Search Results (44,074)

Soil organic carbon fractions and pool management indices are critical for the ecosystem function of bamboo forests; however, their response to varying degrees of expansion of Chimonobambusa utilis (EDCU) and altitudinal gradients remains poorly understood in high-altitude karst regions. In this study, 225 samples (three replicate soil samples, each with five duplicate samples) were collected from 45 typical soil sites in the Jinfo high-altitude karst mountains, China. This study investigated the effects of three EDCUs (low, moderate, and high expansion) and five altitudinal gradients (1300–1500 m, 1500–1700 m, 1700–900 m, 1900–2100 m, and 2100–2300 m) on root elemental composition, soil properties, soil organic fractions, and pool management indices. The results revealed that root total C, N, R_C:P, and R_N:P decreased with increasing altitude, whereas root total C, N, P, and R_C:N also increased significantly with increasing EDCU. Compared with those at low and moderate EDCU, the POC:SOC (34.12%), HFOC (32.73 g kg⁻¹), and HFOC:SOC (37.07%) ratios were highest at high EDCU along the altitudinal gradient of 1700–1900 m. Meanwhile, the L (2.38), LI (2.01), and CMI (174.55) ratios reached their highest values at moderate expansion degrees of Chimonobambusa utilis within the altitudinal gradient of 1900–2100 m. Moreover, redundancy discriminant analysis (RDA) and structural equation modeling (SEM) revealed that the soil carbon pool management index was significantly positively associated with soil properties through direct pathways and negatively correlated with root elemental composition through indirect pathways. In general, the quality of the carbon pool in Chimonobambusa utilis is optimal within the moderate expansion degrees of Chimonobambusa utilis within the altitudinal gradient of 1900–2100 m. The findings of this study establish a theoretical basis for the expansion of Chimonobambusa utilis in high-altitude karst regions and provide scientific evidence to support the increase in the carbon sequestration capacity of bamboo forest ecosystems in these mountainous areas. Full article

Experiential education that connects students with nature and well-being offers a powerful approach to advance sustainability education. Beyond individual benefits, cultivating meaningful human–nature relationships is foundational to fostering environmental stewardship—an increasingly urgent global priority. Universities can play a critical role in preparing students for both professional success and civic, social, and environmental responsibility. This exploratory study examined which components of an experiential learning course most strongly influenced students’ understanding of nature as an asset for their well-being. The course, delivered at a satellite mountain campus of a U.S. university, incorporated Kolb’s stages of experiential learning through forest bathing, reflective journaling, and group outdoor activities. Semi-structured interviews with participants revealed that the coupling of course content with direct experiences in nature, learning alongside peers, and limited technology use were among the most impactful elements. These findings demonstrate that experiential learning environments that intentionally align theory with experience—and situate students in immersive, socially rich, and technology-limited settings—can deepen personal well-being and sustainability understanding. Higher education should embrace nature-based experiential learning to prepare environmentally responsible, critically reflective, and socially connected graduates capable of contributing to a more sustainable future. Full article

Rapid urbanization has intensified resource and population agglomeration while exacerbating urban-rural disparities. To address the long-standing dual structure, China advocates urban-rural integrated development (URID) to achieve common prosperity. However, the long-term evolutionary patterns and explanatory mechanisms of URID remain insufficiently explored, particularly at the county (district)-level in western China. This study constructed an entropy-weighted TOPSIS evaluation system combined with kernel density estimation and an optimal parameters-based geographical detector (OPGD) model to analyze the spatiotemporal evolution and explanatory mechanisms of URID in 26 counties (districts) of the Xi’an metropolitan area from 2010 to 2022. The results showed that: (1) URID levels increased steadily over the study period, forming a pronounced core-periphery gradient with faster improvement in national URID pilot counties. (2) Factor associations evolved from being dominated by a few dimensions to multidimensional coupling. Socioeconomic and geographical factors remained dominant and relatively stable, demographic influences were clearly stage specific, and the interaction between forest coverage and economic variables weakened over time. (3) Enhancing regional transport accessibility, optimizing land use efficiency, and fostering positive population-industry interaction are key pathways for promoting URID in the study area. Methodologically, this study introduces a “significance testing followed by threshold verification” logic into the OPGD model, refining the parameter-setting process and improving the robustness and q-value of factor detection. The findings enrich URID theory, provide county (district)-scale evidence for western China, and offer policy implications for optimizing factor allocation and promoting coordinated regional development. Full article

