Search Results (10,469)

Objectives: To develop a prediction model able to accurately predict which patients will harbor higher risk prostate cancer in the systematic biopsy template compared to the targeted biopsy during MRI/US fusion biopsy. Methods: We included patients who underwent fusion biopsy. Clinical and radiographic variables were collected from patients' records. The outcome of the model was higher risk prostate cancer in the systematic compared with targeted biopsies. An extreme gradient boosting model was trained and tested. We evaluated variable importance and clinical benefit. Results: Five hundred and twenty-nine patients were included. Eighty-two (15.5%) patients had higher risk prostate cancer in the systematic biopsies. The area under the ROC curve and negative predictive value were 0.82 and 0.92, respectively. The four most important features for outcome prediction were prostate volume, PSAD, patient’s age, and PSA. The decision curve showed increased clinical benefit of our model at threshold probabilities of 0–0.5. Limitations include the retrospective design of the study and the lack of external validation of the model. Conclusions: We developed a prediction model able to accurately predict which patients must undergo systematic and targeted biopsy. This prediction model has the potential to help in the decision whether to perform SB and thus may lower the adverse event rate while keeping a high detection rate. Full article

Logistics alliance and integrated passenger-freight transit are two widely adopted collaborative logistics modes in rural areas. With the rapid development of agricultural e-commerce, rural “first-mile” logistics has become critical for agricultural products' upward circulation, but remains constrained by high costs and insufficient service provision. Existing studies mainly focus on a single transportation mode and pay limited attention to logistics service providers’ strategic choice among alternative modes under government intervention. Using a Stackelberg game framework, this study models the interaction among the government, a logistics service provider, and a rural bus company, and analyzes transportation mode choice and subsidy effectiveness. The results show that government subsidies improve rural “first-mile” logistics service levels and stimulate demand for cargo collection services. Transportation mode choice is jointly influenced by market share, service cost coefficient, and subsidy intensity. Large-scale logistics service providers tend to adopt the integrated passenger-freight transit mode when subsidies are insufficient, and prefer the logistics alliance mode when subsidy support becomes adequate. These findings suggest that subsidy policies should consider fiscal capacity and regional operating costs: the integrated passenger-freight transit is more suitable under limited budgets, while the logistics alliance becomes preferable for promoting regional logistics development when sufficient subsidies can be sustained. Full article

Pseudomonas aeruginosa is a Gram-negative pathogen with a remarkable capacity to acquire multiple resistance mechanisms, severely limiting current therapeutic options. Consequently, the identification of new antimicrobial agents remains a critical priority. In this study, an integrated in silico-guided strategy was applied to identify small molecules with antibacterial potential against P. aeruginosa, targeting the quorum-sensing regulator LasR (PDB ID: 2UV0) and elastase (PDB ID: 1U4G). Pharmacophore modeling was performed for both targets, followed by ligand-based virtual screening, structure-based virtual screening (SBVS), and MM-GBSA (Molecular Mechanics-Generalized Born Surface Area) binding free energy calculations. Top-ranked compounds based on predicted binding affinity were selected for in vitro cytotoxicity and antibacterial evaluation. Antimicrobial activity was assessed against three P. aeruginosa strains: an American Type Culture Collection (ATCC) reference strain, a clinically susceptible isolate, and an extensively drug-resistant (XDR) clinical isolate. SBVS yielded docking scores ranging from −6.96 to −12.256 kcal/mol, with MM-GBSA binding free energies between −18.554 and −88.00 kcal/mol. Minimum inhibitory concentration (MIC) assays revealed that MolPort-001-974-907, MolPort-002-099-073, MolPort-008-336-135, and MolPort-008-339-179 exhibited MIC values of 62.5 µg/mL against the ATCC strain, indicating weak-to-moderate antibacterial activity consistent with early-stage hit compounds. MolPort-008-336-135 showed the most favorable activity against the clinically susceptible isolate, with an MIC of 62.5 µg/mL, while maintaining HepG2 cell viability above 70% at this concentration and an half-maximal inhibitory concentration (IC₅₀) greater than 500 µg/mL. In contrast, all tested compounds displayed MIC values above 62.5 µg/mL against the XDR isolate, reflecting limited efficacy against highly resistant strains. Overall, these results demonstrate the utility of in silico-driven approaches for the identification of antibacterial hit compounds targeting LasR and elastase, while highlighting the need for structure–activity relationship optimization to improve potency, selectivity, and activity against multidrug-resistant P. aeruginosa. Full article

