, fire development, and the release of hazardous substances. Although significant attention has been devoted to battery design and fire suppression technologies, less emphasis has been placed on the development of integrated monitoring systems capable of identifying abnormal thermal behaviour during routine storage and handling operations. This paper presents an operational framework for the early detection of thermal anomalies in lithium-ion battery storage facilities based on the integration of thermal imaging technology, multi-level alarm logic, automated notification processes, and predefined response procedures. The proposed framework was developed using a risk-based approach and implemented within an industrial environment where lithium-ion batteries and battery modules are routinely stored and handled. The methodology included hazard identification, determination of critical monitoring zones, configuration of thermal detection devices, establishment of alarm thresholds, and integration with existing fire protection infrastructure. Particular attention was devoted to ensuring rapid identification, localization, verification, and escalation of abnormal thermal conditions before the occurrence of visible fire manifestations. The implemented monitoring framework comprised a total of 21 thermal imaging cameras, including four fixed radiometric thermal imaging cameras and seventeen local thermal monitoring cameras, covering five risk-prioritized monitoring zones within an industrial lithium-ion battery storage facility. During operational deployment, the system recorded 21 Yellow Alerts, 6 Red Alerts, and 4 false alarms, with an average response time of 4.2 min. Experimental verification further demonstrated that, although directly exposed battery modules were measured at approximately 60 °C, enclosure within the battery pack significantly attenuated the externally detectable thermal signature, with surface temperatures decreasing to approximately 23–31 °C after prolonged enclosure. The results demonstrate that the proposed framework enables continuous operational monitoring, supports timely identification of abnormal thermal behaviour, and provides a structured basis for rapid decision-making and emergency response in industrial lithium-ion battery storage facilities. The integration of thermal monitoring with structured alarm management and response procedures creates a comprehensive safety chain that contributes to reducing the probability of delayed incident recognition. The presented approach provides practical guidance for industrial operators seeking to improve lithium-ion battery safety and may serve as a foundation for the future development of operational safety requirements for battery storage facilities. The principal contribution of this study is the documented implementation and operational evaluation of an integrated thermal monitoring and response system under routine automotive production conditions. Full article

Coastal marinas and high-traffic nearshore sites accumulate pollution from vessel movements, tourism, and shifting weather, yet routine monitoring rarely operates at the temporal resolution needed to catch emerging risks before they become acute. This study developed an AI-enabled decision support system and tested it across three European coastal sites over a 90-day window in summer 2025: an urban marina (Site A), a tourism marina (Site B), and a mixed-use port channel (Site C). A composite Water Quality Risk Index (WQRI), combining five normalised environmental and vessel-traffic stressor dimensions, fed a two-layer AI framework in which a gradient boosting model estimated short-term traffic-related stress and a random forest model classified next-day risk. Vessel traffic was heaviest at Site B, but water quality risk followed a different pattern: Site C returned the highest mean WQRI and logged the most hours under red alert despite intermediate traffic volumes, indicating that sustained moderate traffic mattered more than peak volume. Vessel intensity and WQRI were positively correlated at all three sites, most strongly at Site B, and the next-day random forest risk classifier, trained across all three sites, achieved strong discriminative performance (AUC 0.93). When the system indicated elevated risk, managers responded by deploying inspections, issuing traffic advisories and increasing monitoring activity. The pilot shows that connecting vessel tracking, environmental sensing, and ML-based classification into a single decision loop can move coastal pollution management from reactive to anticipatory. Full article

