Search Results (1,426)

Air pollution negatively impacts human health and environmental sustainability, particularly in areas with high industrial activity. This study comparatively evaluated deep learning-based models for estimating PM₁₀ and SO₂ pollutants in Dilovası and Ereğli (Turkey), industrial areas with high pollutant loads. The study utilized Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM), Gated Recurrent Units (GRUs), an RNN–GRU stacked hybrid model, an attention-based hybrid model, and the proposed GRU–LSTM dual-path hybrid model. The proposed method consists of four main stages: data conversion into a time-series format, data preprocessing and feature generation, model architecture development, and model training and performance evaluation. The dataset consisted of 365 daily PM₁₀ and SO₂ observations obtained from the Air Monitoring Center for the Dilovası and Ereğli monitoring stations. Model performance was evaluated using the coefficient of determination (R²), training time, root mean squared error (RMSE), mean squared error (MSE), and mean absolute error (MAE) metrics. The findings showed that the hybrid models provided higher accuracy compared to the single-track models. Specifically, the proposed GRU–LSTM dual-path hybrid model produced the highest R² and lowest error values for both pollutant parameters in both the Dilovası and Ereğli regions. In Dilovası, this model achieved R² = 0.97 for SO₂ and R² = 0.96 for PM₁₀; in Ereğli, it reached R² = 0.92 for SO₂ and R² = 0.98 for PM₁₀. Thus, it has been shown that the GRU–LSTM dual-path hybrid model, which models short-term and long-term temporal dependencies in parallel, is an effective and reliable method for air pollutant forecasting in industrial areas. These findings demonstrate the potential of the proposed model to support air quality monitoring, early warning systems, and environmental decision-making in industrial regions. Full article

Indoor air quality has a significant impact on occupant health, comfort, and productivity in residential and commercial indoor environments. This paper proposes an IoT-edge enabled deep–fuzzy hybrid framework for real-time IAQ prediction and adaptive control. The proposed system integrates IoT-based environmental sensing, Temporal Fusion Transformer-based multivariate forecasting, knowledge distillation, edge-deployed Bi-LSTM inference, and Mamdani fuzzy logic control within a unified IAQ management architecture. A composite Comfort Risk Index is introduced to combine environmental parameters and occupant discomfort feedback into a single adaptive control indicator. Experimental evaluation under varying indoor conditions demonstrated strong forecasting performance, with prediction accuracies reaching 96.3% for CO₂ and 95.7% for PM_2.5 prediction, while reducing inference latency from 575 ms to 295 ms. Comparative analysis against baseline threshold-based control strategies further indicated improved comfort stability, smoother actuator behavior, and reduced estimated actuator operating intensity during deployment. The proposed framework also demonstrated resilient operation under simulated sensor-failure conditions while maintaining low computational overhead suitable for resource-constrained IoT-edge environments. Overall, the results indicate that combining lightweight deep learning models with interpretable fuzzy control can provide an effective, scalable, and energy-aware solution for intelligent real-time IAQ optimization in smart indoor environments. Full article

This narrative review examines how big data analytics supports geospatial decision-making in smart cities through the combined roles of spatial data foundations, GeoAI methods, and urban digital twins. Methodologically, the article follows a structured narrative and critical review design rather than a PRISMA-based systematic review, bibliometric analysis, or meta-analysis. The paper responds to fragmentation across GIScience, smart-city studies, urban analytics, geospatial data engineering, and digital twin research, where related contributions often remain technically rich but weakly integrated from a decision-oriented perspective. Rather than treating geospatial decision-making as an extension of GIS or as a general expression of data-driven governance, the review frames it as a layered socio-technical process through which heterogeneous urban data are transformed into decision-relevant knowledge. The analysis first clarifies the conceptual evolution from GIS to spatial decision support and urban governance, and then examines the spatial data sources, integration problems, and representational limits that shape smart-city evidence. It also reviews GeoAI and geospatial analytics methods, including spatial statistics, machine learning, spatiotemporal forecasting, graph-based modeling, optimization, and explainable GeoAI. Urban digital twins are then analyzed as decision infrastructures that connect sensing, data integration, synchronization, semantic modeling, simulation, visualization, user interaction, and feedback into planning or operations. The review further maps these capabilities across mobility, land use, utilities, risk management, environmental resilience, public health, and cross-domain decision contexts. Overall, the paper argues that the value of smart-city geoinformation systems depends not on data abundance or model sophistication alone, but on their capacity to support interpretable, accountable, and context-sensitive urban decisions. Full article

