Search Results (398)

The spread of Xylella fastidiosa, a xylem-limited bacterial pathogen, has caused widespread mortality among olive trees in Apulian region, Italy in more than a decade, and represents a significant threat to Mediterranean agroecosystems. To encourage evidence-based containment strategies, we developed a stochastic, spatiotemporal simulation model that represents pathogen transmission at the individual-tree level. This work integrates high-resolution georeferenced olive-tree data and implicitly incorporates vector population dynamics through a tree-specific vulnerability index, which considers local host density and landscape connectivity. Vector dispersal is approximated using a radial transmission kernel, which preserves host–vector spatial interactions while avoiding the explicit modeling of insect trajectories. The system’s spatial structure is additionally formulated as a proximity graph, facilitating network-based analysis of spread pathways. A series of Monte Carlo simulation experiments is employed for calibration against the observed epidemic footprint, while validation utilizes independent infection records and global sensitivity analysis of key parameters. The findings indicate that the model effectively replicates realistic propagation patterns, and its calibrated parameters are consistent with out-of-sample data. This makes it an appropriate exploratory tool for scenario testing, assessing the potential impact of intervention strategies, and offering risk-based decision support for handling Xylella fastidiosa outbreaks. Subsequently, graph centrality metrics are used to identify epidemiologically critical trees that function as transmission bridges, thus representing priority targets for surveillance or removal efforts. Thus, multiple tests have been conducted using betweenness and closeness centrality, while comparing both methods leads to effective node-tree removal decisions. Full article

Thermal stress in hot–humid urban environments constitutes a persistent sustainability challenge, driven by the interaction of extreme temperatures, high atmospheric moisture, and heat-retaining urban surfaces, which collectively intensify outdoor discomfort and increase cooling-energy demand. Within this context, soft landscape systems have gained recognition as nature-based solutions capable of moderating microclimates and enhancing residential livability; however, their systematic prioritization based on integrated sustainability performance remains insufficiently addressed, particularly in Gulf-region residential developments. This study proposes a sustainability-oriented decision-support framework that evaluates and prioritizes soft landscape strategies for thermal comfort enhancement using the Analytic Hierarchy Process (AHP) as the core analytical method. Expert judgments were elicited and structured across five sustainability-driven criteria—shading effectiveness, evapotranspiration potential, airflow facilitation, aesthetic–psychological comfort, and implementation and maintenance cost—and applied to five soft landscape alternatives. To verify the physical plausibility of the expert-derived prioritization, microclimate simulations were conducted using ENVI-met under extreme summer conditions, representing the hottest day of the year, at key diurnal intervals. The results reveal a clear dominance of shading-based mechanisms, with tree canopy systems emerging as the most effective and sustainable intervention due to their superior radiative control, ecological cooling capacity, and perceptual benefits. Simulation outputs confirm that canopy-driven strategies achieve the most substantial reductions in mean radiant temperature during peak thermal stress, while surface-based interventions provide secondary benefits primarily related to diurnal heat dissipation. At peak thermal stress (14:00), Scenario 2 reduced mean radiant temperature (MRT) from 71.69 °C to 54.23 °C (≈24% reduction) and PMV from 7.33 to 5.70 (≈22% reduction) relative to existing conditions. By integrating expert-based multi-criteria evaluation with simulation-based thermal verification, the study advances a robust and transferable framework for climate-responsive residential landscape planning. The findings reposition soft landscape systems as essential climatic infrastructure, offering actionable guidance for enhancing thermal resilience, reducing cooling-energy dependence, and supporting sustainable residential development in hot–humid regions. Full article

Reliable tool-wear monitoring is essential for maintaining machining quality and preventing unscheduled downtime in manufacturing. This investigation presents a sound-based classification framework for identifying wear states in the turning of AISI 316L stainless steel using advanced gradient-boosting models. Acoustic signals were recorded under constant cutting parameters to eliminate process-induced variability, and each recording was divided into standardized 2 s segments. A total of 540 multidomain features—including RMS, ZCR, spectral descriptors, Mel-spectrogram statistics, MFCCs and their derivatives, and discrete wavelet energies—were extracted to capture both stationary and transient characteristics of tool–workpiece interactions. Feature selection was performed using a three-stage pipeline comprising Boruta, LASSO, and SHAP analysis, resulting in a compact subset of highly informative descriptors. LightGBM, XGBoost, and CatBoost classifiers were trained using stratified 10-fold cross-validation across three wear states: Unworn, Slight wear, and Severe wear. LightGBM and XGBoost achieved the best performance, with mean accuracies above 0.96 and strong PRC–AUC and ROC–AUC values (0.98–1.00). Although Slight wear remained the most difficult class due to its transitional acoustic characteristics, all models showed clear separability for Unworn and Severe wear conditions. The results confirm that boosted decision-tree methods combined with SHAP-enhanced feature selection provide an effective, low-cost, and non-contact solution for tool-wear classification in 316L turning. Full article