NASA’s Fire Information for Resource Management System (FIRMS) provides near-real-time thermal anomaly detections from VIIRS, but not all detections correspond to wildfire incidents: industrial heat, agricultural burning, and sensor artifacts produce false alarms that contribute to alert fatigue for emergency-management analysts. We study whether contextual machine learning (ML) features improve wildfire-incident classification from FIRMS detections, and—more importantly—whether reported gains survive leakage-controlled evaluation. We construct a labeled dataset by matching 521,395 VIIRS SNPP detections across CONUS in 2024 to 3766 NIFC 2024 wildfire perimeters, yielding 131,771 (25.3%) wildfire-matched and 389,624 candidate non-wildfire detections spanning 1067 distinct wildfire incidents. We benchmark five operational baselines and six classifiers under four validation regimes (random, event-aware, 5° spatial-block, and temporal holdout) with and without raw geographic coordinates. A naive random split inflates LightGBM to F1 = 0.985, but a leakage-controlled event-aware split reduces it to F1 = 0.767, and a spatial-block holdout to F1 = 0.627. Feature attribution shows geographic coordinates account for 88.9% of model gain—the summed share of LightGBM’s total split-gain attributed to the three coordinate features within the full-feature model; removing coordinates improves spatial-block generalization from F1 = 0.627 to 0.818, demonstrating that raw coordinates drive memorization of where 2024 fires occurred rather than transferable discrimination. We further show that spatiotemporal clustering must be causal: a model using full-partition clustering appears strong (F1 = 0.908) but leaks future detections, whereas a properly causal trailing-window version ties plain LightGBM in-distribution (F1 = 0.762). Combining causal clustering with no raw coordinates is the most robust configuration under spatial transfer (spatial-block F1 = 0.868 vs. 0.627 for the coordinate model). Bootstrap 95% confidence intervals show these gaps far exceed statistical uncertainty, and sensitivity analyses show the conclusions are robust to the spatial-block size and to the clustering-window choice. Under natural class prevalence (14%), precision falls to 0.69, and results are sensitive to the labeling buffer. All ML models nonetheless far exceed FIRMS high-confidence thresholding (F1 = 0.128). We argue that spatial leakage—not raw accuracy—is the central methodological issue for FIRMS wildfire-incident classification, and recommend coordinate-free, causal spatiotemporal-clustering features evaluated under spatial holdout. The system is intended as an analyst-prioritization decision-support layer, not autonomous incident confirmation. Full article

Background: Remote patient monitoring (RPM) for chronic heart failure (CHF) management has demonstrated clinical and economic benefits. However, data on its organisational impact remain limited, particularly in the French healthcare context following the integration of RPM into standard care pathways. The objective of this study was to describe the organisational impact of RPM for CHF management from the perspective of healthcare professionals across diverse practice settings in France. Methods: A cross-sectional survey was conducted among 28 French cardiology departments and community practices (14 CHG, 9 CHU, 4 community practices, 1 private clinic) actively using RPM for CHF management. The questionnaire assessed organisational changes across five domains: care process modifications, human resources allocation, professional training and task delegation, coordination with community-based physicians, and perceived impact on patient quality of life and working conditions. Results: Nearly all structures (96.4%) had established a dedicated RPM team (mean size: 4.8 ± 2.3 professionals). A reduction in time between cardiac decompensation and medical care was reported by 96.4% of respondents. Three-quarters (75.0%) had implemented specific consultations for emergency alerts, and 57.1% had established direct admission protocols bypassing emergency departments. Task delegation was widespread (89.3%), primarily involving alert monitoring (100%), patient inclusion (92.0%), and patient contact (84.0%). However, only 53.6% of professionals had received specific training. The perceived impact on patient quality of life was unanimously positive (64.3% very positive, 35.7% positive). Working conditions were perceived as stable by 64.3% and improved by 32.1%. The main advantages cited were reduction in hospitalisations (89.3%) and improved patient communication (85.7%). The primary area for improvement was interprofessional coordination (53.6%). Conclusions: RPM implementation has driven substantial organisational restructuring across French cardiology departments and community practices. While dedicated teams and task delegation are now standard, challenges remain in professional training, coordination with community physicians, and dedicated time allocation. These findings provide an updated organisational assessment that complements prior clinical and economic evaluations. Full article

Accurate wildfire prediction increasingly determines whether emergency resources arrive before a disaster becomes uncontrollable, yet the dominant paradigm reduces the problem to binary occurrence, offering no estimate of the severity that drives suppression planning. This study develops a machine-learning framework for four-class wildfire severity prediction, conditional on ignition, from weather-station observations and calendar terms alone. We construct a composite severity index (CSI) by applying principal component analysis to five damage dimensions (burned area, suppression equipment, personnel, duration, and property loss) recorded for 868 wildfires in Gangwon Province, South Korea (2011–2022) and pair standard observations with effective humidity and six indices of the Canadian Forest Fire Weather Index (FWI) System. Under a leakage-safe protocol, the strongest tree ensembles reach a macro F1 of 0.46 to 0.50 (recommended configuration: 0.41 ± 0.03 across 20 repeated splits) against a four-class chance level of 0.25, and the recommended Random Forest attains an extreme-class recall of 0.474; the CSI target outperforms burned area by 5.5 macro-F1 points under identical inputs. A weather-only screen separates extreme from non-extreme events with an ROC AUC of 0.758, capturing 47% of extreme events at a 20% alert budget. We also quantify how oversampling misplaced before the train-test split inflates the macro F1 to 0.65–0.83, a cause for caution for the severity-prediction literature. Full article

