Search Results (4,020)

Background: Glioblastoma (GBM) is the most aggressive primary brain tumor in adults and remains associated with poor prognosis despite multimodal treatment. Ferroptosis, an iron-dependent form of regulated cell death driven by lipid peroxidation and redox imbalance, has recently emerged as a potential therapeutic vulnerability in glioma. This review summarizes current knowledge on the molecular regulation of ferroptosis in glioma and discusses its implications for tumor progression, therapeutic resistance, and translational targeting. Methods: A structured narrative review of the literature was conducted using PubMed/MEDLINE, Scopus, and Web of Science databases. Experimental, translational, and clinically relevant studies investigating ferroptosis-related mechanisms and therapeutic strategies in glioma and GBM were qualitatively analyzed. Results: Ferroptosis in glioma is regulated by interconnected pathways involving iron metabolism, phospholipid remodeling, oxidative stress, and antioxidant defense systems, particularly the SLC7A11–glutathione–GPX4 axis. Additional protective mechanisms mediated by FSP1 and DHODH, together with regulatory networks involving NRF2, ATF4, p53, and hypoxia-related signaling, contribute to adaptive resistance to ferroptosis. Increasing evidence indicates that ferroptosis interacts bidirectionally with the glioma tumor microenvironment and may exert both antitumor and immunosuppressive effects. Preclinical studies further suggest that ferroptosis induction may enhance the efficacy of temozolomide, radiotherapy, and immunotherapy, although clinical translation remains limited by tumor heterogeneity, blood–brain barrier penetration, and resistance mechanisms. Conclusions: Ferroptosis represents a biologically plausible and therapeutically promising target in glioma. Improved understanding of ferroptosis regulation, tumor microenvironment interactions, and biomarker-guided therapeutic strategies may support the future development of more effective treatments for GBM. Full article

Background/Objectives: Survival prediction in non-small cell lung cancer (NSCLC) remains challenging, particularly in limited-sample settings where end-to-end deep learning models may suffer from limited generalization. This study aimed to develop a data-efficient, multimodal, and explainable framework integrating computed tomography (CT)-derived imaging information with clinical variables for NSCLC survival prediction. Methods: CT images, tumor segmentations, and clinical data from the publicly available NSCLC Radiomics (LUNG1) dataset (377 patients) were used. Tumor-focused regions were extracted using segmentation masks, and pretrained RadImageNet-InceptionV3 embeddings were obtained from the largest tumor-containing slice and neighboring-slice summaries. Deep imaging embeddings, engineered imaging features, and clinical variables were fused into a unified tabular representation. To improve robustness under limited-sample conditions, feature blocks were compressed using principal component analysis. CatBoost, XGBoost, and LightGBM models were trained on a development set and evaluated on a strictly held-out final validation set. Results: In three-class survival stratification, assigning censored/non-event patients to the upper survival group produced the strongest ordinal prognostic performance. Under the EX_PLUS_NON_EX_TOP setting, CatBoost achieved the best holdout score-based class C-index of 0.655. In continuous survival regression, LightGBM achieved the best holdout event-patient C-index of 0.576. Clinical variables provided the dominant prognostic signal, while compact deep image embeddings contributed complementary information, particularly in separating short- and long-survival groups. SHAP analysis confirmed contributions from both clinical and image-derived features. Conclusions: The proposed framework provides a proof-of-concept demonstration of a data-efficient and explainable image-to-tabular approach for NSCLC survival prediction under strict internal holdout validation. The results suggest that pretrained CT embeddings, clinical variables, gradient-boosted trees, and SHAP-based interpretation can be combined in a feasible, limited-sample survival modeling pipeline, while external validation remains necessary before clinical translation. Full article