Amid China’s shift from a model of urban “incremental expansion” to one focused on “stock optimization”, the renewal of streetscapes has taken center stage as a critical approach to improving the human experience within urban environments. However, empirical insight into how visual interventions affect aesthetic perception across different road types remains notably limited. This study addresses that gap through a spatiotemporal investigation of Zhengzhou’s streetscape transformations between 2017 and 2022. Major roads were categorized into four functional types—freeway, under-freeway, regular road, and tunnel—to better capture perceptual variation. Leveraging a Fully Convolutional Network (FCN), we extracted nine visual components from historical street views and paired them with crowd-sourced “beauty” ratings from the MIT Place Pulse 2.0 dataset. Statistical analyses, including paired t-tests and Kernel Density Estimation (KDE), indicated marked improvements in perceived beauty following renewal, with the exception of tunnel segments. Through Random Forest (RF) regression and SHapley Additive exPlanations (SHAP) interpretation, greening emerged as the most influential driver of aesthetic enhancement—most prominently on regular roads (SHAP = 2.246). The impact of renewal was found to be context-specific: green belts were most effective in under-freeway areas (SHAP = +0.8), while improvements to pavement (SHAP = +0.97) and street vitality were key for regular roads. Notably, SHAP analysis revealed non-linear relationships, such as diminishing perceptual returns when green coverage exceeded certain thresholds. These findings inform a “visual renewal–perceptual response” framework, offering data-driven guidance for adaptive, human-centered upgrades in high-density urban settings. Full article

As a key technology in the intelligent interpretation of remote sensing, remote sensing image change detection aims to automatically identify surface changes from images of the same area acquired at different times. Although vision foundation models have demonstrated outstanding capabilities in image feature representation, their inherent patch-based processing and global attention mechanisms limit their effectiveness in perceiving multi-scale targets. To address this, we propose a multi-scale remote sensing image change detection network based on a vision foundation model, termed SAM-MSCD. This network integrates an efficient parameter fine-tuning strategy with a cross-temporal multi-scale feature fusion mechanism, significantly improving change perception accuracy in complex scenarios. Specifically, the Low-Rank Adaptation mechanism is adopted for parameter-efficient fine-tuning of the Segment Anything Model (SAM) image encoder, adapting it for the remote sensing change detection task. A bi-temporal feature interaction module(BIM) is designed to enhance the semantic alignment and the modeling of change relationships between feature maps from different time phases. Furthermore, a change feature enhancement module (CFEM) is proposed to fuse and highlight differential information from different levels, achieving precise capture of multi-scale changes. Comprehensive experimental results on four public remote sensing change detection datasets, namely LEVIR-CD, WHU-CD, NJDS, and MSRS-CD, demonstrate that SAM-MSCD surpasses current state-of-the-art (SOTA) methods on several key evaluation metrics, including the F1-score and Intersection over Union(IoU), indicating its broad prospects for practical application. Full article

This study tests the performance of 30 tree-level models of literature to predict the aboveground biomass (AGB) of trees in 200 1 ha simulated plots representing the following two successional stages of Amazonian forests: Advanced Secondary Forest (ASF) and Mature Forest (MF). This matters because reliable biomass estimates are fundamental to carbon quantification and climate policy. Ensuring consistency between tree-level and plot-level accuracy strengthens transparency and credibility in global reporting. The aim was twofold: (i) to recommend accurate models to predict biomass in the Amazon and (ii) to detect what characteristics of the model calibration dataset can affect the accuracy of the AGB predicted at the plot level. We considered the characteristics of datasets, sample size, minimum, maximum, and range of tree diameters, as well as the coefficient of determination, root mean square error (RMSE), and number of predictors of the 30 models consulted in the literature. These characteristics were correlated with the biomass error per unit area. We listed 11 models based on their acceptable (overall ± 10%) accuracy, whereas four models overestimated and 15 models underestimated the biomass per unit area beyond the acceptable limit. Our analysis pointed out that the strongest (but moderate) correlation (r) was observed for the RMSE of the models, i.e., precision of model predictions. These correlations were r = 0.60 (p = 0.40) for ASF (kg) and r = 0.40 (p = 0.60) for MF (kg) and r = 0.57 (p = 0.18) for ASF (log) and r = 0.21 (p = 0.64) for MF (log), which means that models with greater uncertainty (higher RMSE) tend to produce larger errors in AGB estimation. As a main conclusion, this study cautions that selecting one model among several based on the lowest RMSE is a misleading practice because the precision of predictions at the tree level is not in agreement with the precision at the plot level. Full article

