Search Results (1,358)

The rapid growth of sensor networks, satellite platforms, Internet-of-Things (IoT) devices, and urban monitoring systems has produced vast volumes of spatiotemporal data. Modeling such data is central to applications ranging from air-quality forecasting and climate monitoring to traffic prediction and urban planning, yet it poses challenges, including high dimensionality, spatial autocorrelation, and complex temporal dependencies that classical statistical models struggle to capture jointly. This review provides a comprehensive survey of spatiotemporal machine learning (STML), synthesizing the literature published between 2020 and 2025. We present a taxonomy of STML methods spanning traditional machine learning, recurrent neural networks (LSTM and GRU), convolutional architectures (3D CNN and ConvLSTM), graph neural networks, transformer-based models, and hybrid or ensemble designs. We organize representative studies by application domain, climate and environmental monitoring, transportation and traffic forecasting, and urban planning within smart cities, and provide a comparative analysis of models, datasets, evaluation metrics, and open-source tools. Finally, we identify emerging research trends and open challenges, including real-time processing, interpretability and explainability, scalability and edge deployment, multimodal fusion, federated learning for privacy, self-supervised learning, and ethical concerns such as fairness. The review is intended to serve as a structured reference for researchers and practitioners developing STML systems. Full article

Background/Objectives: episodic memory refers to the ability to place events within a specific spatial and temporal context, making its association with navigation unsurprising. Incidental episodic memory involves encoding and retrieving information without explicit instruction, a common process in everyday life whose sensitivity to emotional valence warrants further investigation. Methods: sixty-three college students (mean age = 25 years; 28 females) completed two versions of the same task, differing only in the instructions given: one directed attention to emotionally charged images encountered along a route through a city environment; the other directed attention to the route itself. Results: Mann–Whitney tests on uncontaminated first-session conditions and linear mixed models all indicated an advantage of intentional over incidental encoding for the spatial localization of emotional images. The advantage was valence-selective, present for negative and neutral stimuli but absent for positive ones. Signal detection analyses indicated that intentional instructions selectively increased sensitivity (d′). The response criterion also shifted, but in opposite directions for negative and neutral stimuli, so the sensitivity gain cannot be attributed to a uniform change in response bias. The effect was domain-specific: sensitivity for the spatial localization of emotional images increased, whereas sensitivity to navigational route did not improve. Conclusions: positive material thus appears to be encoded spontaneously, with potential implications for rehabilitation and clinical assessment. Full article

Synergistic pollution and carbon reduction is central to China’s Dual-Carbon strategy, yet the structural constraints on its regional evolution remain insufficiently examined. Using panel data from 16 prefecture-level cities in Shandong Province over 2010–2023, this study combines a coupling coordination degree model, Dagum Gini coefficient decomposition, and a panel fixed-effects model. Three findings emerge. First, the coupling coordination degree rose from 0.760 to 0.847, but this improvement was driven almost entirely by pollution reduction. The pollution reduction composite index increased by 31.7%, whereas the carbon reduction composite index declined by 6.6%, producing a pronounced “rapid pollution reduction, stagnant carbon reduction” asymmetry. Second, the spatial pattern was characterized by city-level differentiation rather than an east–west gradient, with Binzhou and Rizhao lagging persistently. Third, the Dagum Gini coefficient declined from 0.088 to 0.068, with transvariation intensity remaining the dominant source of disparity throughout the period (contributing 40.6–63.6%), indicating the limited explanatory power of the circle-based classification. Energy intensity and secondary industry share were the most robust drivers, whereas R&D intensity was statistically indistinguishable from zero. Policy should therefore prioritize energy efficiency improvement and city-specific targeting of persistently lagging cities. Full article