Early stage mineral exploration requires the reliable identification of anomalous gold zones from drillhole geochemistry in data-limited environments. This study applies a machine learning (ML) classification framework to detect anomalous gold zones (Au ≥ 0.68 ppm; 90th percentile) from bulk XRF multielement drillhole geochemistry in a Paleoproterozoic Birimian greenstone belt sulphide-hosted orogenic gold system, West African Craton. A total of 53,126 one-metre diamond core samples from 301 drillholes were preprocessed within a compositional data analysis (CoDA) framework, with Au being explicitly excluded from the centred log-ratio (CLR) transformation to eliminate target–predictor circularity. After Minimum Covariance Determinant (MCD) outlier filtering, 40,385 samples were retained to construct a 19-feature matrix of 10 CLR-transformed elements, 1 rock-type feature, and 8 sulphide–lithology interaction features. Drillhole-based block cross-validation (DH-block CV), validated by an experimental along-hole variogram (practical autocorrelation range ≈ 20 m), ensured spatially honest performance estimates. Four nonlinear classifiers—Random Forest (RF), XGBoost, LightGBM, and Multi-Layer Perceptron (MLP)—were benchmarked against a Logistic Regression (LR) linear baseline. All nonlinear classifiers achieved validation AUC of 0.936–0.938, outperforming LR (AUC = 0.931) with F1-score improvements of +0.09 to +0.11 and precision gains of up to +35 percentage points—directly reducing wasted drill holes in applied exploration. MLP recorded the highest F1-score (0.666) and precision (0.765), and XGBoost the highest recall (0.787). Permutation importance identified S-Ti (ΔAUC = 0.028), S-Fe (0.021), and S-Al (0.013) as the top-ranked features, confirming that sulphide enrichment relative to lithological background is the primary discriminating signal. Partial dependence analysis revealed a threshold-driven non-monotonic Fe dependence at CLR(Fe) ≈ 3, marking the transition from lithological dilutant to sulphide co-indicator—a nonlinear pattern inaccessible to linear classifiers. Full article

Solar power plants are among the most widely used renewable energy sources today. Varying radiation levels from region to region, and similarly varying consumption depending on the user within a given region, make the optimal sizing of these plants challenging. In this study, a machine learning-based surrogate model for the real-time sizing optimization of solar power plants, trained with a completely original dataset, has been developed. In the first stage, 500 different solar power plant installation scenarios were synthetically generated and evaluated in HOMER, and the obtained optimal sizing outputs were used as training targets for the proposed surrogate model rather than real operational data. The results obtained by applying various machine learning methods to the generated dataset are presented comparatively. Among 7 different machine learning models, XGBoost, Gradient Boosting, and LightGBM demonstrated the best performance. The developed model achieved an average R² score of 0.9425 for a total of 3 targets, while target-specific performance showed R² scores of 0.9747 for inverters, 0.9365 for PV panels, and 0.9165 for batteries. This model serves as a computationally efficient surrogate of the HOMER optimization process, enabling high-accuracy real-time predictions while significantly reducing the computational burden associated with intensive mathematical calculations, iterative procedures, and complex search spaces. Full article

Operational landslide susceptibility mapping (LSM) remains challenging in regions with pronounced geo-environmental heterogeneity, where single global models often overlook spatially variable landslide-environment relationships. Northern Guangdong, China, is a typical humid mountainous region where steep terrain, diverse lithology, and highly variable rainfall produce non-stationary landslide controls. To address this challenge, we develop a cluster-informed LSM framework that integrates unsupervised consensus K-means sub-zoning with localized Random Forest (RF) models and SHapley Additive exPlanations (SHAP). We use a harmonized inventory of 1510 landslides (2011–2022), together with twelve 30 m conditioning factors, for model training and validation. Compared with logistic regression, Support Vector Machines (SVM), and Light Gradient Boosting Machine (LightGBM), RF consistently achieves higher accuracy across clusters, and the cluster-wise RF ensemble attains pooled ACC = 0.8212, F1 = 0.8176, and AUC = 0.8956. SHAP highlights both regionally consistent predictors (e.g., NDVI, distance to road) and distinct cluster-specific controls linked to geomorphic and hydrologic settings. The proposed framework enhances predictive accuracy, produces finer susceptibility gradients, and yields better-calibrated probability estimates than a single global model. These results demonstrate that explicitly accounting for geo-environmental heterogeneity can generate interpretable, spatially adaptive susceptibility outputs. By identifying high-risk zones for priority monitoring, land-use regulation, infrastructure protection, and mitigation planning, the proposed framework provides a practical decision-support tool for sustainable mountain development and disaster risk reduction in heterogeneous mountainous regions. Full article