Government building projects are particularly complex due to their scale and number of end users, which makes construction prices time-consuming and prone to error. Machine learning is recognized for its ability to process large volumes of complex data quickly with high accuracy, but only a limited number of studies have applied Deep Learning in the early construction stage. Therefore, we aimed to evaluate the potential of Deep Learning to predict construction contract prices for government buildings. Factors were identified through a literature review and interviews with eight experts, and data were collected from 300 government construction projects obtained from Thailand’s Electronic Government Procurement (e-GP) database, the national centralized platform for transparent public bidding. By varying the number of parameters, 80 models were developed and tested. The best-performing model had a three-hidden-layer ratio of 128:64:32 with a Quadratic Loss Function, achieving an R² of 0.918 and an RMSE of 2.022. The results showed 14 significant factors, with the top 5 being (1) usable area, (2) number of sanitary wares, (3) number of rooms, (4) height, and (5) number of elevators. Sensitivity analysis was subsequently conducted to enhance the explainability of the model. The findings demonstrate the potential of Deep Learning to enhance the accuracy of determining construction price and support more effective government budget planning and decision making. Full article

The transient electromagnetic (TEM) method is a key detection and monitoring technology for safe coal-mine production. Surface TEM depth penetration is limited by real geological conditions and transmitter–receiver hardware performance. Compared with the surface TEM method, the tunnel TEM method can enhance the depth of exploration to some extent, but it is constrained by the limited working space of the roadway, which makes it difficult to perform the area-wide and multi-line data acquisition, and thus the lateral detection resolution is directly compromised. Consequently, either surface or tunnel TEM alone suffers inherent limitations. The multidimensional surface and surface-to-tunnel TEM method employs a single large-loop transmitter and records electromagnetic (EM) signals both on the surface and in the tunnel, enabling joint data interpretation. The joint TEM observation method effectively addresses the limitations by using a single observation mode, with the goal of achieving high-precision detection. To investigate the detection capabilities of the joint surface and surface-to-tunnel TEM method, we propose a three-dimensional (3D) joint surface and surface-to-tunnel TEM forward modeling method based on the spectral element method (SEM). The SEM, using high-order vector basis functions, enables high-precision modeling of TEM responses with complex geo-electric earth models. The accuracy of the SEM is validated through comparisons with one-dimensional (1D) TEM semi-analytical solutions. To further reveal TEM response characteristics and multi-dimensional resolution under joint surface and tunnel detection modes, we construct several typical 3D geo-electric earth models and apply the SEM algorithm to simulate the TEM responses. We systematically analyze the horizontal and vertical resolution of 3D earth model targets at different decay times. The numerical results demonstrate that surface multi-line TEM surveying can accurately delineate the lateral extent of the target body, while vertical in-tunnel measurements are crucial for identifying the top and bottom interfaces of geological targets adjacent to the tunnel. Finally, the theoretical modeling results demonstrate that compared to individual TEM methods, the multi-dimensional joint surface and tunnel TEM observation yields superior target spatial information and markedly improves TEM detection efficacy under complex conditions. The 3D TEM forward modeling based on the SEM provides the theoretical foundation for subsequent 3D inversion and interpretation of surface-to-surface and surface-to-tunnel joint TEM data. Full article

Large-scale wind farms operate under highly unsteady atmospheric inflows, where transient turbulence, dynamic wake interactions, and inflow-wake coupling reduce energy production and exacerbate turbine loads. Over the past five years, advances in high-fidelity computational fluid dynamics (CFDs), large eddy simulation (LES), machine learning (ML)-based wake modeling, and multi-objective optimization have reshaped wind farm layout optimization under dynamic inflow conditions. This review synthesizes recent progress in five key areas: dynamic inflow and high-fidelity wake modeling (including LES-driven transient wake evolution and turbulence-resolved inflow generation), data-driven wake prediction, multi-objective layout optimization (considering the annual energy production (AEP), fatigue load constraints, and the levelized cost of energy (LCOE)), blockage modeling for complex terrain and yaw misalignment, and real-time optimization addressing inflow, turbine performance, and modeling uncertainties. Coupling transient wake models with surrogate-assisted multi-objective optimization enables a computationally efficient and physically consistent layout design. Key open challenges (dynamic wake controllability, real-time optimization under uncertainty, and integration with next-generation farm-level control systems) and future directions for enhancing large-scale wind farm resilience and cost-competitiveness are also identified. However, despite significant progress, existing models still face fundamental limitations, such as oversimplified treatment of complex turbulence structures, poor generalization under extreme or atypical conditions, and inadequate capture of long-timescale dynamic responses, which constrain their reliability in practical optimization settings. Full article

