Search Results (21,819)

Using waste materials in the manufacture of concrete has many environmental advantages. However, it can be difficult to estimate structural performance, especially when beams are reinforced with fiber-reinforced polymers (FRP). In order to provide a data-driven approach to sustainable structural design, this work explores the use of machine learning (ML) approaches to forecast the flexural strength of FRP-strengthened waste aggregate concrete beams. A total number of 92 experimental datasets were used to develop and assess four ML algorithms: Random Forest (RF), Decision Tree (DT), Neural Network (NN), and Extreme Gradient Boosting (XGBoost). Regression plots, Taylor diagrams, statistical measures (R2R^2R2, RMSE, MAE, MSE), and explainable AI (XAI) tools, including SHAP, LIME, and partial dependence plots (PDPs), were used to evaluate the model’s performance. RF outperformed NN in terms of predictive accuracy, while XGBoost exhibited similar performance to RF. The most significant predictors, according to a SHAP analysis, were beam length and fiber length, with the lower followed by steel tensile strength, fiber width, and concrete compressive strength. LIME offered local interpretability for individual predictions, but PDPs demonstrated optimal parameter ranges and a nonlinear feature strength relationship. The findings provide engineers with a strong decision-support tool for designing green infrastructure, since they show that ensemble-based models can accurately represent the intricate, nonlinear dynamics controlling flexural behavior in sustainable FRP-strengthened waste aggregate concrete beams. Full article

Safe and sustainable water sources are a serious global concern because of growing population, urbanization, industrialization, and climate change. The conventional water surveillance systems that rely on periodic sampling and laboratory analysis fail to provide time-sensitive and high-resolution data utilized for proactive water management. Artificial Intelligence of Things (AIoT) offers a viable solution, as they can provide tools of constant active monitoring and predictive analytics. The integration of IoT sensor networks with machine learning (ML) methods enables real-time data-driven water resource monitoring and intelligent decision-making, enhances water quality assessment, supports early detection of anomalies, improves predictive capabilities for floods and droughts, and facilitates efficient irrigation and reservoir management, ultimately leading to sustainable and resilient water management systems. The paper presents an extensive overview of AIoT solutions for water quality monitoring and water resource management, including IoT sensor networks for real-time data acquisition, machine learning methods for prediction, classification, anomaly detection, and edge computing platforms for data processing and decision support. This study also highlights existing possibilities, obstacles, and research gaps identified through a review of the recent literature. Key challenges reported across multiple studies include limited data availability, sensor calibration bias, integration of heterogeneous data, and insufficient model interpretability. Advanced paradigms such as digital twin systems, TinyML, federated learning, and explainable AI (XAI) are examined as enabling technologies to enhance system efficiency, flexibility, and transparency. Future research directions are outlined to develop scalable, interpretable, and real-time water management solutions. Full article

Upper troposphere (UT) humidity records are crucial for climate studies. To maximize temporal representativeness and enhance the lidar signal, pseudo-monthly averaging—limited to nighttime measurement—is applied, yielding water vapor mixing ratio (WVMR) profiles up to 16 km. This study evaluates 11 years (2013–2023) of WVMR profiles from a UV Raman lidar (Li1200) at Réunion Island, comparing them with MLS-Aura satellite retrievals, ERA5 reanalysis data, and GRUAN-processed M10 radiosondes. The results reveal a systematic dry shift in MLS of up to 30% above 12 km, particularly during the wet season. The lidar exhibits a slight downward shift in WVMR, approximately 5% lower than ERA5 throughout the UT, with the largest deviations occurring above 14 km and greater variability during the wet season. Calibration-related challenges during the dry season result in lidar WVMR profiles that are up to 10% drier than ERA5. Additionally, comparisons with GRUAN-processed radiosondes show a substantial dry shift relative to the lidar, exceeding 30% above 12 km. We investigate the effect of GNSS-based lidar calibration by applying an alternative calibration method, which produces higher WVMR values. This reveals a dry shift in ERA5 relative to the lidar, increasing with altitude in the UT up to 25%. These measurements contribute to the global effort to monitor and validate tropical and subtropical upper tropospheric humidity. Full article