Accurate day-ahead wind speed forecasting is essential for wind power forecasting and electricity market participation under increasing renewable energy penetration. This study proposes a hybrid forecasting framework that combines raw global forecasts from GFS and IFS, the KMA KIM-RDAPS regional forecast, and dynamically downscaled GFS/IFS forecasts generated with alternative boundary-layer physics. Seven forecast members were synthesized using arithmetic averaging, performance-weighted averaging, and LightGBM-based machine learning (ML) regression. The framework was evaluated over Jeju Island, Republic of Korea, using 10 m Automatic Weather Station observations from 2023 to 2024 and 80 m meteorological mast observations from 2023. For the AWS evaluation, 2023 was used for training and validation, and 2024 was reserved for independent testing. The site-specific LightGBM synthesis achieved the most consistent improvement, reducing the median site-wise MAE across 31 AWS sites to 0.90 m s⁻¹, corresponding to a 39.2% improvement relative to the best non-downscaled member and 47.2% relative to the unweighted multi-model mean. In the 80 m mast-based diagnostic assessment, the same approach reduced derived normalized power MAE to 11.4%. These results indicate that ML synthesis of multi-source NWP forecasts can improve day-ahead wind speed and power-oriented forecast information over complex island terrain. Full article

Introduction: Neuroinflammation is a key feature of both Alzheimer’s disease (AD) and glioblastoma, although it leads to different outcomes in each disorder. In AD, chronic activation of microglia and astrocytes by amyloid-β and tau contributes to neuronal injury and cognitive decline. In glioblastoma, tumor cells exploit inflammatory pathways to create an immunosuppressive microenvironment that supports tumor growth. This review compares the shared and distinct neuroinflammatory mechanisms in AD and glioblastoma and highlights their therapeutic relevance. Materials and Methods: This study was conducted as a narrative review based on a PubMed search performed by three reviewers. English-language articles on AD, glioblastoma, and neuroinflammatory pathways were included, covering original studies, reviews, meta-analyses, and experimental and clinical reports. Keywords included neuroinflammation, microglia, astrocytes, tumor-associated macrophages, inflammasomes, NLRP3, NF-κB, HIF-1α, cytokines, blood–brain barrier, and miRNAs. Due to study heterogeneity, findings were synthesized descriptively. Results: AD and glioblastoma share major neuroinflammatory mechanisms, including microglial and astrocytic activation, cytokine signaling, inflammasome activity, blood–brain barrier dysfunction, hypoxia-related changes, and miRNA regulation. In AD, these pathways promote chronic inflammation, synaptic loss, and neurodegeneration, with NLRP3, NF-κB, and M1-like microglial polarization playing central roles. In glioblastoma, similar pathways are redirected toward tumor progression through tumor-associated macrophages, reactive astrocytes, angiogenesis, immune evasion, and therapy resistance. Key overlapping mediators include IL-1β, TNF-α, NF-κB, HIF-1α, GSK-3β, and selected miRNAs. Conclusions: AD and glioblastoma are connected by common neuroinflammatory pathways, but these processes result in neurodegeneration in AD and tumor support in glioblastoma. Understanding these shared and divergent mechanisms may guide the development of biomarkers and targeted therapies focused on microglia, inflammasomes, cytokines, and immune reprogramming in both diseases. Full article

The mean stand diameter at breast height (DBH) is a key indicator of stand structure and productivity and is widely used in forest resource inventory and management planning. When using regional inventory data, nonlinear interactions between plot-level conditions and predictor variables can undermine the stability of traditional empirical equations across varying site qualities and stand densities. To improve the accuracy and robustness of inventory-scale predictions of mean stand DBH, this study utilized data from 854 forest plots and employed stand age, site class index (SCI), and stand density index (SDI) as independent variables. The predictive performance of traditional growth equations, machine learning models (Random Forest, XGBoost, LightGBM, and support vector machine), and deep learning models (MLP and CNN, ResNet, RNN) was systematically compared, and ensemble learning strategies were further applied to optimize model performance. The results indicated that the Weibull model based solely on stand age achieved the best fit (R² = 0.669). Incorporating SCI and SDI greatly improved model explanatory capability with R² rising to 0.838. XGBoost and CNN further improved predictive performance (R² = 0.852 and 0.861, respectively), while the ensemble model exhibited the highest goodness-of-fit (R² = 0.893), outperforming all individual models. Compared with linear regression, machine learning models demonstrated superior predictive capability. A feature importance analysis indicated that stand age, site quality and stand density together drive mean stand DBH prediction, among which stand age and stand structural characteristics are the dominant influencing factors, whereas SCI and SDI have comparatively weaker effects. Overall, the ensemble model substantially enhanced the prediction accuracy of mean DBH in Pinus yunnanensis stands, thereby providing for precision forest management and ecological function assessment. Full article