Background/Objectives: Intermittent fasting (IF) has shown feature effectiveness in reducing blood pressure, highlighting the need for personalized intervention strategies. Methods: To address this, a machine learning framework was developed to predict the likelihood of blood pressure improvement (≥5 mmHg systolic reduction) across different IF and calorie restriction protocols in premenopausal women without diagnosed hypertension. Results: The model achieved 77% accuracy and an AUC of 0.8 in distinguishing responders from non-responders. Logistic regression analysis identified dietary intervention type as the strongest predictor of success, with Intermittent Energy and Carbohydrate Restriction + free Protein and Fat (IECR + FF) and Intermittent Energy and Carbohydrate Restriction + free Protein and Fat (IECR) protocols showing the highest effectiveness (coefficients 0.55 and 0.41 respectively). Decision tree analysis revealed age in years as a critical stratification factor, with younger patients (≤47 years) responding optimally to IECR + FF combinations, while older patients benefited from IECR, Continuous Energy Restriction (CER), or Intermittent Energy Restriction (IER) approaches. Notably, waist-to-hip ratio emerged as the strongest negative predictor, indicating that central adiposity significantly impedes blood pressure improvement regardless of intervention type. Higher baseline HDL positively predicted success, while elevated LDL and the DER diet were associated with poor outcomes. The complementary analytical approaches demonstrated that logistic regression and decision tree methods highlight different aspects of the data, with the former identifying independent linear associations and the latter suggesting potential non-linear interactions and candidate thresholds involving age years, dietary intervention type, baseline blood pressure, and metabolic markers. Conclusions: This exploratory, hypothesis-generating analysis was conducted in a cohort of premenopausal women without diagnosed hypertension and is not intended to inform clinical decision-making. The observed patterns should be interpreted as preliminary and may reflect sample-specific effects or model instability. Confirmation in larger, independent, and more diverse populations is essential before any clinical relevance can be inferred. Full article

Content/Background: Breast cancer is one of the most fatal cancers among women around the globe. The chances of surviving this cancer increase with early tumor detection, which is necessary for effective treatment. Traditional diagnostic techniques are ineffective and time-consuming, and they yield results that may be accurate or inaccurate. Therefore, our primary objective is to determine how a machine learning model can reduce diagnostic errors and provide accurate results. Objective: The main objective of this project is to build an ML-based classification model that can help doctors detect breast cancer early and more accurately. This project also aims to provide an interactive interface for easy access in healthcare settings. Materials/Methods: For this study, twelve machine learning classification algorithms are implemented and tested: Logistic Regression, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree, Random Forest, Gradient Boosting, XGBOOST, Naive Bayes, AdaBoosting, Light GBM, CatBoost, and the Artificial Neural Network (ANN). This study used the Wisconsin Breast Cancer Dataset (WBCD) from the UCI ML Repository. It contains 569 patient samples and 30 features. This dataset has the following features: Radius, Texture, Area, Perimeter, Smoothness, Compactness, Concavity, and Fractional Dimension. The target variable is diagnosis, which is categorized as malignant vs. benign. Results: The fifteen models were analyzed, evaluated, and compared using five performance metrics: Accuracy, Precision, Recall, F1-Score, and AUC-ROC. Among the evaluated models, CatBoost, LoGR, and AdaBoost outperformed the others, with an Accuracy of 97.%, Precision of 97%, Recall of 97%, and AUC-ROC score of 99%. The AUC-ROC is nearly 99%, and the model has a high ability to differentiate between malignant and benign tumors. Full article