Soil moisture content (SMC) is a critical factor in agricultural management; however, traditional monitoring methods face limitations regarding spatial resolution and the acquisition of regional dynamics. Unmanned Aerial Vehicle (UAV) remote sensing offers new opportunities for precision monitoring. This study proposes a UAV-based multi-modal remote sensing method for soil moisture estimation. Specifically, novel dual-band and three-band hyperspectral (HS) indices were constructed, and visible (RGB) and thermal infrared (TIR) information were integrated to form a multi-modal data system; simultaneously, multi-modal estimation models were developed by combining four AutoML methods: TPOT, AutoGluon, H2O AutoML, and FLAML. The results indicate that the H2O AutoML model, fusing multi-modal data, exhibited the best performance in estimating soil moisture at depths of 0–20 cm and 20–40 cm (R ≥ 0.72, RMSE 1.99–2.17%), demonstrating superior stability and generalization capabilities compared to other models. This study has made progress in hyperspectral index construction, multi-modal fusion, and soil moisture retrieval, providing a new technical approach for the refined management of agricultural water resources. Full article

Drought stress severely limits maize growth and productivity worldwide. In this study, we examined the effects of foliar-applied carbon nanoparticles (CNPs) on morphological and physiological traits in maize plants exposed to drought stress for 25 days. Two maize hybrids, one drought-tolerant (LG 36745 PRO4) and one drought-sensitive (AG 8088 PRO2), were fertilized with 0 or 1.0 mL L⁻¹ of a CNP-based nanofertilizer at the V₂ growth stage and exposed to three drought levels: 0 MPa (control), −0.4 MPa (moderate stress), and −0.8 MPa (severe stress). The experiment followed a 2 × 2 × 3 factorial design (hybrid × CNP treatment × drought level) with four replicates. Results indicated that drought stress adversely affected most morphological and physiological traits, particularly in the drought-sensitive hybrid. However, foliar CNP application significantly alleviated the adverse effects of drought in maize plants under moderate and severe stress, primarily by preserving plant water status, enhancing water use efficiency, carboxylation efficiency, photosynthetic rate, and initial growth in challenging environments. These findings will provide the basis for future research on management practices adopted to control drought and ensure the development of modern and sustainable agriculture. Full article

Industrialization has led to the substantial release of heavy metals and organic pollutants into soil and groundwater, resulting in severe contaminated site issues that pose significant threats to ecosystems and human health. This review aims to systematically review the current development status and challenges of contaminated site remediation technologies, and explore the potential of artificial intelligence (AI) applications in site remediation, to provide a theoretical reference for advancing intelligent remediation. Conventional remediation technologies mainly include physical methods (e.g., solidification/stabilization (S/S), soil vapor extraction (SVE), thermal desorption, pump and treat (P&T), groundwater circulation wells (GCWs)), chemical methods (e.g., chemical oxidation/reduction, electrokinetic remediation (EKR), soil washing), and biological methods (phytoremediation, microbial remediation), along with combined strategies that integrate multiple approaches. Although these technologies have achieved certain successes in engineering practice, they still face common challenges such as risks of secondary pollution, long remediation periods, high costs, poor adaptability to complex hydrogeological conditions, and insufficient long-term stability, making it difficult to fully meet the remediation demands of complex contaminated sites. Subsequently, the potential of emerging technologies—including nanomaterial-based remediation, bioelectrochemical systems, and molecular biology-assisted remediation—is introduced. On this basis, the forefront applications of AI in contaminated site remediation are discussed, covering site monitoring and characterization, risk assessment, remedial strategy selection, process prediction and parameter optimization, material design, and post-remediation intelligent stewardship. Machine learning (ML), explainable AI (XAI), and hybrid modeling approaches have markedly improved remediation efficiency and decision-making. Looking forward, with advancements in XAI, mechanism-data fusion models, and environmental foundation models, AI is poised to drive a paradigm shift toward intelligent and precision remediation. However, challenges related to data quality, model interpretability, and interdisciplinary expertise remain key barriers to overcome. Full article