As emerging infectious diseases and human–wildlife interactions increasingly transcend political boundaries, the ability to detect patterns, compare data, and respond in a coordinated manner depends on the availability of interoperable, well-structured data systems. Wildlife health data are often collected by diverse stakeholders—including field biologists, wildlife rehabilitators, hunters, veterinarians, and diagnostic laboratories—using inconsistent methods, terminology, and levels of detail. Data stored in isolated systems results in duplication, information loss, and barriers to integration. This paper describes the design and development of the Shared Wildlife Health Information System (SWHIS), an online database system designed to facilitate wildlife health data management. Development incorporated continuous stakeholder feedback, including input from state wildlife management agencies, to ensure the system aligned with real-world workflows and user needs. Launched in October 2022, SWHIS is supported by a dedicated development team with a growing suite of tools to meet diverse user needs. Priority features under development include a mobile field application, a public reporting tool, and an event alerting system. By improving data standardization, accessibility, and long-term preservation, systems like the SWHIS strengthen wildlife health surveillance capacity and support timelier, evidence-based decision-making across jurisdictional boundaries. Full article

The safety of non-food products is embedded in the United Nations 2030 Agenda for Sustainable Development and the European Union (EU) framework, supporting health protection, responsible production and consumption, and market surveillance. The EU Rapid Alert System for dangerous non-food products, known as Safety Gate (formerly RAPEX), was established in 2005 to facilitate the exchange of information on products posing risks within the internal market. The aim of this study was to present the interdependencies reported in the Safety Gate system/RAPEX in 2005–2025, considering: product category, type of risk, country of origin, notifying country and year, as well as measures taken. The VOSviewer 1.6.20 and Statistica 13.3 were used. The results highlighted the following problems: toys from China with chemical, choking and injury risks; electrical appliances also from China with electric shock hazards; motor vehicles from Germany with injury risks; cosmetics from Italy with chemical and microbiological risks; and clothing from Turkey with suffocation risks. Reporting is expected to continue under existing regulatory frameworks, although changing the name of the system from RAPEX to “Safety Gate” may reduce its recognition. The findings highlight the need for targeted enforcement, improved risk profiling by product category and origin, and ongoing monitoring of emerging safety risks. Full article

Background and Objectives: Immediately after thyroidectomy, retching driven by postoperative nausea and vomiting (PONV) may cause wound rupture, severe bleeding, and airway obstruction. Although inhaled anesthetics are widely used in thyroidectomy, they may increase the postoperative risk of PONV. Therefore, this study aimed to compare PONV incidence and recovery patterns according to the characteristics of sevoflurane (Sevo) and desflurane (Des) when remifentanil was continuously infused until extubation. Materials and Methods: This retrospective cohort study involved 70 female patients undergoing elective thyroidectomy, who were categorized into the Sevo (n = 35) and Des (n = 35) groups. Remifentanil was administered at an effect-site concentration of 2 ng/mL during emergence. Results: PONV incidence during emergence was 20% in both groups (p = 1.000). The Des group had shorter times to recovery of consciousness and extubation than the Sevo group (p < 0.001 and p < 0.001, respectively). At 5 min after extubation, patients in the Des group were more alert (p = 0.001), with 54.3% awake and responsive. Postanesthesia care unit stay was also shorter in the Des group (16.89 ± 3.22 vs. 23.74 ± 4.80; p < 0.001). Additionally, perioperative hemodynamic status, surgical site pain, and residual sedation did not differ between inhaled anesthetics. Conclusions: When remifentanil was infused until extubation after thyroidectomy, the choice of inhaled anesthetics did not affect the incidence of acute PONV. Des provided faster early recovery without additional side effects than Sevo; nonetheless, acute recovery profiles did not differ between inhaled anesthetics. Full article