The MOPLUS project, funded by the Portuguese Recovery and Resilience Plan (PRR), aims to enhance soil organic matter, soil structure, and water retention in apple orchards located in the “Maçã de Alcobaça” Protected Geographical Indication area through organic fertilization based on locally available livestock effluents, thereby reducing reliance on synthetic fertilizers under Mediterranean climatic conditions. This study evaluated the physiological and biometric responses of apple trees subjected to four fertilization strategies (M1–M4) in three commercial ‘Gala’ orchards in central Portugal over three growing seasons (2023–2025). Measurements included leaf functional traits, gas exchange, chlorophyll fluorescence, spectral indices, vegetative growth, fruit production per tree and mean fruit weight. Interannual climatic variability and orchard-specific conditions were the dominant drivers of tree response, while fertilization effects were smaller and mainly expressed through interactions with year and orchard. When analyzed within the same orchard, fertilization strategies M2 and particularly M3 maintained physiological performance, vegetative growth, and fruit production per tree at levels comparable to full mineral fertilization. Among treatments, M3 showed the most consistent responses across sites and years, indicating that partial mineral substitution with pig slurry can sustain tree functioning while maintaining or enhancing fruit production per tree. The most restrictive strategy (M4) occasionally showed reduced photosynthetic performance under specific orchard–year combinations, suggesting a threshold effect associated with stronger mineral reduction, but without evidence of generalized physiological stress. Overall, these findings demonstrate that partial substitution of mineral fertilizers by organic amendments—especially pig slurry (M3) and, to a lesser extent, composted cattle manure (M2)—is agronomically viable, allowing apple tree performance and productivity to be maintained while enhancing system resilience under Mediterranean climatic variability. These results also provide practical decision support for site-adapted fertilization management in commercial drip-irrigated apple orchards, supporting reduced mineral fertilizer dependence without compromising productivity. Full article

This article proposes an automated approach, based on artificial intelligence techniques, for detecting indicators of depression in texts written in Spanish. Among the main contributions is the construction of a new specialized corpus, supervised by mental health professionals and based on the Beck Depression Inventory. Text processing included linguistic techniques such as lemmatization, stopword removal, and structural transformation using trigrams. As part of the work, SpADE-BERT was designed, a model based on multilingual BERT with a tokenization scheme adapted to incorporate trigrams directly from the input phase. This modification allowed for more robust interaction between the local context and semantic representations. SpADE-BERT was evaluated against multiple approaches reported in the literature, which employ algorithms such as logistic regression, support vector machines, decision trees, and Random Forest with advanced configurations and specialized preprocessing. In all cases, our model showed consistently superior performance on metrics such as precision, recall, and F1-score. The results show that integrating deep language models with adapted tokenization strategies can significantly strengthen the automated identification of linguistic signals associated with depression in Spanish texts. Full article

Public passenger transport systems increasingly face the challenge of balancing economic efficiency with ecological sustainability, reflecting both policy objectives and passenger expectations. This study examines passenger perceptions of the economic and environmental aspects of public transport services and the factors influencing these perceptions, primarily based on survey data collected in Slovakia. The Slovak dataset was analysed using contingency analysis, namely Chi-square tests of independence, contingency coefficients, and sign scheme, and C5.0 decision tree classification models to identify key determinant of behavioural and attitudinal outcomes. In addition, descriptive comparisons with a complementary Polish sample are provided to illustrate potential differences in preference patterns across national contexts, without formal statistical inference. The results identify key socio-demographic and behavioural factors influencing passenger perceptions and usage patterns in Slovakia, while the complementary Polish sample is used to provide contextual descriptive comparison without formal testing. The study enhances scientific understanding of public transport by exploring the interaction between economic efficiency and ecological sustainability of transport services and provides practical recommendations for the strategic management of transport companies, especially in service modernisation, marketing communication, and support for sustainable mobility. The findings are relevant not only to Slovakia but also to broader European discussions on integrating economic and environmental dimensions into public transport development. Full article

This study explores the application of multiple predictive algorithms under general versus route-specialized modeling strategies to estimate passenger boarding demand in public bus transportation systems. Accurate estimation of boarding patterns is essential for optimizing service planning, improving passenger comfort, and enhancing operational efficiency. This research evaluates a range of predictive models to identify the most effective techniques for forecasting demand across different routes and times. Two modeling strategies were implemented: a generalistic approach and a specialized one. The latter was designed to capture route-specific characteristics and variability. A real-world case study from a medium-sized metropolitan region in Brazil was used to assess model performance. Results indicate that ensemble-tree-based models, particularly XGBoost, achieved the highest accuracy and robustness in handling nonlinear relationships and complex interactions within the data. Compared to the generalistic approach, the specialized approach demonstrated superior adaptability and precision, making it especially suitable for long-term and strategic planning applications. It reduced the average RMSE by 19.46% (from 13.84 to 11.15) and the MAE by 17.36% (from 9.60 to 7.93), while increasing the average R² from 0.289 to 0.344. However, these gains came with higher computational demands and mean Forecast Bias (from 0.002 to 0.560), indicating a need for bias correction before operational deployment. The findings highlight the practical value of predictive modeling for transit authorities, enabling data-driven decision making in fleet allocation, route planning, and service frequency adjustment. Moreover, accurate demand forecasting contributes to cost reduction, improved passenger satisfaction, and environmental sustainability through optimized operations. Full article