Background/Objectives: Minor neurocognitive disorder (minor NCD) represents a heterogeneous and potentially modifiable stage along the continuum from normal aging to dementia, offering a critical window for targeted health promotion interventions. Early identification of individuals at increased risk of progression is essential for implementing preventive strategies that may delay functional decline. This study developed a transparent machine learning (ML) framework to predict diagnostic change from minor to major NCD at 12 and 24 months using baseline demographic, clinical, and multidomain neuropsychological data. Methods: A retrospective cohort of 162 memory clinic patients was analyzed using a rigorously controlled pipeline incorporating nested stratified cross-validation, SMOTE-based imbalance correction, and sequential forward feature selection. Logistic regression, support vector machines (SVMs), and XGBoost were evaluated, with SHapley Additive exPlanations (SHAPs) applied to ensure interpretability. Results: SVM achieved the most balanced predictive performance at both 12 months (accuracy = 0.90) and 24 months (accuracy = 0.81). Short-term progression was primarily driven by subtle multidomain cognitive inefficiencies, while longer-term risk reflected continued cognitive vulnerability modulated by metabolic factors such as diabetes. Conclusions: These findings highlight the potential of explainable ML as a health promotion tool and suggest that explainable ML can uncover clinically meaningful cognitive risk signatures at the earliest stages of NCD. By identifying modifiable systemic contributors alongside cognitive risk profiles, this framework supports precision-oriented preventive strategies and proactive longitudinal monitoring in minor NCD. Full article

Accurate crop yield prediction is fundamental to sustainable agricultural management, enabling optimized resource allocation and informed decision-making. However, a critical gap exists in current prediction models: existing approaches overlook the temporal alignment between meteorological conditions and production management phases—defined as the intervals between consecutive agronomic operations (e.g., sowing, fertilization, thinning). This oversight results in suboptimal predictive performance, as conventional whole-season weather aggregation fails to capture phase-sensitive crop–weather interactions. While machine learning (e.g., XGBoost) and deep learning approaches (e.g., CNN, LSTM) have been applied to yield prediction, these models typically treat weather variables as temporally homogeneous inputs, inadequately modeling the correlation between historical yields and phase-specific meteorological patterns. To address this gap, this study proposes CNN-LSTM-AM, an innovative hybrid deep learning model that integrates convolutional neural networks (CNNs), long short-term memory (LSTM), and attention mechanisms (AMs), utilizing weather data explicitly aligned with production management phases as input. The CNN component extracts cross-phase weather patterns, the LSTM captures sequential dependencies across growth stages, and the attention mechanism dynamically weights phase importance based on meteorological conditions. The proposed model is validated using a real-world case study of Bok choy production from an agricultural cooperative in Yunlin County, Taiwan, encompassing 1714 production cycles over eight years (2011–2019). Experimental results demonstrate that CNN-LSTM-AM achieves an RMSE of 1448.24 kg/ha, MAPE of 3.60%, and R² of 0.98, outperforming five baseline models—CNN (RMSE = 2919.18), LSTM (RMSE = 2529.74), CNN-LSTM (RMSE = 1516.44), LSTM-AM (RMSE = 2284.64), and XGBoost (RMSE = 3452.47)—representing a notable reduction in prediction error (58% lower RMSE) compared to XGBoost. Furthermore, prediction accuracy improves progressively as harvest time approaches, and phase-specific weather encoding enhances accuracy by 16.5% compared to whole-season averaging. These findings underscore the critical importance of integrating agronomic domain knowledge into data-driven prediction frameworks. Full article

This paper reviews contemporary developments in systems analysis applied to water resources and agricultural management, highlighting the growing influence of artificial intelligence (AI) and machine learning (ML). The literature in this field encompasses a wide range of approaches, methods, and applications, including hydrological simulation models, decision-support systems, and participatory governance frameworks. In recent years, increasing attention has been devoted to systematically reviewing and categorizing these approaches, particularly in light of rapid advances in AI- and ML-based technologies. The present study focuses on the contributions and impacts of AI and ML on systems analysis methodologies compared with the state of the field approximately a decade ago. By revisiting and classifying key groups of approaches, methods, and software tools, the paper provides an updated overview of the current status of systems analysis in water resources and irrigation management. This overview also serves as a reference framework for assessing future methodological and technological developments. Adopting a systems-thinking perspective, the review spans multiple spatial and management scales, from plot-level irrigation practices to river-basin water allocation. The paper aims to support a more holistic understanding and improved design and evaluation of water–agriculture systems, while also strengthening policy support for sustainable resource management. Finally, it highlights the need for continued interdisciplinary integration, enhanced stakeholder participation, and the development of operational tools capable of translating complex systems insights into actionable water management strategies in the emerging context shaped by AI and ML. Full article

