Search Results (511)

Early and non-destructive identification of fungal contamination in cereals is essential to support post-harvest management, reduce economic losses, and mitigate food safety risks along the wheat supply chain. Among filamentous fungi, Alternaria spp. are widespread contaminants of durum wheat and producers of toxic secondary metabolites such as alternariol (AOH), whose early detection remains analytically challenging. The aim of this study was to evaluate the potential of near-infrared transmittance (NIT) spectroscopy as a rapid, non-destructive pre-screening tool for the early identification of Alternaria-contaminated durum wheat lots and associated AOH risk. Samples from three durum wheat cultivars were artificially inoculated with Alternaria spp. and monitored over time. NIT spectra (570–1100 nm) were acquired in transmittance mode and analyzed using partial least squares (PLS) regression, focusing on the 870–1100 nm spectral region. Clear and time-dependent spectral differences were observed between inoculated and control samples, with the strongest discriminative features at 834 and 966 nm. Classification performance was high, with area under the curve (AUC) values between 0.96 and 0.97. ELISA analysis confirmed progressive AOH accumulation in inoculated kernels, consistent with the observed spectral changes, while control experiments excluded autoclaving and visual grain damage as confounding factors. From an applied perspective, the results indicate that NIT spectroscopy can support post-harvest decision-making as a rapid pre-screening approach, enabling the prioritization of suspect wheat lots for confirmatory analytical testing. Multivariate analysis further confirmed the consistency of spectral differences across datasets. Full article

Background/Objectives: Early progression (EP) occurs in a subset of patients with locally advanced pancreatic cancer (LAPC), limiting the clinical benefit of treatment, and it remains difficult to predict. Methods: We developed a multiparametric predictive model integrating baseline ¹⁸F-FDG PET/CT radiomic features with clinical and biological data. A total of 242 radiomic features were extracted from each imaging modality (CT and PET), including first-order, gray-level co-occurrence matrix (GLCM), and local binary pattern (LBP-TOP) features, and combined with PET-derived metrics and clinical variables. Model development included cross-validation procedures and rigorous feature selection, followed by the training of a two-level decision tree classifier. Results: The model achieved an accuracy of 80.7% and an area under the curve (AUC) of 0.83. Integrated analysis of CT and PET texture enabled the identification of patients at high risk of EP prior to treatment initiation. Conclusions: PET/CT-based radiomic biomarkers, combined with clinical data, can non-invasively capture tumor heterogeneity and improve risk stratification in LAPC, supporting more personalized therapeutic decision-making. Full article

Background: Lung adenocarcinoma (LUAD) is the most common type of lung cancer and a major cause of cancer death. Zinc finger proteins (ZNFs) have been implicated in LUAD progression, functioning either as oncogenes or tumor suppressors. Therefore, an in-depth investigation of ZNFs may contribute to the development of novel diagnostic and therapeutic strategies for LUAD. Methods: Transcriptomic and clinical data were obtained from the TCGA and GEO databases. Prognosis-related ZNF genes were identified using univariate Cox, LASSO, and multivariate Cox regression analyses. An eight-gene ZNF-based prognostic signature was constructed and validated in two independent external cohorts (GSE50081 and GSE26939). A nomogram integrating independent prognostic factors was developed. Immune infiltration, somatic mutation profiles, and drug sensitivity were systematically analyzed. We further focused on FGD3, a key gene from the signature, examining its expression in LUAD cells and tissues, including lorlatinib-resistant models. Results: The prognostic signature comprising TRIM6, TRIM29, CTCFL, FGD3, GATA4, CASZ1, TRAF2, and ZNF322 effectively stratified patients into distinct risk groups with significantly different overall survival (p < 0.05). The risk score, together with T and N stage, served as independent prognostic predictors (n = 500, p < 0.05). High-risk patients exhibited an immune-desert phenotype, increased tumor mutational burden, and distinct drug sensitivity patterns. Notably, FGD3 expression was downregulated in LUAD tissues (n = 14, p < 0.0001) and lorlatinib-resistant cells, and its restoration suppressed resistant cell proliferation and partially reversed drug resistance. Conclusions: This study establishes a promising ZNF-based prognostic model for LUAD, providing a potential tool for risk stratification and individualized therapeutic decision-making. The identification of FGD3 as a potential mediator of drug resistance highlights its promise as a candidate biomarker and therapeutic target in LUAD. Full article