As the automotive industry undergoes a green transformation, digital upgrading, and increasingly intensive supply chain collaboration, the supply chain finance credit risks faced by small and medium-sized enterprises (SMEs) in the sector exhibit characteristics such as multi-source interaction, nonlinear transmission, and class imbalance. This study uses 210 SMEs in China’s A-share automotive sector from 2020 to 2024 and constructs a credit risk evaluation system covering 56 indicators across the macro environment, financing enterprises, supply chain characteristics, and core enterprise credit support. Methodologically, DE-LightGBM is employed for feature selection to reduce redundancy and noise, while TabPFGen is introduced to generate synthetic risk-class samples. Business logic constraints and a Nearest Neighbor Distance Ratio filtering mechanism are further applied to improve the plausibility and fidelity of generated samples. Empirical results show that the TabPFN model achieves superior predictive performance after feature selection and data augmentation, and the Wilcoxon signed-rank test confirms the effectiveness and stability of sample augmentation. In addition, the ablation experiment demonstrates that green-related features provide significant incremental predictive value for supply chain finance credit risk identification. The proposed framework provides a useful reference for SME credit assessment, risk early warning, and green financial resource allocation in the automotive industry. Full article

Background: As glucagon-like peptide-1 (GLP-1) receptor agonist use expands globally, reports of psychiatric adverse events (AEs) have increased in spontaneous reporting databases. However, which case-level characteristics are associated with these reporting patterns remains insufficiently characterized. This study aimed to characterize case-level features associated with psychiatric AE reporting in GLP-1 receptor agonist users through pharmacovigilance and explainable machine learning methods. Methods: The FDA Adverse Event Reporting System (FAERS) data (2021 Q2–2025 Q3) were analyzed using a comparator-based approach comprising other antidiabetic and anti-obesity agents. Disproportionality analyses (PRR, ROR, and IC) were performed to detect consolidated safety signals at the Preferred Term (PT) level, with additional drug-specific analyses for individual GLP-1 receptor agonists. Three classification models (logistic regression, XGBoost, and LightGBM) were developed, and SHAP values were used to identify case-level features associated with psychiatric AE reporting patterns. Results: A total of 211,195 unique cases were included (111,105 for GLP-1 receptor agonists; 100,090 for comparators). Sixteen PTs met consolidated signal criteria, with suicidal ideation being the most frequently reported (ROR 2.95). Drug-specific analyses indicated that signal magnitudes were consistently higher for semaglutide than tirzepatide. The XGBoost model achieved an AUROC of 0.816. SHAP analysis showed that age ≥65 years had the highest mean |SHAP| value (0.57) with a negative direction, corresponding to a lower predicted probability of psychiatric AE reporting in older adults. Semaglutide use ranked second (0.35) and showed a positive direction. Absence of concomitant medications (0.20) and diabetes indication (0.10) showed negative directions, while younger age (19–44 years, 0.08) showed a positive direction. These patterns remained consistent in sensitivity analysis excluding concomitant psychotropic medication users (AUROC 0.797). Conclusions: Older age status was associated with decreased predicted probability of psychiatric AE reporting, while semaglutide use and younger age showed positive contributions. These case-level patterns, identified through pharmacovigilance analysis using a comparator-based approach and explainable machine learning, suggest that reporting patterns may differ across subgroups and that the observed reporting heterogeneity among younger adults and semaglutide users merits further investigation in prospective studies. Full article