Affective computing has emerged as a pivotal field in human–computer interaction. Recognizing human emotions through electroencephalogram (EEG) signals can advance our understanding of cognition and support healthcare. This study introduces a novel subject-independent emotion recognition framework by integrating multiple EEG emotion databases (DEAP, MAHNOB HCI-Tagging, DREAMER, AMIGOS and REFED) into a unified dataset. EEG segments were transformed into feature vectors capturing statistical, spectral, and entropy-based measures. Standardized pre-processing, analysis of variance (ANOVA) F-test feature selection, and six machine learning models were applied to the extracted features. Classification models such as Decision Tree, Discriminant Analysis, Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), Naive Bayes, and Artificial Neural Networks (ANN) were considered. Experimental results demonstrate that SVM achieved the best performance for arousal classification (70.43%), while ANN achieved the highest accuracy for valence classification (68.07%), with both models exhibiting strong generalization across subjects. The results highlight the feasibility of developing biomimetic brain–computer interface (BCI) systems for objective assessment of emotional intelligence and its cognitive underpinnings, enabling scalable applications in affective computing and adaptive human–machine interaction. Full article

Trees are known to modify radiation on building façades via shading effects. However, the combined influence of tree morphological traits and street canyon geometry on façade solar exposure remains inadequately quantified. This paper will fill this gap by using an integrated field measurement, ENVI-met simulations and theoretical analysis of an east–west street canyon in Hangzhou, China. We present the stratified cumulative shortwave radiation disturbance (SRD) and the mean value (MSRD) of R as indices for assessing the influence of the tree height (TH), canopy diameter (D_C), leaf area density (LAD), and under-canopy height (UH) on the shortwave radiation profile of the south façade. Using 54 parametrized simulation scenarios, it was found that tree height is the most sensitive parameter to affect MSRD in the 1114 m range, with under-canopy height defining the building layers below. An LAD of 2 m²/m³ will be an optimal shading and daylighting. When discussed in terms of space, a canopy diameter of 5 m and a wall-to-canopy distance of 1 m (D_W-T) provides better shading in asymmetric canyons where the buildings in the south are lower. Further, canyon building height on either side of the canyon is found to be a decisive factor that mediates tree impacts on radiation, which allows specific approaches to greening canyons of diverse kinds. Through this work, there is a theoretical basis for understanding how trees and canyons interact, and this work gives scientific principles for a tree-planting initiative to reduce urban heat islands. Full article

Genome-wide association studies (GWAS) are essential for identifying genomic regions associated with agronomic traits, but Linear Mixed Model (LMM)-based GWAS face challenges in capturing complex gene interactions. This study explores the potential of machine learning (ML) methodologies to enhance marker identification and association modeling in plant breeding. Unlike LMM-based GWAS, ML approaches do not require prior assumptions about marker–phenotype relationships, enabling the detection of epistatic effects and non-linear interactions. The research sought to assess and contrast approaches utilizing ML (Decision Tree—DT; Bagging—BA; Random Forest—RF; Boosting—BO; and Multivariate Adaptive Regression Splines—MARS) and LMM-based GWAS. A simulated F₂ population comprising 1000 individuals was analyzed using 4010 SNP markers and ten traits modeled with epistatic interactions. The simulation included quantitative trait loci (QTL) counts varying between 8 and 240, with heritability levels set at 0.5 and 0.8. These characteristics simulate traits of candidate crops that represent a diverse range of agronomic species, including major cereal crops (e.g., maize and wheat) as well as leguminous crops (e.g., soybean), such as yield, with moderate heritability and a high number of QTLs, and plant height, with high heritability and an average number of QTLs, among others. To validate the simulation findings, the methodologies were further applied to a real Coffea arabica population (n = 195) to identify genomic regions associated with yield, a complex polygenic trait. Results demonstrated a fundamental trade-off between sensitivity and precision. Specifically, for the most complex trait evaluated (240 QTLs under epistatic control), Ensemble methods (Bagging and Random Forest) maintained a Detection Power (DP) exceeding 90%, significantly outperforming state-of-the-art GWAS methods (FarmCPU), which dropped to approximately 30%, and traditional Linear Mixed Models, which failed to detect signals (0%). However, this sensitivity resulted in lower precision for ensembles. In contrast, MARS (Degree 1) and BLINK achieved exceptional Specificity (>99%) and Precision (>90%), effectively minimizing false positives. The real data analysis corroborated these trends: while standard GWAS models failed to detect significant associations, the ML framework successfully prioritized consensus genomic regions harboring functional candidates, such as SWEET sugar transporters and NAC transcription factors. In conclusion, ML Ensembles are recommended for broad exploratory screening to recover missing heritability, while MARS and BLINK are the most effective methods for precise candidate gene validation. Full article