In the context of evolving antifungal resistance and increasing reports of clinical outbreaks of non-albicans Candida spp. invasive infections, the rapid detection of resistant patterns is of the utmost importance. Currently, an azole-resistant Candida parapsilosis clinical outbreak is ongoing at Pisa University Hospital. Resistant isolates bear both Y132F and S862C amino acid substitutions. Based on the data and isolates retrieved during the clinical outbreak, mass spectrometry was used to investigate the differences between fluconazole-resistant and -susceptible clinical strains directly from yeast colonies isolated from agar culture media. A total of 39 isolates, 16 susceptible and 23 resistant, were included. Spectra were processed following a standardized pipeline. Several supervised machine learning classifiers such as Random Forest, Light Gradient Boosting Machine, and Support Vector Machine, with and without principal component analysis were implemented to discriminate resistant from susceptible isolates. Support Vector Machine with principal component analysis showed the highest sensitivity in detecting fluconazole resistance (100%). Despite these promising results, external prospective validation of the algorithm with a higher number of clinical isolates retrieved from multiple clinical centers is required. Full article

Valued above all others, the white truffle species (Tuber magnatum Picco) is highly dependent on the forest ecosystem and its underground biology. Despite its economic importance, knowledge of its biology and mycorrhizal symbioses remains limited; moreover, natural yields have sharply declined, and cultivation efforts have produced inconsistent results. This study evaluated various forest and mycorrhizal inoculation techniques to promote T. magnatum mycelium development in three Tuscan sites converted to truffle cultivation, using qPCR analysis. Alongside conventional practices like irrigation, mulching, and tillage, an experimental method with a sterile, spore-inoculated soil barrier was tested to improve host root establishment, enhance mycorrhization, and maintain long-term symbiosis for healthy truffle ecosystems. Soil analyses nine months after planting Quercus robur L. seedlings showed significant differences in Tuber magnatum mycelium abundance across sites and treatments. The MA treatment—mycorrhized seedlings combined with a sterile, inoculated substrate and separation diaphragm—produced the highest mycelial levels, underscoring the importance of initial mycorrhization and soil manipulation. These findings provide valuable insights for optimizing forest management and improving truffle cultivation by enhancing mycelial development, a key step toward increasing truffle production. Full article

Saproxylic insects are key forest components but highly vulnerable to practices that reduce deadwood quality and diversity. We investigated the response of Platycerus caraboides (Coleoptera: Lucanidae)—an elusive, cool-adapted stag beetle associated with moist, small-diameter decayed wood—to forest coppicing in the Euganean Hills (northeastern Italy). Surveys were conducted both before (2017–2020) and after coppicing (2021–2025) to compare plots that had undergone coppicing with those that remained uncoppiced. Field investigation focused on the volume of downed woody debris and on P. caraboides occurrence, quantified as encounter rates of deadwood bearing stag beetle oviposition scars or the evidence of stag beetle presence. Coppicing and wood harvesting reduced the overall volume of deadwood but did not significantly affect the amount of small-diameter downed woody debris. Nevertheless, P. caraboides showed consistently lower encounter rates in coppiced areas, particularly during the initial survey period, suggesting that altered microclimatic conditions and reduced debris quality may hinder colonization or larval development. These findings underscore the need to retain small- and medium-diameter woody debris, maintain partial canopy cover, and enhance structural heterogeneity to conserve overlooked, cool-adapted saproxylic species—especially under climate change and in line with EU biodiversity restoration goals. Full article