For regions with limited geothermal geological data, regional reservoir parameters are often difficult to directly incorporate into numerical models, while the linkage between heterogeneous zoning and detailed simulations remains insufficient and full-domain multi-scenario modeling requires substantial preprocessing effort. Taking the medium-to-deep hydrothermal geothermal reservoir in Anguang Town, Da’an City, Jilin Province, China, as the study area, this study proposes a regional geothermal development assessment framework integrating the digital reconstruction of parameter maps, finite-element parameter mapping, spatially constrained K-means clustering, and three-dimensional coupled groundwater flow and heat transport simulation. Regional reservoir parameter fields were reconstructed from published geological contour maps through georeferencing, contour vectorization, and continuous parameter-field reconstruction, and were subsequently mapped onto a three-dimensional FEFLOW finite-element mesh. Spatial cross-validation showed that the R² values for all predicted parameters exceeded 0.99 and the NRMSE values were below 2.18%, indicating that the reconstructed parameter fields reliably preserved the spatial variations represented by the original maps. Based on total reservoir thickness, hydraulic conductivity, porosity, and initial temperature, a spatially constrained K-means algorithm was then applied to delineate the study area. Comparison of different values of K and spatial constraint coefficients identified five hydrothermal zones, and the influence of parameter weighting on the zoning results was further evaluated. Representative locations from different zones were subsequently selected for constant-drawdown pumping tests and typical production–reinjection simulations. The results show that differences in development response among the zones are primarily controlled by regional hydrothermal conditions. At a drawdown of 100 m, pumping rates at typical representative locations ranged from 1255 to 3369 m³/d, while the 50-year cold-front propagation distances ranged from 348 to 674 m. Zones with higher hydraulic conductivity exhibited greater production capacity but also faster cold-front propagation. Further multi-well scenario analysis showed that increasing the number of wells did not necessarily improve long-term heat extraction performance. Among the tested configurations, the one-production–two-reinjection scheme achieved a cumulative heat extraction of 17.19 PJ over 50 years, with a production-temperature decline of only 0.31 °C, indicating good long-term thermal stability. Compared with directly conducting computationally intensive detailed simulations over the entire study area, the proposed framework establishes a continuous linkage among regional geological maps, heterogeneous zoning, and representative numerical scenarios, thereby reducing the domain requiring detailed simulation while retaining the major spatial heterogeneity. It provides a practical approach for candidate-area screening and comparative evaluation of production–reinjection schemes in data-limited geothermal regions. Full article

Background: Chronic subdural hematoma (CSDH) is the most common bleeding disorder in the nervous system and has a high recurrence rate, posing a serious threat to the lives of elderly individuals. Early identification and adjustment of the treatment regimen in the perioperative period is crucial. The aim of this study was to create a perioperative predictive nomogram for the early prediction of recurrent CSDH (R-CSDH). Methods: A predictive model was developed including 632 patients with CSDH hospitalized between 1 January 2021 and 31 December 2025 from Wuhan University People’s Hospital. The 206 patients from the First People’s Hospital of Xiantao City who were hospitalized between 1 January 2022 and 31 December 2025 formed an external validation cohort. Minimum absolute shrinkage was used to optimize feature selection, and logistic regression analysis was applied to further select features and construct a risk prediction model for recurrent CSDH. The calibration, discrimination, and clinical utility of the predictive models were assessed using calibration plots, the C-index, the area under the curve (AUC), and decision curve analysis. External validation was performed by an external validation cohort. Findings: Age, CT hematoma size, hematoma location, atrophy, density, surgical procedure and D-dimer were the predictors for constructing the predictive nomogram, and the results showed that a logistic regression model based on the above seven factors had good calibration and discriminative ability (C-index: 0.925; 95% confidence interval: 0.893–0.957). The AUC of this predictive model was 0.92487. The calibration curve of the nomogram was close to the ideal diagonal. In addition, the decision curve analysis showed that the net benefit of the model was significantly better. External validation demonstrated the reliability of the predictive nomogram. Interpretation: D-dimer was found to be a novel factor for recurrent chronic subdural hematoma. Our study provides clinicians with an accurate tool for identification and early intervention and therapy to reduce the incidence of R-CSDH. Full article

