Search Results (111)

An accurate assessment of chronic kidney disease (CKD) requires understanding both renal morphology and functional decline, yet most deep learning approaches treat segmentation and eGFR prediction as separate tasks. This paper proposes the Functionally Guided CKD U-Net (FG-CKD-UNet), a dual-headed multitask architecture that integrates multi-class kidney segmentation with end-to-end eGFR prediction using a structure–function consistency loss. The model incorporates a morphological biomarker extractor to derive cortical thickness, kidney volume, and cortex–medulla ratios, enabling explicit coupling between anatomy and physiology. Experiments on T2-weighted MRI and colorized CT datasets demonstrate that the proposed method surpasses state-of-the-art segmentation baselines, achieving a Dice score of 0.94 and an HD95 of 9.8 mm. For functional prediction, the model achieves an MAE of 0.039, an RMSE of 0.058, and a Pearson correlation of 0.92, outperforming CNN, MLP, and ResNet baselines. The structure–function consistency mechanism reduces the consistency error from 0.071 to 0.042, confirming coherent physiological modeling. The results indicate that the FG-CKD-UNet provides a reliable, interpretable, and physiologically grounded framework for comprehensive CKD assessment. Full article

This paper presents results of numerical modeling of tunneling using mechanized tunnel boring machines (TBMs) based on a methodology for determining the volume loss cohesionless (loose) soils, denoted as V_L,LSR, for shallow tunnels in dispersive soils to estimate surface and foundation on settlement natural ground. Existing methods for estimating ground surface and structural settlements have significant drawbacks, caused by several factors, including the complexity of determining volume loss using the proposed methodologies, a limited number of empirical parameters describing the technological features of TBM operations, the absence of methods in Russian regulatory documentation for determining volume loss in tunnels with diameters of 6 m or more, among other issues. The study aims to validate a previously developed method for estimating V_L,LSR and an empirical equation for predicting surface settlements, S_max, to assess additional settlements induced by tunneling. The proposed volume loss methodology and the modified S_max expression from Peck R.B. (1969), derived from monitoring data, are used in empirical calculations and numerical modeling of surface and building settlements during TBM tunneling. Validation results include back-analysis of geotechnical “tunnel–ground–structure” interaction models, comparisons of additional settlements from design calculations and field monitoring data, as well as comparisons with existing empirical relationships and relevant regulatory documents, followed by recommendations for their integrated application. The validated methods demonstrate good agreement with observed monitoring data, while providing sufficient engineering safety margins, confirming the applicability of the V_L,LSR and the modified S_max expression by Peck R.B. (1969) for predicting settlements of tunneling and identifying directions for further research. Full article

Air-assisted sprayers are widely used in orchards to ensure deep canopy penetration and effective pesticide coverage, yet excessive or misdirected airflow often causes spray drift and ground losses. This study evaluated spray deposition efficiency, drift, and environmental performance of a novel double-tower orchard sprayer (DIVENT) equipped with two independently driven axial fans allowing separate airflow adjustment on each side. Field experiments were conducted in apple orchards under crosswind conditions using the following three airflow emission scenarios (air volume to the LEFT/RIGHT side of sprayer): symmetrical (100%/100%), compensating crosswind (30%/100%), and one-sided (0%/100%). Measurements of spray deposition within the canopy, ground losses, and off-target deposition drift were performed using fluorescent tracer, and power consumption was recorded to estimate fuel use and CO₂ emissions. The compensating airflow setting significantly improved spray targeting, reducing both in-orchard ground losses and off-target drift by up to 60%, while maintaining uniform canopy coverage comparable to the conventional symmetrical mode. The one-sided emission scenario achieved the highest drift reduction (67.8%) and the lowest power and CO₂ emissions, though at the cost of reduced canopy deposition. Overall, the study demonstrates that independent fan control allows effective adaptation of spraying to weather and canopy conditions, providing substantial environmental and energy benefits without compromising spray efficiency. Full article

In Europe, there is a growing interest in crop spraying using unmanned aerial vehicles (UAVs, drones), although current legislation imposes significant limitations on this technique. Spraying of orchard crops with drones remains particularly challenging due to the risks of spray drift and insufficient deposition uniformity. This study evaluated spray deposition within tree canopies (in two application terms), airborne and sediment drift losses, and contamination of the spraying equipment. The performance of a medium-sized drone (ABZ Innovation L10, maximum take-off weight 29 kg) was compared at flight speeds of 1.8, 2.7, and 3.6 m·s⁻¹ with that of a conventional orchard sprayer (Munckhof axial sprayer with column attachment, operating at 1.7 m·s⁻¹). A fluorescent tracer (BF7G, 1200 g·ha⁻¹) was used in all trials, with spray volume rates of 27 or 40 L·ha⁻¹ for the drone and 400 L·ha⁻¹ for the sprayer. In most cases, deposition within the tree canopy was significantly lower for the drone. Poor uniformity of spray distribution was observed, especially between the upper and lower surfaces of collector plates with attached filter papers and between the top and bottom canopy zones. Airborne drift increased significantly with higher drone flight speeds, while sediment drift decreased. At 1.8 m·s⁻¹, both drift types were comparable to those from the conventional sprayer. Drone surface contamination was several times lower than that of the ground sprayer, even when accounting for differences in equipment surface area. Full article