Background/Objectives: Adult spinal deformity (ASD) correction is a complex surgery to restore spinal alignment and relieve patients’ symptoms. Modern techniques and technologies allow for aggressive surgical correction in tissue-friendly ways that preserve anatomy and may enable faster recovery. Robotic-assisted posterior spinal stabilization may be used as an adjunct to complex ASD reconstruction to facilitate a minimally invasive approach, reduce perioperative morbidity and physiological insult, and allow for the performance of procedures traditionally reserved for large academic centers to be effectively performed by qualified surgeons in optimized patients at smaller hospitals with fewer resources. The objective of this study is to assess realignment, perioperative complications, and patient-reported outcomes of complex, minimally invasive, robotic-assisted adult spinal deformity correction in a surgical specialty hospital. Methods: Demographic, surgical, and perioperative data were collected from the medical record. The Oswestry Disability Index (ODI) and Numeric Rating Scale (NRS) for pain scores were collected preoperatively and at regular post-op visits. X-rays were captured preoperatively before hospital discharge and at follow-up visits. Results: Fifty consecutive deformity patients were corrected with a two-stage approach (anterior column reconstruction followed by posterior stabilization with robotic-assisted screw placement on the next day) at a 48-bed (eight operating rooms), surgeon-owned, subspecialty hospital. The average patient age was 70 years, and 64% were female. The average estimated blood loss (EBL) values for the first and second stages were 62 mL and 205 mL, respectively. The average operative time was 172 min during the first stage and 210 min for the second stage. Three interbody spacers (first stage) and 16 screws (second stage) were inserted on average in each procedure. The average length of stay (LOS) in the hospital was 5 days, and the average follow-up period was 10.6 months. No patients required a transfer to another facility with intensive care unit (ICU) capabilities, and none required a revision of hardware placement. There was an average reduction in the lumbar coronal scoliotic curve of 14.5° and an increase in lumbar lordosis of 14.8° at the latest follow-up (p < 0.01). The average mismatch between pelvic incidence and lumbar lordosis (PI-LL) preoperatively was 17.6°, which was reduced to 9.6° at the latest postoperative follow-up (p < 0.01). Mean ODI (%) and NRS scores were significantly improved by 33.8% (46.7 ± 13.3 to 30.9 ± 19.8; p < 0.01) and 55% (6.0 ± 2.2 to 2.7 ± 2.6; p < 0.01), respectively, at last follow-up. Conclusions: This study demonstrates the feasibility of performing complex, robotic-assisted ASD corrective surgery in a surgical specialty hospital, achieving significant correction of sagittal and coronal deformities, relieving patients’ symptoms, and offering efficiency and consistency to pedicle screw placement. This study demonstrates that a minimally invasive approach to complex deformity reconstruction reduces perioperative morbidity with decreased operative times, EBL, and LOS when compared to historic controls. This approach allows for the democratization of deformity care in that procedures typically reserved for large academic centers can be successfully accomplished at smaller institutions in optimized patients by qualified surgeons with appropriate perioperative support staff. Full article

The adoption of instrumented wheelsets on diagnostic trains offers the possibility of continuous monitoring of wheel–rail contact forces. The collection of large datasets can be exploited for diagnostic purposes, aiming to localize specific track defects, allowing significant improvements in terms of safety and maintenance costs. Machine learning (ML) techniques can be used to automate anomaly detection. In this work, the authors compare the application of various ML algorithms based on the identification of different frequency or time-based features of analyzed signals. To perform the activity, a significant number and variety of local defects have been included in the recorded data. From a practical point of view, the insertion of real known defects into an existing line is extremely time-consuming, expensive, and not immune to safety issues. On the other hand, the design of anomaly detection algorithms involves the usage of relatively extended datasets with different faulty conditions. The authors propose deliberately adding real contact force profiles of healthy lines to a mix of synthetic signals, which substantially reproduce the behavior and the variability of foreseen faulty conditions. The results of this work, although preliminary and still to be completed, offer a contribution to the scientific community both in terms of obtained results and adopted methodologies. Full article

Cardiac fibrosis is a hallmark of cardiac aging and a major contributor to development of heart failure. However, therapeutic strategies that specifically target cardiac fibrosis remain limited. In this study, we demonstrate that small-molecule compound ML216 exerts protective effects against aging-associated or β-adrenoceptor agonist isoproterenol-induced cardiac fibrosis in vitro or in vivo. Mechanistically, ML216 inhibits transforming growth factor-β1 (TGF-β1) signaling by reducing TGF-β1 protein levels, thereby attenuating Mothers against decapentaplegic homolog (SMAD) phosphorylation and downstream induction of connective tissue growth factor (CTGF). This leads to a marked suppression of fibrotic genes Col1a1, Cnn2, and Acta2, ultimately resulting in reduced fibrosis. Additionally, the inhibition of the TGF-β1 pathway alleviates cardiomyocytes apoptosis, which may further limit inflammatory responses and contributes to the overall attenuation of cardiac fibrosis. Collectively, these findings demonstrate that ML216 mitigates cardiac fibrosis through the inhibition of TGF-β1 pathway-mediated fibrotic signaling and apoptosis, highlighting its potential as a therapeutic candidate for the treatment of cardiac fibrosis. Full article