Conservation of the gallery forest in Berenty Reserve is becoming increasingly urgent. Any deterioration threatens its increasingly rare lemur species. Following a trial planting programme started in 2016 on three plots, with measurement of seedling growth in 2017 and 2018, we returned in 2025 to measure the changes in height, canopy cover and stem diameter. Key insights were that growth had accelerated markedly after 2018. Trees in the forest can be divided into three main species groups—upper canopy, lower canopy and dryland species—but we found scant relationship between species growth and their eventual canopy height, which could have consequences for future planting schemes and management. The plots in the mid-forest showed the highest growth rates. Mortality of seedlings was highest on the riverside plot, but there was also wild recruitment from the forest. The plots by the river and in the mid-forest received the largest number of recruits. The chief problem for the study was that we were only in Berenty for short periods and could not oversee ongoing activities in the plant nursery and in the forest. Consequently, there were problems arising from nursery treatment, unrecorded replanting and difficulties tracking the growth of individuals across years. Future work, based on our results, will focus on identifying and planting species best suited for recovery on the varied sites. Overall, temporal depth is essential for making appropriate restoration decisions based on long-term ecological functioning. Full article

The development of rapid and intelligent methods is urgently needed for wheat quality evaluation. Using the prediction of wet gluten content as a case study, this work systematically investigated the performance of various machine learning algorithms and their optimization for content prediction, based on hyperspectral data from the visible and near-infrared ranges of wheat grains and flour. The results revealed that the random forest regression (RFR) algorithm delivered the best predictive performance under two conditions: first, when applied directly to visible spectra; and second, when applied to fused visible and near-infrared spectral data. This held true for both grains and flour. Conversely, its direct application to NIR spectra alone yielded relatively worse performance. Following data optimization, the first-derivative (FD) visible spectra of wheat grains were smoothed using a Savitzky–Golay (SG) filter and subsequently used as input for the RFR model. This optimized approach achieved a coefficient of determination (r²) of 0.8579, a root mean square error (RMSE) of 0.0216, and a relative percent deviation (RPD) of 2.6978. Under the same conditions, for wheat flour, the corresponding values were 0.8383, 0.0231, and 2.5293, respectively. Similarly, for wheat flour, the RFR model was applied to the SG-filtered FD spectra derived from the fused visible and near-infrared data, yielding an r² of 0.8474, an RMSE of 0.0224, and an RPD of 2.6034. Under the same conditions, wheat grains yielded an r² of 0.8494, an RMSE of 0.0223, and an RPD of 2.6208. This efficient and rapid intelligent prediction scheme demonstrates considerable potential for the quality assessment and control of relevant food products. Full article

The expansion of electronic health records (EHRs) has generated a large amount of unstructured textual data, such as clinical notes and medical reports, which contain diagnostic and prognostic information. Effective classification of these textual medical notes is critical for improving clinical decision support and healthcare data management. This study presents a statistically rigorous comparative analysis of four traditional machine learning algorithms—Random Forest, Logistic Regression, Multinomial Naive Bayes, and Support Vector Machine—for multiclass classification of medical notes into four disease categories: Neoplasms, Digestive System Diseases, Nervous System Diseases, and Cardiovascular Diseases. A dataset containing 9633 labeled medical notes was preprocessed through text cleaning, lemmatization, stop-word removal, and vectorization using term frequency-inverse document frequency (TF–IDF) representation. The models were trained and optimized through GridSearchCV with 5-fold cross-validation and evaluated across five independent stratified 90-10 train–test splits. Evaluation metrics, including accuracy, precision, recall, F1-score, and multiclass ROC-AUC, were used to assess model performance. Logistic Regression demonstrated the strongest overall performance, achieving an average accuracy of 0.8469 and high macro and weighted F1 scores, followed by Support Vector Machine and Multinomial Naive Bayes. Misclassification patterns revealed substantial lexical overlap between digestive and neurological disease notes, underscoring the limitations of TF–IDF representations in capturing deeper semantic distinctions. These findings confirm that traditional machine learning models remain robust, interpretable, and computationally efficient tools for textual medical note classification, and the study establishes a transparent and reproducible benchmark that provides a solid foundation for future methodological advancements in clinical natural language processing. Full article