To effectively combat wildfires, ground teams must reach the fire site via road network within critical response time. However, low-standard forest roads can reduce firetruck speeds and delay fire response times. This study aimed to investigate how improving road standards affects firefighting access within critical response time and contributes to reducing timber losses. This study was conducted in Antalya, the city most affected by wildfires in Türkiye. In the study, highly fire-prone forests were first identified based on a fire probability map of Antalya, developed through a GIS-based MCDA model incorporating the Fuzzy-AHP method. Then, the highly fire-prone forests and their corresponding timber volume were determined. Finally, the economic value of timber saved from fire and the present net value of total road costs were determined. As a result of improving forest roads, the forest areas that could be reached in time increased by 11.04%, making an additional 81,867.53 hectare of highly fire-prone forests accessible. The amount and economic value of timber products saved in this area were 971,195.55 m³ and €37,689,301, respectively. The cost of improved roads was €37,386,622 while the resulting positive net economic value of €302,679 indicates that investing in forest roads improvements is a viable option. Full article

The rapid growth of Internet of Things (IoT) deployments has created an urgent need for energy-efficient communication strategies that can adapt to dynamic operational conditions. This study presents a novel adaptive protocol selection framework that dynamically optimizes IoT communication energy consumption through context-aware decision making, achieving up to 34% energy reduction compared to static protocol selection. The framework is grounded in a comprehensive empirical evaluation of three widely used IoT communication protocols—MQTT, CoAP, and HTTP—using Intel’s Running Average Power Limit (RAPL) for precise energy measurement across varied network conditions including packet loss (0–20%) and latency variations (1–200 ms). Our key contribution is the design and validation of an adaptive selection mechanism that employs multi-criteria decision making with hysteresis control to prevent oscillation, dynamically switching between protocols based on six runtime metrics: message frequency, payload size, network conditions, packet loss rate, available energy budget, and QoS requirements. Results show MQTT consumes only 40% of HTTP’s energy per byte at high volumes (>10,000 messages), while HTTP remains practical for low-volume traffic (<10 msg/min). A novel finding reveals receiver nodes consistently consume 15–20% more energy than senders, requiring new design considerations for IoT gateways. The framework demonstrates robust performance across simulated real-world conditions, maintaining 92% of optimal performance while requiring 85% less computation than machine learning approaches. These findings offer actionable guidance for IoT architects and developers, positioning this work as a practical solution for energy-aware IoT communication in production environments. Full article

The solar energy reaching the immediate surroundings of a single-family house throughout the year is sufficient to repeatedly and fully cover its heating needs during the heating season in a temperate climate. Nevertheless, modern technology is not yet able to fully solve the problem of thermal self-sufficiency in single-family houses. It is therefore advisable to seek solutions that improve the thermal efficiency of domestic solar installations. Efficient use of solar radiation heat accumulated during the summer months for heating requires the use of high-volume storage tanks. Another option is to discharge excess heat outside the system during the summer. This publication focuses on the latter solution. A model of the solar heating system for a residential building and pool with a storage tank powered by solar energy has been developed. Simulation calculations were performed, showing that the removal of excess heat is a beneficial solution, especially when this energy can be used to heat water in the pool. The calculations concerned the heating of a single-family house in a temperate climate. Lowering the temperature of the water in the storage tank reduces heat losses from the tank to the environment (ground), while supplying the solar collectors with lower-temperature fluid increases the driving force of the heat transfer process. Full article