Urban air pollution, especially the problem of PM_2.5, is one of the major health challenges facing the planet today. To provide accurate PM_2.5 predictions despite data noise and missing data, the authors proposed a deep learning model. We constructed a Stacked Autoencoder–Convolutional Neural Network–Bidirectional Long Short-Term Memory–Long Short-Term Memory (SAE-CNN-BiLSTM-LSTM) model that (1) utilises convolutional layers to extract spatial features from the input data, (2) employs bidirectional LSTM layers to capture long-term temporal dependencies, and (3) utilises an autoencoder to learn latent representations of the data to mitigate the effects of missing data. The model was trained on a large dataset of hourly measurements of air quality and meteorological parameters collected between 2018 and 2020 in Klang, Malaysia. The performance of the model on data that were not used during training was evaluated using a range of metrics. The SAE-CNN-BiLSTM-LSTM model achieved a test RMSE of approximately 11.97 µg/m³ and an R² statistic of approximately 0.85 for PM_2.5 concentrations, outperforming the other models tested on the same datasets. The additional metrics of MAE, MAPE, Mean Bias Error, and Index of Agreement confirmed the model’s accuracy and low bias in the prediction of air pollution levels. Statistical tests, such as the Diebold–Mariano test, confirmed the significance of the model’s accuracy over the CNN-LSTM models. These findings indicate that the proposed model effectively captures the dynamics of the air pollution data. The proposed model structure efficiently achieved an accurate and lightweight model for urban air pollution forecasting. Full article

Modern inpatient care generates irregular streams of heterogeneous clinical events, yet most predictive models require fixed feature matrices, predefined time windows, or discretization of continuous measurements. We developed CHaRT, a decoder-only autoregressive transformer designed to jointly forecast the identity of the next clinical event and, when applicable, its associated continuous value. CHaRT was trained and internally validated on structured electronic health record data from adult acute-care encounters across a 12-hospital health system in Minnesota from 2001 to 2025. The final corpus included 4,447,625 encounters from 1,301,502 patients and 701,556,877 non-padding clinical event tokens spanning vital signs, laboratory values, medications, diagnoses, microbiology, virology, imaging, fluids, and outcomes (ICU transfer or death). Encounters were split into training, validation, and test sets before vocabulary construction, normalization, and windowing. On the held-out test set, CHaRT achieved Top-1, Top-5, and Top-10 next-event accuracies of 51.61%, 87.34%, and 93.22%, respectively, with perplexity 4.50 and expected calibration error 0.0109. For numeric prediction, z-score MSE was 0.3812 for vital signs and 0.5713 for laboratory values. Seeded examples generated clinically coherent trajectories. Using model representations, a linear probe predicted deterioration (ICU transfer or in-hospital death) at a 6 h landmark with AUROC 0.95–0.97, indicating that learned representations transfer to downstream clinical risk prediction. Full article

Power systems continuously generate large-scale load time series data for forecasting, consumption analysis, and equipment health monitoring. However, real-world load measurements are often contaminated by anomalies caused by sensor faults, communication errors, and abnormal consumption behaviors, which may degrade data quality and affect operational decision-making. To address this issue, this paper proposes an integrated anomaly detection framework named PEL, which combines Prophet-based seasonal-trend decomposition, ensemble empirical mode decomposition (EEMD), and a multilayer long short-term memory (LSTM) network. Prophet is first employed to decompose the original series into trend, seasonal, holiday, and residual components. Sample entropy analysis and white noise tests are then adopted to evaluate whether the residual component still contains complex structured information requiring secondary decomposition. Next, EEMD is applied to the residual component to extract multi-scale intrinsic mode functions. Finally, all decomposed components are normalized and fed into a multilayer LSTM model for anomaly detection. Experiments on a real-world power load dataset demonstrate that the proposed PEL framework achieves an accuracy of 99.92%, a precision of 97.33%, a recall of 100%, an F1-score of 98.65%, and an AUC of 0.9996, outperforming or matching several baseline and hybrid models. Full article