This paper examines the combination of traditional banking credit assessment techniques with contemporary internal sales accounting systems in Kazakhstan, aiming to augment the precision and resilience of financial assessments pertaining to SMEs. The proposed model consists of two discrete components: a traditional credit scoring module that employs logistic regression and a supplementary sales analytics module that leverages ensemble machine learning methodologies — random forests and gradient boosting algorithms. The outputs generated by these components are amalgamated through an ensemble strategy, where optimal weighting coefficients are ascertained via cross-validation. An empirical analysis was conducted on a dataset encompassing 41,000 SME records from a prominent Kazakhstan bank alongside daily transactional sales data from 150 SMEs gathered between the years 2021 and 2024. The integrated hybrid model demonstrated a statistically meaningful enhancement in predictive efficacy, as evidenced by an increase in the area under the ROC curve from 0.76 to 0.87 and a decrease in mean squared error from 0.12 to 0.08 relative to the traditional methodology. The investigation delves into the transformative influence of digitalization on innovation within SMEs, elucidating that improved real-time data integration not only sharpens risk assessment processes but also promotes adaptive lending strategies and operational efficiencies. Full article

Timely and accurate urban change detection is vital for sustainable urban development, infrastructure management, and disaster response. Traditional two-dimensional approaches often overlook vertical and structural variations in dense urban areas. This study proposes a three-dimensional (3D) change detection framework that integrates high-resolution optical imagery and Digital Surface Models (DSMs) from two time points to capture both horizontal and vertical transformations. The method is based on a deep learning architecture combining a ResNet34 encoder with a UNet++ decoder, enabling the joint learning of spectral and elevation features. The research was carried out in two stages. First, a binary classification model was trained to detect areas of change and no-change, allowing direct comparison with conventional methods such as Principal Component Analysis (PCA), Change Vector Analysis (CVA) with thresholding, K-Means clustering, and Random Forest classification. In the second stage, a multi-class model was developed to categorize the types of structural changes, including new building construction, complete destruction, building height increase, and height decrease. Experiments conducted on a high-resolution urban dataset demonstrated that the proposed CNN-based framework significantly outperformed traditional methods, achieving an overall accuracy of 96.58%, an F1-score of 96.58%, and a recall of 96.7%. Incorporating DSM data notably improved sensitivity to elevation-related changes. Overall, the ResNet34–UNet++ architecture offers a robust and accurate solution for 3D urban change detection, supporting more effective urban monitoring and planning. Full article

Soil-borne pathogens cause devastating root rot diseases in forest ecosystems, often by inducing dysbiosis in the rhizosphere microbiome. While antagonistic bacteria can suppress disease, their effects frequently extend beyond direct inhibition to include ecological restructuring of resident microbial communities. However, the causal relationships between such microbiome restructuring and disease suppression in tree species remain poorly understood. Here, we show that the antagonistic bacterium B. mojavensis dxk33 effectively suppresses F. solani-induced root rot in C. lanceolata, and that this disease suppression coincides with a partial reversal of pathogen-associated dysbiosis in the rhizosphere. Inoculation with dxk33 significantly promoted plant growth and reduced the disease index by 72.19%, while concurrently enhancing soil nutrient availability and key C-, N- and P-cycling enzyme activities. High-throughput sequencing revealed that dxk33 inoculation substantially reshaped the rhizosphere microbiome, counteracting the pathogen’s negative impact on microbial diversity and coinciding with a shift toward a more stable community structure. Under pathogen stress, dxk33 enriched beneficial bacterial taxa such as Pseudomonas and Sphingomonas and suppressed pathogenic fungi while promoting beneficial fungi such as Mortierella. Linear discriminant analysis and functional prediction further indicated that dxk33 remodeled ecological guilds enriched for mycorrhizal and saprotrophic fungi, and reactivated bacterial metabolic pathways and signaling networks that were suppressed by the pathogen. Taken together, our findings are consistent with a multi-tiered mode of action in which direct antagonism by B. mojavensis dxk33 operates alongside associated changes in the rhizosphere microbiome that resemble a disease-suppressive state, although the present experimental design does not allow a strictly causal role for microbiome reconfiguration in disease suppression to be established. This study provides a mechanistic framework for understanding how microbiome engineering may mitigate soil-borne diseases in perennial trees and highlights the potential of targeted microbial interventions for sustainable forest management. Full article