A therapeutic antibody, CSL305, has been developed, which combines inhibition of the complement classical and lectin pathways via complement C2 binding with an ability to act as an antagonist of the neonatal Fc receptor (FcRn). CSL305 binds to human C2 (huC2) but shows no binding or activity against mouse C2 precluding its use in mouse models of disease to fully assess in vivo efficacy. To circumvent this, a mouse strain was developed that replaced the expression of mouse C2 with huC2 by homologous recombination. These mice (huC2 “knock-in”; KI) were shown to express huC2 protein and to have complement activity. Interestingly, male huC2-KI mice showed much stronger complement activity compared to female mice and were also sensitive to inhibition by CSL305. Two models of disease using male huC2-KI mice were then used to assess the in vivo efficacy of CSL305. The first was an attenuated passive anti-glomerular basement membrane (GBM) glomerulonephritis model involving complement activation as its primary mechanism of action. CSL305 showed dose-dependent inhibition of disease as measured by urine albumin, with reductions in kidney cellular infiltration and plasma C3 cleaved fragments C3b/C3c/iC3b also observed. The second model was a collagen autoantibody-induced arthritis (CAIA) mouse model. Here, CSL305 showed a significant and dose-dependent inhibition of clinical score in both prophylactic and therapeutic settings, mediated exclusively via its FcRn mechanism of action. Although the animal models used in this study were found to preclude the demonstration of a synergistic effect on both mechanisms, CSL305 does act in vivo as both a complement inhibitor and as a FcRn antagonist to ameliorate disease. Full article

The Robot Operating System (ROS) is widely used in modern robotics, but its open architecture makes it vulnerable to numerous cyber threats. Although machine learning (ML)-based intrusion detection systems (IDSs) demonstrate strong classification performance on ROS-specific datasets, reliance on topology-dependent identifiers such as source and destination IP addresses, port numbers, and Flow IDs remains a critical limitation in current research. This reliance may encourage algorithms to exploit scenario-specific endpoint signatures instead of relying primarily on transferable behavioral patterns. Consequently, classification scores may be artificially inflated due to data leakage. This study addresses this issue by quantitatively measuring the impact of data leakage and introducing a topology-independent, explainable ROS framework that provides a more realistic, leakage-aware, and topology-independent evaluation framework. The evaluation involved testing the LightGBM, XGBoost, and CatBoost algorithms on ROSIDS23. Additionally, Random Forest and Gradient Boosting were included to verify the presence of data leakage. In our ablation study, models that included topology features achieved near-perfect Macro-F1 values of 0.999 to 1.000. In contrast, removing topology-dependent features reduced the Macro-F1 score to about 0.66. This finding shows that topology descriptors, rather than just transferable attack behaviors, can significantly influence the near-perfect scores seen with topology-preserving protocols. Even without topology data, ML models effectively captured temporal behavioral patterns and detected DoS attacks with nearly perfect performance, reaching F1 scores of 0.99 or higher. However, semantic attacks like Unauthorized Subscribe remained tough to classify, with F1 scores of 0.43 or lower. Additionally, SHapley Additive exPlanations (SHAP) analysis improves the interpretability of IDSs by identifying the main behavioral features that drive model decisions and suggesting feature-level directions for rule-based defense configurations in ROS environments. Full article

Background: Acquired tolerance to temozolomide (TMZ) remains one of the main obstacles to enduring therapeutic success in glioblastoma (GBM). While tumor-derived extracellular vesicles are known to orchestrate therapy evasion by horizontally transferring molecules across the tumor microenvironment, the precise regulatory roles of specific exosomal circular RNAs (circRNAs) in establishing this refractory state require further elucidation. Methods: The expression of circ_0050688 in TMZ-resistant GBM clinical tissues and cell lines was evaluated. Exosomes derived from resistant cells were isolated and confirmed via transmission electron microscopy (TEM) and marker analysis. PKH67 fluorescent tracking was utilized to visually demonstrate exosome internalization by sensitive recipient cells. Biological functions, including the expression of the multidrug resistance protein P-glycoprotein (P-gp) and the proliferation marker Ki-67, were evaluated. The competing endogenous RNA mechanism was validated using RNA FISH, dual-luciferase reporters, and functional rescue experiments. In vivo efficacy was determined using subcutaneous xenograft mouse models. Results: Clinical and in vitro analyses revealed that circ_0050688 is upregulated in TMZ-refractory GBM, predicting adverse patient survival. Through PKH67-based tracing, we confirmed that resistant cells actively secrete circ_0050688-enriched exosomes, which are subsequently engulfed by drug-sensitive bystander cells. This vesicular transfer directly instigates a chemoresistant and highly proliferative phenotype, marked by elevated P-gp and Ki-67 levels. At the molecular level, circ_0050688 operates as a molecular decoy for miR-508-5p, thereby preventing the suppression of its downstream target, MDM2. Functionally, circ_0050688 depletion eradicated these aggressive traits and restored TMZ vulnerability across both cellular and murine xenograft models. Furthermore, rescue assays confirmed that this circ_0050688-driven chemoresistance is fundamentally dependent on the miR-508-5p/MDM2 signaling axis. Conclusions: Current data uncover an intercellular signaling network driven by vesicular circ_0050688, which functions as a mobile oncogene to reshape the TMZ-refractory microenvironment. Targeting this exosomal circ_0050688/miR-508-5p/MDM2 network to suppress P-gp and Ki-67 expression represents a highly promising therapeutic strategy for refractory GBM. Full article