Rapid urbanization, a hot climate, and poor urban planning have intensified the Urban Heat Island (UHI) effect in Egyptian cities. However, remote-sensing-based mapping of UHIs at a micro-scale for mitigation purposes remains challenging. Reasons include the cost of high-resolution satellite imagery and legal restrictions on Unmanned Aerial Vehicles (UAVs) usage. To address this dilemma, this study introduced a framework for UHI spatial distribution mapping at a micro-scale for Aga City, Egypt, by simulating Brightness Temperature (BT) using the Discrete Anisotropic Radiative Transfer (DART) model. Optical and thermal UAV imagery—used here for the first time for urban 3D modeling in Egypt—were utilized for model parameterization. Multiple GIS-based approaches for reconstructing an accurate 3D model of the city were examined. The framework accurately simulated the BT spatial distribution, showing strong agreement with observations derived from Landsat 9 thermal data. UHI spatial distribution was generated, revealing moderate to severe hot spots across the city. UHI mitigation scenarios were tested by carrying out additional simulations under prescribed surface temperature assumptions. Results showed that white asphalt and green roofing reduced mean BT by 4 K and 5 K, respectively, making them promising solutions for UHI mitigation in dense Egyptian cities. Full article

The spatial differentiation of street-block vitality and its built-environment correlates provide an empirical basis for implementing refined street-block governance in the context of the renewal of existing urban areas. Based on the human–environment relationship theory, this study constructs an analytical framework of “built environment condition–street-block vitality response” and establishes a built environment indicator system based on a five-dimensional (5D) framework encompassing density, diversity, design, destination accessibility, and distance to transit. Empirically, street-block vitality is measured by the block-level average heat value derived from Baidu Huiyan data, representing the relative intensity of population activity within each street block. By integrating objective data—such as point-of-interest (POI) data and road networks—with residents’ subjective perceptions, this study employs spatial autocorrelation analysis, hot spot analysis, and multiscale geographically weighted regression (MGWR) models to identify the spatial patterns of street-block vitality in Baoding during the observation week of 8–14 April 2024 and their local associations with built environment factors. Model performance was evaluated through model comparison, sensitivity analysis, and spatial holdout validation. The results indicated that street-block vitality exhibited significant spatial clustering. High–high clusters were primarily distributed in urban development zones and their adjacent areas in the central-eastern, northeastern, and southwestern parts of the city, whereas low–low clusters were mainly concentrated in ecological protection and ecological control zones in the western, northwestern, and northern parts of the city. The density of scenic facilities (Den_sce), facility-density principal component (PC_FAC), road network density (Den_road), and distance to the nearest public transit stop (Dist_transit) generally showed a positive association with street-block vitality, whereas satisfaction with public transit stop placement (Convenience_sat) showed a negative association. The direction of the association between satisfaction with traditional cityscape design (Style_sat) and street-block vitality varied across locations, and spatial differences were evident in the local coefficients, the spatial extent of statistical significance, and optimal bandwidths for each variable. The research findings provide empirical evidence for formulating differentiated street-block renewal strategies based on territorial spatial functional zoning, street-block vitality types, and local association patterns. Full article

To address the challenges of reliance on manual expertise and insufficient automation and intelligence in settlement generalization during scale transformation, this paper proposes a multi-scale settlement generalization model (DSG-Trans) based on the fusion of dual-source data: raster maps and remote sensing images. Leveraging the cartographic correspondence between remote sensing image resolution and map scale, this model integrates source-scale maps with specific low-resolution remote sensing imagery to automatically generate target-scale raster maps that adhere to cartographic specifications. The model employs a bidirectional cross-attention mechanism and a Transformer architecture to facilitate the semantic integration of dual-source data, while a progressive upsampling strategy is utilized to reconstruct spatial details. Additionally, a composite loss function incorporating four constraints is designed to enhance the quality of settlement generalization. Experiments were conducted in New York City, USA, using a multi-source dataset constructed from 3D-GloBFP building footprint data and multi-source remote sensing imagery, focusing on two scale transformation scenarios: 1:5000 to 1:10,000 and 1:5000 to 1:25,000. The results indicate that the proposed method effectively learns and simulates expert generalization knowledge, significantly improving the automation level and scale adaptability of map generalization while preserving geometric morphology and spatial distribution patterns. Full article