The subject of the research was a prototype two-row sprayer, equipped with a centrifugal fan and directed air-jet emission system, dedicated to the chemical protection of berry plantations, and, in particular, blackcurrants. The prototype was set up with two configurations: “offset”, in which the opposing air streams were “offset” by 0.5 m, and “face-to-face”, when they were positioned opposite each other. The field experiments were carried out on a blackcurrant plantation (Tisel cv.; bush spacing of 4.0 × 0.5 m; height 1.2 m; width 2.5 m). The spray deposition within the crop canopies as well as spray drift to the air and to the ground were assessed using the fluorescence method in order to compare the quality of treatments performed with the two-row sprayer and a conventional axial fan sprayer with radial air discharge system. Spray applications were performed at spray volume 300 L∙ha⁻¹ and working speed 6 km h⁻¹ by both sprayers. The plantation was sprayed with 0.25% water solution of a fluorescent tracer BF7G. The in-canopy spray deposit and spray drift were evaluated using artificial targets made of filter paper. Although directed air-jet sprayer in two configurations (“offset” and “face-to-face”) and conventional one produced similar deposits within the bushes, the spray loss from the directed air-jet sprayer was considerably lower (25.1–32.2%) than that from the conventional sprayer (76.9–81.8%) generating considerably greater airflow volume. Lower PPP losses mean lower environmental impact, which is in line with integrated plant protection. The research responds to numerous inquiries from sprayer manufacturers and blackcurrant growers regarding the most appropriate configuration of the air flow outlet planes. The results obtained will contribute to increasing the efficiency of spraying and facilitate the implementation of the European Green Deal and the achievement of the target of a 50% reduction in the use of plant protection products after 2030 in the EU. Full article

Pedestrian detection, as a core function of an intelligent vision system, plays a key role in obstacle avoidance during driverless navigation, intelligent traffic monitoring, and other fields. In this paper, we optimize the YOLO11 detection algorithm to solve the problem of insufficient accuracy of pedestrian detection in complex scenes. The C3K2-lighter module is constructed by replacing the Bottleneck in the C3K2 module with the FasterNet Block, which significantly enhances feature extraction for long-distance pedestrians in dense scenes. In addition, it incorporates the Triplet Attention Module to establish correlations between local features and the global context, thereby effectively mitigating omission problems caused by occlusion. The Variable Focus Loss Function (VFL) is additionally introduced to optimize target classification by quantifying the variance in features between the predicted frame and the ground-truth frame. The improved model, YOLO11-Improved, achieves a synergistic optimization of detection accuracy and computational efficiency, increasing the AP value by 3.7% and the precision by 2.8% and reducing the parameter volume by 0.5 M while maintaining real-time performance. Full article

The conversion of wetlands into croplands often leads to significant losses of peat soil salinity and soil organic matter (SOM), though quantifying these changes is challenging due to limited historical data. In this study, we compared current soil physicochemical properties with rare historical data from the Mezzano Lowland (ML) in Northeastern Italy, a former wetland drained over 60 years ago. The transformation, which affected approximately 18,100 hectares, was achieved through the construction of a network of drainage canals and pumping stations capable of removing large volumes of water, enabling intensive agricultural use. Results showed a marked decrease in electrical conductivity (EC) and sulphate concentration, indicating extensive salt leaching from the upper peat soil layers. EC dropped from historical values up to 196 $\mathrm{m}$$\mathrm{S}$/cm (1967–1968) to a current maximum of 4.93 $\mathrm{m}$$\mathrm{S}$/cm, while sulphate levels declined by over 90%. SOM also showed significant depletion, especially in deeper layers (50–100 cm), with losses ranging from 50 to 60 $\mathrm{wt}\%$, due to increased aeration and microbial activity post-drainage. These climatic and environmental changes, including a marked reduction in soil salinity and sulphate concentrations due to prolonged leaching, have likely shifted the Mezzano Lowland from a carbon sink to a net source of CO₂ and CH₄ by promoting microbial processes that enhance methane production under anaerobic conditions. To detect residual peat layers, we used Ground-Penetrating Radar (GPR), which, combined with soil sampling, proved effective for tracking long-term peat soil changes. This approach can inform sustainable land management strategies to prevent further carbon loss and maintain peat soil stability. Full article

Ground subsidence is a significant geotechnical hazard in urban areas, leading to property damage, casualties, and broader societal issues. This study investigates the mechanisms of cavity formation and ground subsidence through laboratory model tests using Korean standard sand and marine clay under controlled conditions. A transparent soil box apparatus was fabricated to simulate sewer pipe damage, with model grounds prepared at various relative densities, groundwater levels, and fines contents. The progression of cavity formation and surface collapse was observed and quantitatively analyzed by measuring the time to cavity formation and ground subsidence, as well as the mass of discharged soil. Results indicate that lower relative density accelerates ground subsidence, whereas higher density increases cavity volume due to greater frictional resistance. Notably, as the fines content increased, a tendency was observed for ground subsidence to be increasingly suppressed, suggesting that cohesive clay particles can limit soil loss under seepage conditions. These findings provide valuable insights for selecting backfill materials and managing subsurface conditions to mitigate ground subsidence risks in urban infrastructure. Full article