Background/Objectives: The trajectory of influenza hospitalization burden from pre-COVID-19 pandemic baseline through post-pandemic recovery remains poorly characterized at the national level. This study characterized phase-stratified burden and seasonal structure, quantified racial and ethnic disparities, and assessed whether post-pandemic seasons represent anomalous departures from pre-pandemic expectations. Methods: Sixteen complete seasons of FluSurv-NET surveillance data (2009–2010 through 2024–2025; 509 observation weeks) were analyzed across pre-pandemic, disruption, and recovery phases using OLS regression with effect-size estimation, bootstrapped age-adjusted rate ratios, seasonal-trend decomposition (STL), Prophet time-series forecasting, and Isolation Forest anomaly detection. Results: Mean peak weekly hospitalization rate nearly doubled from pre-pandemic to recovery (5.1 to 11.1 per 100,000), cumulative seasonal burden increased from 46.3 to 87.0 per 100,000, and median peak timing advanced from MMWR week 9 to week 50. STL decomposition revealed a marked shift from weak pre-pandemic seasonality (Fs = 0.14) to substantially stronger annual regularity (Fs = 0.98) across three recovery seasons, with threefold amplitude increase. Non-Hispanic Black persons had rate ratios of 1.72, 2.16, and 1.99 relative to White persons across phases; American Indian and Alaska Native persons showed the highest disruption-phase ratio (2.24, 95% CI 1.90–3.53), based on two contributing seasons. A flat-growth Prophet model detected first exceedance in February 2020, outperforming a linear-growth specification on held-out validation. Isolation Forest identified 2017–2018, 2023–2024, and 2024–2025 as robust anomalies across all contamination thresholds. Conclusions: Post-COVID-19 pandemic influenza recovery is characterized by intensified and restructured seasonality, persistent racial and ethnic disparities, and anomalous burden exceeding pre-pandemic projections, identified independently by time-series forecasting and unsupervised anomaly detection. Full article

Accurate air pollutant concentration prediction is crucial for public health and sustainable urban development. Existing methods predominantly rely on single-modal data, resulting in inadequate representation of pollutant spatiotemporal evolution, poor prediction accuracy, and limited generalization capabilities. To address these challenges, this research proposes a novel PM_2.5 prediction framework termed MFD-DF that integrates ground-station time series and satellite remote sensing images. In feature extraction, learnable decomposition and deformable convolution are introduced, and a Cross-Modal Slot Attention module explicitly decomposes features to resolve information blurring. Subsequently, a dynamic cross-modal alignment mechanism is designed alongside a learnable Time-Expansion Network (TEN) to ensure fine-grained interaction. Furthermore, a local-global attention feature fusion mechanism is proposed to optimize data integration efficacy. Experimental results demonstrate that in single-step PM_2.5 prediction tasks, the proposed MFD-DF achieves significant improvements of approximately 10–20% in MAE, RMSE, and MAPE compared to state-of-the-art baselines. In multi-step PM_2.5 prediction, it effectively alleviates the error accumulation problem in long-sequence forecasting, demonstrating superior robustness and accuracy. Full article

Air quality forecasting and environmental health research at urban and regional scales depend on the combination of measurements from heterogeneous sensor networks, yet the construction of integrated multi-source datasets is rarely described or released as a self-contained deliverable. This paper presents an open dataset that combines four sensor-derived sources covering the whole of Spain over the period from 2022 to 2024: hourly air quality observations from the 588 stations of the national network operated by the Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO), daily meteorological records from the Agencia Estatal de Meteorología (AEMET), daily mobility indicators derived from anonymised mobile telephony events published by the Ministerio de Transportes y Movilidad Sostenible (MITMA) at the municipality level, and a calendar of national and Autonomous Community public holidays. The processing pipeline harmonises sources that differ in temporal resolution, spatial codification and quality regime into a tidy hourly table indexed by station and timestamp, with a fixed feature schema of 56 variables per record. Air quality stations are paired with their nearest AEMET station through a three-tier distance rule, and the daily exogenous features are aligned to the air quality time axis through a two-variant temporal-alignment scheme (lag-and-expand to the hourly grid for the hourly release, same-calendar-day join for the daily release). A complementary daily resolution variant of the dataset is also released, with 72 columns and the same feature schema except for the air quality block, which is aggregated to daily mean, minimum and maximum. The integrated dataset contains approximately 15 million hourly records across the 588 stations and is released on Zenodo (DOI 10.5281/zenodo.20196221) under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence. It is intended as a substrate for research on air quality forecasting, environmental epidemiology and multi-source data fusion at the nationwide scale. Full article