Forest structural parameters, such as canopy closure, stand density, diameter at breast height, tree height, leaf area index, stand age, and biomass, are fundamental for quantifying forest ecosystem functioning and supporting sustainable forest management. Remote sensing has become an indispensable tool for forest structural parameter estimation. Commonly used data sources include optical imagery, synthetic aperture radar (SAR), light detection and ranging (LiDAR), unmanned aerial vehicles (UAVs), and multisource data fusion. Correspondingly, modeling approaches have evolved from empirical and statistical methods to machine learning, deep learning, and hybrid physical-data-driven models, enabling improved characterization of nonlinear and complex forest structures. Each data source and modeling strategy offers unique strengths and limitations with respect to accuracy, scalability, interpretability, and transferability. This review provides a concise synthesis of recent advances in remote sensing data sources and model algorithms for forest structural parameter estimation, evaluates the strengths and limitations of different sensors and algorithms, and highlights key challenges related to uncertainty, scalability, transferability, and model interpretability. Finally, future research directions are discussed, emphasizing cross-scale integration, multisource data fusion, and physically informed deep learning frameworks as promising pathways toward more accurate, robust, and ecologically interpretable forest structural parameter modeling at regional to global scales. Full article

Anemone davidii Franch. is an herbaceous plant with high ornamental and medicinal value belonging to the Ranunculaceae family. Understanding its genetic diversity and predicting its potential habitat shifts are crucial for its germplasm conservation. In this study, we analyzed the genetic diversity of 164 individuals from A. davidii and its relatives using genotypic sequencing (GBS) technology. The results indicated that the expected heterozygosity (He) of 12 A. davidii populations ranged from 0.074 to 0.095, while the observed heterozygosity (Ho) ranged from 0.105 to 0.127. Phylogenetic, principal component (PCA), and population structure analyses revealed clear genetic separation among A. davidii, A. griffithii, and A. scabriuscula. The 12 A. davidii populations were grouped into three genetic clusters. Six populations—CQ, ES, SNJ, SZ, TR, and WX—of Central China were clustered together. Southwestern region populations were divided into two clusters (DG, PZ, SF and DY, EMS, HY). Low genetic differentiation values (Fst, 0.018–0.053) and high levels of gene flow (Nm, 4.4678–13.639) between populations were observed in this study, indicating that genetic differentiation was lower between adjacent populations. We also used the Maximum Entropy (MaxEnt) model to predict changes in suitable distribution areas of A. davidii across four time periods and two climate scenarios (RCP4.5, RCP8.5). Compared to the Last Glacial Maximum (LGM), the current suitable habitat area has contracted. Future climate projections indicated a progressive range contraction under both scenarios. Therefore, appropriate conservation measures are needed to address its limited genetic diversity and projected habitat loss under climate change. Our findings provide insights into the population genetics of A. davidii and the impact of climate change on plants of Southwestern China. Full article

Recurrent flooding poses a persistent and growing threat to West African watersheds facing rapid urbanization and climate change. Despite advances in machine learning and geospatial datasets, urban planning and flood prevention often rely on limited datasets and traditional analysis. This study addresses this research gap in the Bonoumin-Palmeraie watershed (Abidjan, Côte d’Ivoire) by developing an integrated approach leveraging remote sensing, Geographic Information Systems (GIS), and the Random Forest algorithm to assess and map flood susceptibility. Twelve conditioning factors related to topography, hydrology, land use, and climate were derived from Sentinel-1, ALOS PALSAR, and multi-source earth observation datasets. Historical flood extents were mapped in Google Earth Engine to train the Random Forest model in a Google Colab environment. The model demonstrated high discriminatory power, yielding an Area Under the Curve of 0.94 and Overall Accuracy of 0.83. Drainage density, rainfall, and altitude were identified as the primary explanatory drivers. The resulting flood susceptibility map indicates that 39% of the watershed exhibits medium to very high susceptibility, with critical hotspots in the neighborhoods of Palmeraie, Attoban, Akouedo, Djorogobité, and Riviera-Sogefiha. While limited by the exclusion of certain anthropogenic variables and ground truth constraints, the study provides a reproducible, data-driven framework for flood risk assessment in tropical urban environments. These findings offer essential scientific support for urban planners and decision-makers to enhance territorial planning and sustainable flood management in Abidjan. Full article