Pozzolanic concrete demonstrates intricate, highly nonlinear material interactions that pose significant challenges for the accurate prediction of compressive strength (CS). This study introduces a novel, interpretable ensemble machine learning (ML) framework for predicting CS based on 759 mixture records encompassing cement, aggregates, supplementary cementitious materials (pozzolans), water/binder (W/B), superplasticizer, water, and curing age. Descriptive analysis and ANOVA were used to identify key predictors, followed by an 80/20 train–test split with 10-fold cross-validation to ensure robust and generalizable modeling. To further enhance model reliability, 5% of outliers were removed using an isolation forest algorithm, after which data were normalized and ensemble hyperparameters optimized. Among the evaluated models, the extra trees algorithm with standard scaling demonstrated the most stable generalization, achieving a coefficient of determination (R²) of 0.978 and a root mean square error (RMSE) of 4.197 MPa on the test set, and R² = 0.966 (RMSE = 5.053 MPa) under 10-fold cross-validation. Feature importance, SHAP, and partial dependence analyses consistently demonstrated that W/B, curing age, and cement are the principal determinants of CS. Finally, multi-objective optimization generated high-strength, low-impact mixtures, confirming the framework’s effectiveness as a transparent decision-support tool for performance- and sustainability-oriented pozzolanic concrete design. This study is novel in combining interpretable ensemble ML with multi-objective optimization to simultaneously achieve precise CS prediction and the formulation of sustainable, performance-optimized pozzolanic concrete mixtures. Full article

Dynamic response measurements support bridge monitoring and structural assessment because they are obtainable under operational loading and are sensitive to changes in stiffness, boundary conditions, and mass distribution. This article presents a structured scoping review of dynamic-response-based bridge monitoring and assessment. It covers damage-sensitive indicators, stiffness/capacity proxy inference, interpretation under operational and extreme loading, sensing with acquisition (contact, and indirect/drive-by), and data processing, machine learning and digital-twin integration for decision support. Evidence was identified through targeted searches in Scopus and The Lens with duplicate resolution in Zotero. The cited studies are compiled into a traceable evidence inventory linked to method families and decision objectives. The synthesis shows that global modal properties enable change screening but are highly confounded by environmental/operational variability. Localization and state characterization typically require denser or higher-fidelity sensing and signal conditioning. Finally, capacity-related inference using calibrated conversion models or machine learning (ML) surrogates remains context-bounded and validation-dependent. This review provides an end-to-end pipeline, evidence-maturity rubric, and conservative failure-mode checks with escalation logic that tie SHM outputs to inspection and analysis rather than direct condition declarations for bridge owners. This review is intentionally scoped and does not claim PRISMA-style comprehensiveness. Full article

With the explosive growth of the Internet of Things (IoT) and cloud-computing traffic, Intrusion Detection Systems (IDSs) have become a cornerstone of network security. However, modern traffic data often exhibits extreme class imbalance and long-tailed distributions, leading to persistently high miss rates for minority attack categories in Machine Learning (ML)-based IDSs. Conventional oversampling may introduce decision noise, whereas standard Generative Adversarial Networks (GANs) can suffer from training instability and mode collapse when modeling high-dimensional tabular traffic features. To address these challenges, we propose a high-fidelity traffic augmentation framework based on an Adaptive Hybrid-loss Conditional GAN (AH-CGAN). Specifically, AH-CGAN introduces an iteration-dependent adaptive gradient penalty (AGP) schedule to enforce the Lipschitz continuity constraint more effectively during training and incorporates a feature-matching objective to align intermediate critic representations between real and synthetic traffic. Experiments on the CIC-IDS2017 benchmark show that AH-CGAN generates distribution-consistent synthetic samples and that augmentation improves downstream detection across multiple classifiers. In particular, the weighted F1-score of Logistic Regression increases from 0.8237 to 0.8697 (Δ = +0.0460, i.e., +4.6%). Overall, the proposed approach enhances minority coverage in the feature space and can improve class separability, providing a practical solution for long-tailed IDS. Full article