Dengue Virus (DENV) circulates as four antigenically distinct serotypes whose dominance fluctuates over time in many endemic regions, a phenomenon known as serotype shift that is frequently associated with large outbreaks and increased disease severity. This review, through a synthesis of epidemiological, virological, immunological, entomological, and environmental evidence, observes that serotype shift likely arises from the interaction of multiple determinants rather than solely from viral evolution, with population immunity playing a central role. The accumulation of serotype-specific herd immunity, together with short-lived cross-protection and Antibody-Dependent Enhancement (ADE), reshapes population susceptibility and creates ecological space for heterologous serotypes with higher transmission potential. The synthesis of global dengue studies indicates that these immune dynamics interact with viral genetic diversity, vector competence, climate variability, and human factors such as demography, socioeconomic status, population density and mobility to drive cyclical and sometimes abrupt changes in serotype dominance. Notably, the review indicates that serotype changes often precede or coincide with more clinical severity and patterns of outbreaks, with direct implications for the process of forecasting outbreaks, vaccine performance, and preparedness to respond with appropriate health measures. On the whole, this review confirms the opinion that the change of dengue serotype occurrence becomes a consequence of interconnected biological and ecological processes involved in the transmission of dengue serotype shifts in hyperendemic areas. Full article

Background: Pancreatic cancer is characterized by increasing incidence and high mortality in Poland and worldwide. The aim of this study was to assess the relationship between selected risk factors and the age-standardized incidence rate of pancreatic cancer at the voivodeship level in Poland, and to evaluate the accuracy of a prediction model. Methods: Age-standardized incidence rate data for 16 Polish voivodeships in 2011–2023 were obtained from the Polish National Cancer Registry. The risk factor burden for 2011–2019, expressed as disability-adjusted life years (DALYs) per 100,000 population, was obtained from the System Analysis and Implementation Database of the Polish Ministry of Health. A generalized estimating equation model was constructed to predict the age-standardized incidence rate, with multicollinearity addressed using variance inflation factor analysis. Predictions for 2020–2023 were validated against observed data, and forecasts for 2024–2030 were subsequently calculated. Results: The number of new pancreatic cancer cases in Poland increased in eight out of 16 voivodeships. The highest burden was recorded in the Masovian, Subcarpathian, Świętokrzyskie and Greater Poland voivodeships. Air pollution was positively associated with pancreatic cancer incidence. Predictions for 2020–2023 showed satisfactory agreement with observed data, with the largest discrepancy being equal to 4.1 in terms of the age-standardized incidence rate. Based on the models, the incidence of pancreatic cancer was projected for all of 16 voivodeships through to 2030. Conclusions: Air pollution is associated with the regional burden of pancreatic cancer in Poland. The generalized estimating equation prediction approach demonstrated acceptable accuracy and can support monitoring and public health planning at the voivodeship level. Full article

To address the challenges of complex multi-channel signal coupling and insufficient long-term temporal dependency characterization in remaining useful life (RUL) prediction of rotating machinery, this paper proposes a multivariate time series forecasting framework integrating multi-channel information fusion and a self-attention gated augmentation unit (SGAU). First, a multilayer perceptron (MLP) explicitly models nonlinear coupling among channels; SGAU replaces the conventional feed-forward network in the Transformer encoder, using multi-head self-attention outputs as gating signals to adaptively regulate feature transformation. Second, multi-channel signals are predicted via this framework; high-dimensional feature vectors are extracted to construct multi-channel Gaussian mixture models (GMMs). Third, Jensen–Shannon divergence (JSD) quantifies deviations between the target and initial data clusters; centroid distance evolutionary trajectory is fused with JSD to construct the health indicator (HI). Continuous HI predictions yield the RUL prediction curve. Experiments on a self-designed wind turbine gearbox platform and the XJTU-SY bearing dataset demonstrate that the proposed framework outperforms baseline methods on Mean Square (MS), Root Mean Square (RMS), and Energy metrics, with average error reductions of 6.6% and 12.1% in the horizontal and vertical directions on the gearbox dataset and 20.9% and 32.3% on the bearing dataset, confirming its effectiveness and generalization capability. Full article