In Mexico, since June 2025, human cases of myiasis caused by Cochliomyia hominivorax (the screwworm) have been identified, prompting health authorities to issue a surveillance alert for potential infections by this fly. Here we report the identification and interdisciplinary management of a case of nasal myiasis caused by Lucilia sericata. In September 2025, a 40-year-old Hispanic man was admitted to the emergency room due to an altered mental status; after a 17-day hospital stay, nasal larvae were detected. The larvae were submitted for morphological identification by microscopy and molecular identification by polymerase chain reaction, using ITS-2 and 16S loci as nuclear and mitochondrial markers, respectively; the amplicons were purified and sequenced, and a Bayesian phylogenetic analysis was performed. Morphological analysis showed that they were L. sericata larvae at different developmental stages (1st to 3rd instars). Molecular analysis confirmed the morphological result, as phylogenetic inferences showed a clear grouping of the sequences within the L. sericata cluster. While remaining vigilant for possible human cases of myiasis caused by C. hominivorax, here, we confirmed by morphological and molecular analysis the identification of an unexpected case of nasal myiasis caused by L. sericata, demonstrating the importance of care in all aspects of thorough epidemiological surveillance and the interdisciplinary management to address these alerts and any related cases. Full article

This study examines how the relationship between perceived financial accounting disclosures (FAD) and investor reactions to negative financial news (IRNFN) is conditioned by two individual-level moderators among 310 retail investors holding shares in project-based organisations (PBOs) listed on the Dubai Financial Market and Abu Dhabi Securities Exchange. The two moderators are framing bias susceptibility, a cognitive predisposition to be influenced by presentational form, and AI channel reliance (AICR), the extent to which investors rely on AI-mediated information channels—including algorithmic news aggregators, robo-advisory tools, AI-curated social media feeds, and automated sentiment-scored financial alerts—for receiving and interpreting corporate disclosures. Drawing on Behavioural Finance Theory and the Theory of Planned Behaviour, the study investigates whether the strength of the FAD–IRNFN association depends on these cognitive and informational processing conditions. The measurement model was estimated using confirmatory factor analysis in AMOS 25, and the moderation hypotheses were tested through path analysis with mean-centred composite scores and bias-corrected bootstrap inference, with a latent interaction robustness check reported in parallel. AI channel reliance emerged as a substantial moderator of the FAD–IRNFN relationship, while framing bias provided a smaller, marginally significant moderating effect. The findings are consistent with the theoretical expectation that, in AI-mediated information environments, the perceived quality and presentation of complex disclosures are associated with stronger, rather than weaker, investor reactions to negative news. Because the design is cross-sectional and based on self-reported data, the results are interpreted as associations rather than causal effects, with implications for disclosure regulation, corporate communication, and AI platform design in the UAE and comparable emerging markets. Full article

Earthquake-induced landslides represent a critical threat to transportation infrastructure in tectonically active mountainous regions, particularly in tropical volcanic settings where weak, highly weathered geomaterials dominate. This study develops an integrated framework that directly links physically based seismic threshold modelling with real-time landslide monitoring and operational early warning. The approach is demonstrated in the Cugenang area of Cianjur Regency, West Java, Indonesia, which was severely impacted by the moment magnitude (Mw) 5.6 earthquake in 2022. Slopes composed of highly weathered pyroclastic deposits [Plasticity Index (PI) = 54–68%; porosity > 60%] exhibit low shear strength and high sensitivity to seismic loading. Limit equilibrium analysis using the Morgenstern–Price method that combines the influence of seismic loading and groundwater conditions suggests that a horizontal seismic coefficient (kh) of approximately 0.06, corresponding to a Peak Ground Acceleration (PGA) of about 0.12 gravitational acceleration (g), is a critical threshold for initial landsliding. This comparatively low threshold challenges commonly reported values and demonstrates that slope failure in tropical volcanic terrains can occur under moderate ground shaking, reinforcing the need for site-specific hazard characterisation. The derived thresholds are operationalised within a multi-sensor early warning system integrating Micro-Electro-Mechanical Systems (MEMS) accelerometers and inclinometer measurements. Three hazard levels—Normal (<0.06 g), Alert (0.06–0.12 g), and Emergency (≥0.12 g)are combined with deformation thresholds [<10 milimeter (mm), 10–30 mm, >30 mm] to capture progressive failure processes and minimise false alarms. By coupling geotechnical modelling and real-time monitoring, this study provides a transferable and scalable framework for enhancing infrastructure resilience in landslide-prone regions. Full article