Soil organic carbon (SOC) is closely linked to soil fertility, agricultural carbon cycling, and the functioning of cotton field ecosystems, and it provides essential information for sustainable soil management. Rapid and accurate SOC estimation is therefore important for assessing carbon sequestration potential and supporting low-carbon agricultural management. This study focused on cotton fields in northern Shawan City and used optical imagery, Synthetic Aperture Radar (SAR) imagery, and 140 ground-collected SOC samples to estimate SOC content with three machine learning models: Random Forest (RF), Light Gradient Boosting Machine (LightGBM), and Extreme Gradient Boosting (XGBoost). The Kennard–Stone algorithm was applied to partition the 140 SOC samples into training and validation subsets at a 7:3 ratio, ensuring a more representative distribution of samples. Model performance was evaluated using the coefficient of determination (R²) and root mean square error (RMSE), and SHapley Additive exPlanations (SHAP) was used to interpret feature contributions and SOC spatial variability. The results showed that: (1) optical features performed better than SAR features, while fused optical-SAR features achieved the highest accuracy; (2) XGBoost consistently outperformed RF and LightGBM, with the optimal model achieving R² = 0.726 and RMSE = 1.252% on the validation set; (3) SHAP analysis confirmed the dominant contribution of optical features to SOC estimation; and (4) the predicted SOC distribution showed higher values in the central study area, lower values in the northern and southern parts, and high-value zones mainly along both sides of the Manas River. By comparing optical, SAR, and fused features for SOC estimation in arid-zone cotton fields, this study provides methodological support for rapid SOC monitoring and sustainable soil management, and offers practical guidance for variable-rate fertilization and soil carbon sequestration planning along the Manas River corridor. Full article

Source apportionment and the elucidation of driving mechanisms are essential for targeted soil pollution management. This study investigated surface soils across six towns in southern Shimen County, northwestern Hunan Province, where 662 samples were collected to determine the concentrations of As, Cd, Cr, Cu, Ni, Pb, and Zn. Multivariate statistics and the APCS-MLR receptor model were integrated to quantify pollution sources, while three machine learning models (RF, XGBoost, and LightGBM) were applied to identify key drivers of the spatial enrichment of Cd. Results showed that Cd was significantly enriched, with a mean concentration of 0.43 mg/kg (3.41 times the provincial background value). The mean concentrations of As, Cr, Cu, Ni, Pb and Zn were 11.97 mg/kg, 81.01 mg/kg, 24.15 mg/kg, 49.25 mg/kg, 29.56 mg/kg and 76.77 mg/kg, respectively, and these PTEs remained at normal background levels. Significant inter-element correlations indicated common sources. Three primary sources were quantified—natural parent material (43.83%), mining activities (30.99%), and mixed sources of coal mining and agricultural inputs (7.84%), with 17.34% attributed to unidentified mixed sources. Natural sources dominated the geogenic enrichment of Cd, Cu, Ni, Pb, and Zn; mining activities governed the accumulation of As, Cr, Cu, and Pb; a mixed source of coal mining and agricultural practices contributed substantially to Cd enrichment. Machine learning identified PM10, topography, strata, and soil type as dominant drivers, with their total feature importance reaching 70.05%. Among these factors, natural factors and anthropogenic factors accounted for 44.23% and 55.77% of the total feature importance, in turn revealing coupled natural–anthropogenic controls. This study establishes an integrated framework linking source apportionment and driver identification, providing scientific insights for potentially toxic elements (PTEs) control in analogous mining–agricultural regions. Full article