Accurate reconstruction of digital surface models (DSMs) from multi-date high-resolution urban satellite imagery supports large-scale Earth observation and three-dimensional city modeling, with 3D Gaussian Splatting (3DGS) emerging as an efficient solution. Existing shadow-aware methods provide shadow-related spatial evidence, whereas geometric priors are generally applied across the image or scene without using this evidence to allocate supervision. Consequently, regions where shadows overlap abrupt elevation transitions may remain geometrically under-constrained, producing spatially non-uniform elevation errors whose severity varies with the geometry covered by shadow. To address this issue, we propose Shadow-Adaptive Geometric Supervision for Earth Observation Gaussian Splatting (SAGE). SAGE is primarily designed for urban scenes dominated by regular human-made structures and retains image-wide altitude guidance while allocating additional geometric supervision at two levels. At the spatial level, patch-level shadow weights determine where a shadow-targeted local altitude prior is applied to constrain relative altitude agreement. At the scene level, SAGE computes the reconstruction-derived Shadow Geometric Complexity (SGC) from multi-view shadow evidence and gradient statistics of the predicted DSM. SGC is used as a scene-level descriptor of shadow–gradient contrast and determines how strongly the local altitude prior and the complementary Manhattan orientation regularization are applied. Experiments on seven predominantly urban scenes from the DFC2019 and IARPA 2016 benchmarks show that SAGE achieves the best mean DSM accuracy on both benchmarks while retaining minute-level processing. Full article

Fixed urban sensing infrastructure underpins smart-city monitoring, transportation, environmental sensing, and digital twins, yet placement decisions often reduce deployment to geometric coverage or connectivity. This systematic mapping review is informed by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. It examines geospatial and three-dimensional (3D) placement methods for fixed urban sensors and synthesizes them through task-aware observability. The retained Scopus export dated 19 May 2026 contained 83 candidates; 20 were excluded before full-text assessment. After eligibility screening, 51 Scopus studies, four citation-identified studies, and eight studies from a focused assessment of known references formed a 63-publication analytical corpus comprising 56 direct-placement and seven placement-enabling studies. Studies were coded by spatial representation, objective, method, evaluation context, and reproducibility assets. Twenty-eight use 3D scenes or digital-twin representations, including 23 direct-placement studies. Only three report operational field evaluation, and four provide documented public-code links. Stream proportions are sensitive to the twelve non-database additions, all concentrated in the roadside, low-power wide-area network (LPWAN), and acoustic stream. Coding was collaborative rather than independently duplicated, so frequencies are descriptive of the agreed corpus. The review provides a cross-domain evidence map, reporting checklist, and research agenda for reproducible, transferable, task-aware urban sensing placement. Full article

This study develops a design procedure for electromechanical regenerative dampers using recorded damper stroke, velocity, and force from city, highway, country, and rough-road driving. The signals are converted into directional force, mechanical-power, energy-coverage, and operating-duty requirements. Motor size, gear ratio, lever ratio, power-network limits, packaging, and reflected inertia are screened, and 12 V and 48 V configurations are selected separately. Force tracking is checked with a prescribed-motion actuator model, and the selected designs are compared with a hydraulic reference in a quarter-car model under paired ISO Class B/80, C/60, and D/40 km/h inputs. The design points include 0.4 kW at 0.30 m/s, 0.6 kW for continuous-load screening, and 2.3 kW for 95% mechanical-energy coverage per damper. On the selected Class D/40 km/h input, the interface-force NRMSE is 52.3% for the 12 V configuration and 23.5% for the 48 V configuration. This is a comparison of two complete system designs, and prototype validation remains necessary. Full article