Background/Objectives: Fluid overload in critically ill patients is linked to adverse outcomes. While resuscitation strategies are well established, guidance for the de-resuscitation phase remains limited. This study aimed to identify clinical factors associated with diuretic response and achieving negative fluid balance (FB) in critically ill patients. Methods: We conducted a single-center, retrospective cohort study of ICU patients who received intravenous furosemide between 2017 and 2023. A CHAID (Chi-square Automatic Interaction Detector) decision tree identified clinical variables associated with fluid removal after the first dose, and a mixed-effects model analyzed repeated measurements. Results: The cohort comprised 1764 patients over 6632 ICU days. Mean arterial pressure (MAP) was the strongest predictor of negative FB. MAP ≤ 75 mmHg yielded minimal negative FB (−33 ± 1054 mL/24 h); MAP 75–90 mmHg yielded intermediate negative FB (−467 ± 1140 mL/24 h); and MAP > 90 mmHg produced the greatest negative FB (−899 ± 1415 mL/24 h; p < 0.001). Secondary associations varied by MAP: creatinine at low MAP, blood urea nitrogen at mid-range MAP, and SOFA score at high MAP, all inversely related to negative FB. In mixed-effects analyses, each 1 mmHg MAP increase was associated with 23.3 mL greater fluid removal (p < 0.001). Independent factors linked to reduced negative FB included vasopressor use (noradrenaline), elevated creatinine, and higher SOFA scores. Conclusions: In this cohort, MAP was significantly associated with the likelihood of achieving a negative fluid balance during de-resuscitation. Conversely, vasopressor use, renal dysfunction, and higher illness severity were linked to reduced diuretic responsiveness. These findings support individualized de-resuscitation strategies. Full article

As distributed generators (DGs) and flexible adjustable loads (FALs) further penetrate distribution networks (DNs), to reduce regulation complexity compared with traditional centralized control frameworks, DGs and FALs in DNs should be packed in several clusters to enable their dispatch to become standard in the industry. To mitigate the negative influence of DGs’ and FALs’ spatiotemporal distribution and uncertain output characteristics on dispatch, this paper proposes an intelligent dynamic cluster partitioning strategy for DNs, from which the DN’s resources and loads can be intelligently aggregated, organized, and regulated in a dynamic and optimal way with relatively high implementation efficiency. An environmental model based on the Markov decision process (MDP) technique is first developed for DN cluster partitioning, in which a continuous state space, a discrete action space, and a dispatching performance-oriented reward are designed. Then, a novel random forest Q-learning network (RF-QN) is developed to implement dynamic cluster partitioning by interacting with the proposed environmental model, from which the generalization and robust capability to estimate the Q-function can be improved by taking advantage of combining deep learning and decision trees. Finally, a modified IEEE-33-node system is adopted to verify the effectiveness of the proposed intelligent dynamic cluster partitioning and regulation strategy; the results also indicate that the proposed RF-QN is superior to the traditional deep Q-learning (DQN) model in terms of renewable energy accommodation rate, training efficiency, and portioning and regulation performance. Full article

Background: The COVID-19 pandemic has put other infectious diseases, especially in children, into a new perspective. Our study focuses on two important viral infections: COVID-19 and influenza, which often present with similar clinical symptoms. Taking into consideration that the pathophysiology and systemic impact of the two viruses are distinct, which can lead to measurable differences in laboratory values, this study aimed to analyze laboratory features that differentiate between COVID-19 and influenza virus infections in pediatric patients. Methods: We statistically analyzed the routinely available laboratory data of 98 patients with influenza virus and 78 patients with COVID-19. Afterwards, the classification and regression tree (CART) method was performed to identify specific clinical scenarios, based on multilevel interactions of different features that could assist clinicians in evidence-based differentiation. Results: Significant differences between the two groups were observed in ALT, eosinophils, hemoglobin, and creatinine. Influenza-infected infants presented significantly higher leukocyte, neutrophil, and basophil counts compared to infants infected with COVID-19. Regarding children (over 12 months), significantly lower levels of ALT and eosinophil counts were observed in those with influenza compared to those with COVID-19. Furthermore, the CART decision tree model identified distinct profiles based on a combination of features such as age, leukocytes, lymphocytes, platelets, and neutrophils. Conclusions: After further refinement and application, such machine learning-based, evidence-driven models, considering the large scale of clinical and laboratory variables, might help to improve, support, and sustain healthcare practices. The differential decision tree may contribute to enhanced clinical risk assessment and decision making. Full article