The species, quantity, and tree diameter at breast height (DBH) are important indicators for assessing species distribution, individual growth status, and overall health in the forest. The existing tree information collection mainly relies on manual labor, which results in low efficiency and high labor intensity. To address these issues, we propose a method for tree species identification and diameter estimation by combining deep learning algorithms with binocular vision. First, an image acquisition platform is designed and integrated with a weeding machine to capture images during weeding operation. Images of seven types of trees are captured to develop a dataset. Second, a tree species identification model is established based on the YOLOv8n network, achieving 98.5% accuracy, 99.0% recall, and 99.2% mAP. Then, an improved YOLOv8n-seg model is proposed. It simplifies the network by introducing VanillaBlock in the backbone. FasterNet with a CCFM structure is added at the neck to enhance the model’s multi-scale expression capability. The mIoU of the improved model is 93.7%. Finally, the improved YOLOv8n-seg model is combined with binocular vision. After obtaining the segmentation mask of the tree, the spatial position of the two measurement points is calculated, allowing for the measurement of tree diameter. Verification experiments show that the average error for tree diameter ranges from 4.40~6.40 mm, and the proposed error compensation method can reduce diameter errors. This study provides a theoretical foundation and technical support for intelligent collection of tree information. Full article

Mediterranean dehesa ecosystems are highly valuable agroforestry systems from ecological, social and economic perspectives. Their structural characterization has traditionally relied on resource-intensive field inventories. This study assesses the applicability of low-density airborne LiDAR data from the Spanish National Aerial Orthophotography Plan (PNOA) for the automated morphological characterization of Quercus ilex dehesas. This novel workflow integrates the DBSCAN clustering algorithm for unsupervised segmentation of tree vegetation units and Concaveman for crown perimeter delineation and slicing using concave hulls. The technique was applied over 116 hectares in Santibáñez el Bajo (Cáceres), identifying 1254 vegetation units with 99.8% precision, 97.3% recall and an F-score of 98.5%. A field validation on 35 trees revealed strong agreement with the LiDAR-derived metrics, including crown diameter (R² = 0.985; bias = −0.96 m) and total height (R² = 0.955; bias = −0.34 m). Crown base height was overestimated (+0.77 m), leading to a 20.9% underestimation of crown volume, which was corrected using a regression model (R² = 0.952). This methodology allows us to produce scalable, fully automated forest inventories across extensive Iberian dehesas with similar structural characteristics using publicly available LiDAR data, even with a six-year acquisition gap. Full article

Sudden cardiac death (SCD) remains a major clinical challenge, with implantable cardioverter-defibrillators (ICDs) serving as the primary preventive intervention. Current patient selection guidelines rely on limited and imperfect risk markers. This study explores the potential of machine learning (ML) models to improve SCD risk prediction using tabular clinical data that include features derived from medical sensing devices such as electrocardiograms (ECGs) and ICDs. Several ML models, including tree-based models, Naive Bayes (NB), logistic regression (LR), and voting classifiers (VC), were trained on demographic, clinical, laboratory, and device-derived variables from patients who underwent ICD implantation at a Croatian tertiary center. The target variable was the activation of the ICD device (appropriate or inappropriate/missed), serving as a surrogate for high-risk SCD detection. Models were optimized for the F2-score to prioritize high-risk patient detection, and interpretability was achieved with post hoc SHAP value analysis, which confirmed known and revealed additional potential SCD predictors. The random forest (RF) model achieved the highest F2-score (F2-score 0.74, AUC-ROC 0.73), demonstrating a recall of 97.30% and meeting the primary objective of high true positive detection, while the VC classifier achieved the highest overall discrimination (F2-score 0.71, AUC-ROC 0.76). The predictive performance of multiple ML models, particularly the high recall they achieved, demonstrates the promising potential of ML to refine ICD patient selection. Full article