COVID-19, caused by the SARS-CoV-2 virus, has become a global public health crisis with diverse clinical manifestations affecting multiple organ systems, including the integumentary system. One notable cutaneous manifestation, referred to as “COVID toes,” involves the development of pernio-like chilblains, characterized by red-to-violet macules, plaques, or nodules, primarily on toes and fingers. This characteristic clinical feature gained significant attention due to its apparent association with COVID-19, especially during the early stages of the pandemic when individuals with mild or asymptomatic cases exhibited these symptoms. Concurrently, digital platforms such as Google Trends have emerged as tools for tracking public interest in health-related topics, offering insights into real-time patterns of disease awareness. Previous research has demonstrated that Google Trends data may correlate with the incidence of infectious diseases, suggesting that search interest can be a proxy for disease outbreaks. In this study, we sought to explore the potential relationship between public interest in COVID toes, as reflected in Google Trends, and the incidence and mortality rates of COVID-19. Specifically, we examined whether peaks in search interest for “COVID toes” corresponded with surges in COVID-19 cases and deaths. By analyzing trends in search data, we aimed to assess the utility of digital platforms as an epidemiological tool for monitoring disease progression and public awareness. Our findings provide insights into the potential role of digital search data in forecasting outbreaks and highlight the interplay between public perception and the clinical burden of COVID-19, emphasizing the importance of real-time data in public health surveillance and response. Full article

Intensifying heatwaves threaten the reliability of electric distribution systems, yet the quantitative relationship between heatwave characteristics and observed power outage behavior remains poorly understood at multi-year, statewide scales. This study develops an event-based, spatiotemporal framework to quantify heatwave-induced outage risk across 254 Texas counties from 2014–2021 by integrating county-level EAGLE-I outage records with reanalysis-derived heat index measurements. An adaptive percentile-based threshold identifies 3048 heatwave events; logistic regression quantifies the probabilistic relationship between heat intensity and major-outage occurrence under three severity definitions. Across 3048 identified heatwave events, 51% involved at least one outage, a rate significantly above the non-heatwave warm-season baseline and revealing widespread heat-related reliability challenges. Outage severity and duration exhibit heavy-tailed distributions, with a small number of extreme events disproportionately affecting customers. Logistic regression models under three severity definitions (P90, P95, and ≥500 customers) demonstrate that heat intensity is a statistically robust probabilistic predictor of major outages, with each +1 °F increase in mean event heat index raising the odds by approximately 43–52%. The predicted probability of a P90-severity major outage approximately doubles across the interquartile range of event heat intensity (~7% to ~14%), providing actionable guidance for utility pre-staging decisions during forecast heatwave episodes. These findings offer a scalable methodology for climate-related reliability assessment, supporting grid hardening, resource planning, and public health preparedness. Full article

Background/Objectives: Melanoma is a major and growing public health concern in Poland, with a five-year survival around 60–70%. While UV radiation and genetic susceptibility are well-known risk factors, lifestyle and environmental exposures may also contribute. This study examined how selected risk factors relate to one-year melanoma prevalence across Poland’s 16 voivodeships and assessed whether these factors can support short-term prediction. Methods: Annual melanoma prevalence for 2011–2021 was obtained from the Polish National Cancer Registry, and voivodeship-level estimates of metabolic risk factors, physical inactivity, alcohol consumption, smoking, high BMI, air pollution, water pollution and limited data on UV exposure were used to build a general estimating equations model. Model predictions for 2020–2021 were compared with observed data, and forecasts were generated through 2030. Results: Melanoma cases increased in every voivodeship between 2011 and 2021. Metabolic risk factors, high BMI, low physical activity and smoking were associated with higher melanoma prevalence. When other factors were considered, air pollution showed an inverse association, suggesting complex relationships that warrant further analysis. Forecasts indicated increasing prevalence in all of 16 voivodeships through 2030, although three regions showed large prediction errors for 2020–2021. A key limitation was the lack of sufficient UV exposure data. Conclusions: The findings support further evaluation of public health actions targeting the reduction of unhealthy lifestyle regarding diet, low physical activity, and smoking to help slow the projected rise in melanoma. Full article