Windthrow determines major changes in tree stand evolution due to the felling or breaking of either isolated trees or entire stands. It has a major ecological, social and economic impact. Wood loss resulting from tree felling and cross-cutting operations is a less-studied aspect related to windthrow. Wood loss is represented by high stumps, broken or split stems, wood lost in the felling of trees that remain standing, wood lost in felling cuts that attempt to remove the stem from the root plate of partially or totally uprooted trees and wood lost as a result of stem cross-cutting. The study focused on estimating losses and their indices in a spruce tree stand located in the Curvature Carpathians. Windthrow took place in this tree stand in February 2020. The results showed that the total wood loss index is 7.747%. The main losses are represented by wood losses in high stumps (5.319%). The amount of wood loss depends on the proportion of uprooted or standing trees, ground inclination and the uprooting direction of trees as opposed to ground inclination, as well as on tree dimension. Tree volume significantly influences wood loss in high stumps (p < 0.001). The closer the uprooting direction is to the highest slope, the higher the tree stump becomes. Wood loss caused by stem breaking and splitting represents 2.280%, tree felling cuttings and stem removal from the root plate in uprooted trees account for 0.138% while loss resulting from stem cross-cutting represents 0.10%. Full article

Conductive polymer composites (CPCs) are widely used in flexible electronics due to their tunable electrical properties and mechanical deformability. However, accurately predicting the evolution of conductive networks, particularly under compressive strain, remains a significant challenge. In this study, we developed a statistical mechanics model and an extended dynamic statistical mechanics model to quantitatively describe percolation behavior in CPCs. The static model incorporates filler geometry, aspect ratio (AR), and surface-to-volume ratio, and was validated using Monte Carlo simulations. Results show that the percolation threshold for spherical fillers was 0.11965, while significantly lower values of 0.00669 and 0.00203 were observed for plate- and rod-shaped fillers, respectively, confirming the enhanced connectivity of anisotropic particles. To capture strain-dependent behavior, a dynamic model was constructed using a Smoluchowski-type gain–loss framework. This model separates conductive network formation (gain) from network disconnection (loss) caused by filler alignment and Poisson-induced expansion. At high Poisson’s ratios (0.3 and 0.5), the model accurately predicted the reduction in connectivity, particularly for anisotropic fillers. Across all tested conditions, the model exhibited strong agreement with simulation data, with RMSE values ranging from 0.0004 to 0.0449. The results confirm that high AR fillers enhance conductivity under compression, while large Poisson’s ratios suppress network formation. These findings provide a reliable, physically grounded modeling framework for designing strain-sensitive devices such as flexible pressure sensors. Full article

Recent advancements in space exploration have significantly increased the volume of astronomical data, heightening the demand for efficient analytical methods. Concurrently, the considerable energy consumption of machine learning (ML) has fostered the emergence of Green AI, emphasizing sustainable, energy-efficient computational practices. We introduce the first large-scale Green AI benchmark for galaxy morphology classification, evaluating over 30 machine learning architectures (classical, ensemble, deep, and hybrid) on CPU and GPU platforms using a balanced subset of the Galaxy Zoo dataset. Beyond traditional metrics (precision, recall, and F1-score), we quantify inference latency, energy consumption, and carbon-equivalent emissions to derive an integrated EcoScore that captures the trade-off between predictive performance and environmental impact. Our results reveal that a GPU-optimized multilayer perceptron achieves state-of-the-art accuracy of 98% while emitting 20× less CO₂ than ensemble forests, which—despite comparable accuracy—incur substantially higher energy costs. We demonstrate that hardware–algorithm co-design, model sparsification, and careful hyperparameter tuning can reduce carbon footprints by over 90% with negligible loss in classification quality. These findings provide actionable guidelines for deploying energy-efficient, high-fidelity models in both ground-based data centers and onboard space observatories, paving the way for truly sustainable, large-scale astronomical data analysis. Full article

The Tunnel Boring Machine (TBM) method has gained attention as an eco-friendly tunneling technique, effectively reducing noise, vibration, and carbon emissions compared to conventional blasting methods. However, ground settlement and volume loss are inevitable during TBM excavation due to the deformation of the surrounding ground, which may even lead to ground collapse in severe cases. In this study, a Shield TBM model, validated using field data, was employed to perform numerical analyses on parameters such as tunnel diameter, ground elastic modulus, face pressure, and backfill pressure. Based on the simulation results, the influence of each parameter on settlement was evaluated, and a predictive model for estimating maximum settlement was developed. The proposed model was statistically validated using p-value assessment, variance inflation factor (VIF), coefficient of determination (R²), and residual analysis. Furthermore, the prediction model showed high agreement with the field data, yielding a prediction error of 8.25%. This study emphasizes the applicability of verified numerical modeling for accurately predicting ground settlement in Shield TBM tunneling and provides a reliable approach for settlement prediction under varying construction conditions. Full article