Background/Objectives: Evidence regarding the effects of proton pump inhibitors (PPIs) on glucose metabolism is conflicting, and real-world data on the association between chronic PPI use and glycemic control in older adults with type 2 diabetes mellitus (T2DM) are limited. This study aimed to evaluate the association between chronic PPI use and glycemic control in patients with T2DM aged 65 years or older. Methods: This single-center, retrospective cross-sectional study included 462 T2DM patients attending the Family Medicine Outpatient Clinic of Ankara Bilkent City Hospital between April 2025 and April 2026. Patients with regular PPI use for ≥6 months before the index HbA1c were classified as chronic PPI users. Suboptimal glycemic control was defined as HbA1c ≥ 8.0%. Associations were evaluated using logistic regression for suboptimal control and linear regression for HbA1c level. Results: Of the 462 patients, 316 (68.4%) were chronic PPI users. Mean HbA1c (7.11 ± 1.21% vs. 6.23 ± 0.99%; p < 0.001) and fasting plasma glucose (148.70 ± 64.87 vs. 116.23 ± 43.54 mg/dL; p < 0.001) were higher among PPI users. In multivariable logistic regression, chronic PPI use was independently associated with suboptimal glycemic control (adjusted odds ratio [aOR] = 3.416; 95% CI: 1.572–7.422; p = 0.002). This association remained significant in a sensitivity model additionally adjusted for age, sex, body mass index (BMI), estimated glomerular filtration rate (eGFR), smoking status, diabetes duration, insulin use, metformin use, and heart failure/coronary artery disease (aOR = 3.834; 95% CI: 1.734–8.478; p = 0.001). In multivariable linear regression, chronic PPI use was independently associated with a 0.793-percentage-point higher HbA1c level (95% CI: 0.570–1.016; p < 0.001). Conclusions: Chronic PPI use was independently associated with higher HbA1c levels and suboptimal glycemic control in older adults with T2DM. Because of the observational design, causality cannot be established. Prospective studies accounting for PPI indication, agent-specific effects, dose, and duration of exposure are needed to clarify the nature of this association. Full article

Using selected summer Landsat 8 scenes for Taiyuan, Chengdu, and Yinchuan, this study compared functional zone land surface temperature (LST) patterns, green-space–LST associations, and the relationships of 2D landscape configuration, blue-green composition, 3D building form, and terrain with LST. Separate city-specific Light Gradient Boosting Machine (LightGBM) models with SHapley Additive exPlanations (SHAP) interpretation characterized within-model predictor contributions, SHAP dependence patterns, and selected representative pairwise interactions. Polygon-level LST distributions differed among functional zones within every city (Kruskal–Wallis, all p < 0.001), and Commercial zones had higher LST than Park/green-space zones in all three cases with large pairwise effect sizes. Other pairwise contrasts were city-specific. Most city-zone regressions showed negative green-space–LST associations, whereas Chengdu Industrial showed no clear association. Building volume ranked among the Top-3 predictors in every city model, but the co-dominant structures differed: building volume–green-space proportion–elevation in Taiyuan, mean building height–building volume–green-space proportion in Chengdu, and building volume–aggregation index–elevation in Yinchuan. Five-fold fixed-specification spatial block cross-validation provided a more conservative spatially held-out sensitivity assessment than ordinary random splitting, while reference and repeated spatial-block out-of-fold SHAP analyses showed stable within-model predictor rankings and contribution shares. An 85-scene multi-date Landsat comparison with selected-scene leave-one-out sensitivity supported broad summer spatial consistency in Taiyuan and Yinchuan, with partial, context-dependent support in Chengdu. After excluding each city’s selected 2019 scene from its own reference median, the relative–LST Spearman correlations remained 0.895, 0.861, and 0.909 for Taiyuan, Chengdu, and Yinchuan, respectively. These findings distinguish recurring comparative anchors from context-dependent predictor structures, supporting context-sensitive planning hypotheses rather than universal predictor rankings. Full article