Magnetron sputtering serves as a key method for fabricating functional thin films used in transparent film heaters. However, as heater designs become more intricate, achieving uniform film deposition on nonrectangular areas induces localized overheating owing to current density crowding, compromising long-term reliability of the device. To address this limitation, a simulation-assisted design and fabrication strategy is presented to realize a uniform temperature profile through the precise regulation of the sheet resistance distribution of the film. Initially, an electrothermal-coupled finite element model was established using COMSOL Multiphysics to inversely determine the spatial gradient of sheet resistance required for achieving a uniform thermal distribution. Subsequently, a custom-designed mesh shadow mask was used to locally adjust the flux of indium tin oxide (ITO) sputtered particles, enabling the establishment of a relationship between the mask’s aperture geometry and the resulting particle deposition profile. The magnetic field and plasma simulations were integrated to model particle transport and design a specialized gradient aperture-based shadow mask, enabling the deposition of an ITO film with a controlled sheet resistance gradient in a single magnetron sputtering step. Experimental results demonstrated that the proposed method decreased the maximum temperature variation by 8.25 °C and reduced the standard deviation of the surface temperature by 82.1% at an average temperature of 45 °C within a defined nonrectangular heating region, demonstrating a substantial improvement in temperature uniformity relative to conventional uniform coating processes. Full article

Illegal dumping poses serious risks to ecosystems and human health, requiring effective and timely monitoring strategies. Advances in uncrewed aerial vehicles (UAVs), photogrammetry, and deep learning (DL) have created new opportunities for detecting and characterizing waste objects over large areas. Within the framework of the EMERITUS Project, an EU Horizon Europe initiative supporting the fight against environmental crimes, this study evaluates the performance of pre-trained prompt-based multimodal (PBM) DL models integrated into ArcGIS Pro for object detection and segmentation. To test such models, UAV surveys were specially conducted at a semi-controlled test site in northern Italy, producing very high-resolution orthoimages and video frames populated with simulated waste objects such as tyres, barrels, and sand piles. Three PBM models (CLIPSeg, GroundingDINO, and TextSAM) were tested under varying hyperparameters and input conditions, including orthophotos at multiple resolutions and frames extracted from UAV-acquired videos. Results show that model performance is highly dependent on object type and imagery resolution. In contrast, within the limited ranges tested, hyperparameter tuning rarely produced significant improvements. The evaluation of the models was performed using low IoU to generalize across different types of detection models and to focus on the ability of detecting object. When evaluating the models with orthoimagery, CLIPSeg achieved the highest accuracy with F1 scores up to 0.88 for tyres, whereas barrels and ambiguous classes consistently underperformed. Video-derived (oblique) frames generally outperformed orthophotos, reflecting a closer match to model training perspectives. Despite the current limitations in performances highlighted by the tests, PBM models demonstrate strong potential for democratizing GeoAI (Geospatial Artificial Intelligence). These tools effectively enable non-expert users to employ zero-shot classification in UAV-based monitoring workflows targeting environmental crime. Full article

Remote sensing semantic segmentation encounters several challenges, including scale variation, the coexistence of class similarity and intra-class diversity, difficulties in modeling long-range dependencies, and shadow occlusions. Slender structures and complex boundaries present particular segmentation difficulties, especially in high-resolution imagery acquired by satellite and aerial cameras, UAV-borne optical sensors, and other imaging payloads. These sensing systems deliver large-area coverage with fine ground sampling distance, which magnifies domain shifts between different sensors and acquisition conditions. This work builds upon DeepLabV3+ and proposes complementary improvements at three stages: input, context, and decoder fusion. First, to mitigate the interference of complex and heterogeneous data distributions on network optimization, a feature-mapping network is introduced to project raw images into a simpler distribution before they are fed into the segmentation backbone. This approach facilitates training and enhances feature separability. Second, although the Atrous Spatial Pyramid Pooling (ASPP) aggregates multi-scale context, it remains insufficient for modeling long-range dependencies. Therefore, a routing-style global modeling module is incorporated after ASPP to strengthen global relation modeling and ensure cross-region semantic consistency. Third, considering that the fusion between shallow details and deep semantics in the decoder is limited and prone to boundary blurring, a fusion module is designed to facilitate deep interaction and joint learning through cross-layer feature alignment and coupling. The proposed model improves the mean Intersection over Union (mIoU) by 8.83% on the LoveDA dataset and by 6.72% on the ISPRS Potsdam dataset compared to the baseline. Qualitative results further demonstrate clearer boundaries and more stable region annotations, while the proposed modules are plug-and-play and easy to integrate into camera-based remote sensing pipelines and other imaging-sensor systems, providing a practical accuracy–efficiency trade-off. Full article