Smart mobility is a core component of the smart city; it comprises various intelligent systems that make the city more safe, livable, efficient and sustainable. One of these system is the smart parking system (SPS). A smart parking system is known as the application of advanced information and communication technologies to the traditional parking system. It is a technology-driven approach that plays a crucial role in smart cities, particularly in improving traffic management. The smart parking system is a multi-faceted concept that integrates parking reservation, automatic payment, parking space detection, outside/inside guidance, data analytics, and user interaction interfaces. While various advanced technologies, such as Wireless Sensors Networks (WSNs), Internet of Things (IoT), Cloud Computing (CC), FoG computing, and Artificial Intelligence (AI), are involved in engineering and implementing these functional facets, Vehicular Ad Hoc Network (VANET) and its derivatives have received relatively limited interest in the literature of smart parking systems. Moreover, much of the literature primarily focuses on reducing parking search time outside parking facilities. However, drivers in practice often waste a significant amount of time inside the parking facility navigating for a free parking spot. In this work, we address two main functions in smart parking: a parking spot allocation mechanism and on-site parking real-time guidance. We developed an automatic parking spot allocation and in-parking guidance approach where the communication layer is modelled using IEEE 802.11p DRSC as the reference infrastructure. The simulation assumes a reliable channel and excludes lower-layer performance evaluation such as packet loss, congestion, latency, interferences, or coverage sensitivity. The reported improvements therefore characterize the application-layer benefit of pre-assignment reservation and spatial guidance, independent of the specific communication technology employed. The designed system, called in this paper VANET-guided, aims to significantly decrease the parking spot search time inside the parking facility. This paper presents a comparative evaluation study of a VANET-guided smart parking framework against a fully unguided baseline using NS-3.36.1, evaluating their performance across four key operational dimensions: lot occupancy, number of vehicles within facility, inter-arrival gap, and driver decision delay, with R = 30 statistically validated replications per experimental cell, using a series of Monte Carlo simulations under diverse traffic and mobility conditions. The guided system combines atomic server-side spot reservation, spatial allocation strategies, and a Kalman-filtered RSSI-based in-lot positioning pipeline, while the unguided baseline models realistic driver behavior, including cognitive decision delays, spot rejection, and concurrent spot competition. To address the structural asymmetry between the two systems, two intermediate baselines are evaluated: (i) an information-only configuration and (ii) a reservation-only configuration, providing an exploratory comparison of the respective roles of availability information, reservation, and spatial targeting. The findings indicate that the fully guided system (called in this paper VANET or VANET-guided) reduces mean parking search time by 28% to 71% relative to the fully unguided baseline across all tested occupancy levels, with statistical validation through 95% confidence interval analysis, paired t-test and Wilcoxon signed-rank test, practical effect sizes reported through Cohen’s d, and distributional statistics including standard deviation, coefficient of variation, and 90th–95th percentile search times. Full article

Africa remains the most under-addressed continent on the planet, with an estimated 60–70% of streets and dwellings lacking formal, standardised addresses. This deficit has profound consequences for governance, emergency response, e-commerce, logistics, and the daily lives of over 1.4 billion people. We present a hybrid graph-theoretic and machine learning framework for automated address generation, trained on high-quality open data from three European cities (Stuttgart, Paris, and Bern) and progressively adapted to seven African cities (Kigali, Dakar, and Kampala as training cities; Dar es Salaam, Nairobi, Kinshasa, and Lagos as unseen test cities). The framework models street networks as primal and dual graphs, applies the Intersection Continuity Negotiation algorithm for stroke detection, and deploys four supervised classifiers for connectivity error detection, road type classification, building-to-street assignment, and address quality scoring. Building footprints are fused from three complementary sources (OpenStreetMap, Google Open Buildings, Microsoft Building Footprints) and augmented with SRTM elevation data and the World Settlement Footprint 3D (WSF 3D) dataset. Across ten study cities, the pipeline produces 3.2 million merged building footprints, 467,000 street edges, and 880,464 quality-scored addresses, and a five-version iterative improvement cycle reduces the EU–Africa address quality gap by 54.2%. To validate African address outputs independently, we construct a cross-validation database of 1000 verifiable addresses of known organisations across the seven African cities; cross-validation achieves a geometric match rate of 79.6% within 100 m and 83.5% within 150 m. The system is implemented as an open-source QGIS plugin and ArcGIS Toolbox. Full article