Background/Objectives: Vaccine-preventable diseases remain a persistent public health challenge in regions characterized by structural vulnerabilities, including suboptimal vaccination coverage, socioeconomic deprivation, and limited access to healthcare. In structurally vulnerable regions, such as the South-West Romanian region, characterized by persistent vaccination gaps and recurrent outbreaks, these conditions generate a sustained public health burden that requires ongoing preventive risk management strategies. In such contexts, digital risk stratification tools may support preventive decision-making by enabling early identification of patients at increased risk of severe outcomes. This study applied machine learning techniques to routinely collected measles surveillance data from South-West Romania to identify severe disease cases and determine key predictors of severity, offering a pragmatic alternative to outbreak forecasting in a resource-constrained setting. Methods: An open epidemiological dataset of laboratory-confirmed measles cases reported by the Regional Center for Public Health Surveillance Craiova was analyzed. The dataset defined severe cases as those with pneumonia, thrombocytopenia, a hospital stay exceeding three days, or other documented complications requiring medical intervention. Random Forest (RF) and Logistic Regression (LR) classifiers were trained and compared using a 10-fold cross-validation framework across 200 resampling iterations. Model performance was assessed using accuracy, AUC-ROC, sensitivity, specificity, positive predictive value, and F1-score. Feature importance was quantified using permutation-based measures, and the highest-ranked predictors were further evaluated through chi-square tests of independence. Results: RF significantly outperformed LR in accuracy (0.84 vs. 0.82), AUC (0.87 vs. 0.80), specificity (0.87 vs. 0.84), positive predictive value (0.89 vs. 0.86), and F1-score (0.84 vs. 0.83), with p ≤ 0.001 for most metrics. Sensitivity was equivalent between models (approximately 0.81; p = 0.328). Feature importance analysis identified seven key predictors: county of residence, vaccination status, outbreak status, presence of other symptoms, occupation, cough, and conjunctivitis. All seven were significantly associated with disease severity, and six showed significant geographic variation across counties. Vâlcea County had the highest concentration of severe cases. The model was trained on a regional surveillance cohort in which symptomatic and hospitalized cases are over-represented and should be interpreted as a triage-support tool within this surveillance context rather than as a population-level severity estimator. Conclusions: Machine learning, particularly RF, can effectively identify severe measles cases using routinely collected surveillance data in settings where robust outbreak prediction is not feasible. The county of residence functioned as a composite proxy for structural determinants, including healthcare access, vaccination coverage, and socioeconomic deprivation. These findings support the use of ML-based severity classification as a pragmatic tool for clinical risk stratification and targeted public health intervention in resource-constrained environments. Full article

The transition towards alternative marine fuels introduces new safety challenges related to onboard storage, distribution, and fuel management, due to the markedly different physical and chemical properties of methane, methanol, ammonia, and hydrogen. While numerous studies address the risks of individual fuels, there is a lack of structured and comparable risk-assessment methodologies to support early-stage fuel selection and preliminary system design under a unified framework. This study introduces the Methodology to Alternative-fuels Hazardous Identification, a hybrid framework that integrates HAZOP-based deviation analysis with HAZID-style risk classification to enable a consistent qualitative–quantitative comparison of alternative marine fuel systems. The methodology is applied to representative storage and distribution architectures for methane, methanol, ammonia, compressed hydrogen, and liquefied hydrogen, allowing the identification of dominant risk drivers and system-level vulnerabilities across fuel options. The results reveal distinct fuel-specific risk profiles. Methane and methanol are mainly associated with moderate risks linked to operational temperature deviations and system controllability. Ammonia exhibits the most severe risk profile due to the high consequences of toxic releases, particularly under pressure-related failures. Compressed hydrogen is dominated by high-risk scenarios driven by extreme storage pressures, while liquefied hydrogen presents a mixed profile governed by the interaction between cryogenic temperature control and pressure regulation. By providing a comparative and scalable risk-assessment framework, the Methodology to Alternative-fuels Hazardous Identification (MAHI) supports informed decision-making in early design phases and complements existing regulatory safety analyses, contributing to a safer energy transition in maritime transport. Full article

Background/Objectives: Sepsis is a dynamic clinical syndrome characterized by a rapidly evolving inflammatory response, where early identification of patients at risk for adverse outcomes remains a major challenge. While inflammatory biomarkers are widely used, their prognostic value at baseline is limited. This study aimed to evaluate whether early changes in inflammatory biomarkers, particularly the neutrophil-to-lymphocyte ratio (ΔNLR), provide additional prognostic value in predicting short-term outcomes in patients with sepsis. Methods: A retrospective longitudinal observational study was conducted, including 168 adult patients admitted with sepsis at a tertiary infectious diseases hospital. Inflammatory biomarkers (CRP, procalcitonin, leukocyte subpopulations, and NLR) were assessed at admission and at 48–72 h. Early changes (Δ values) were calculated and analyzed in relation to a composite adverse outcome, including ICU admission, vasopressor requirement, mechanical ventilation, or in-hospital mortality. Logistic regression and ROC curve analyses were used to evaluate predictive performance. Results: Patients with adverse outcomes had significantly higher baseline inflammatory markers and severity scores. Early reductions in CRP and NLR were more pronounced in survivors, whereas non-survivors showed persistently elevated or minimally decreasing values. In multivariate analysis, ΔNLR remained independently associated with in-hospital mortality (OR 0.91, 95% CI 0.84–0.98, p = 0.015), alongside Sequential Organ Failure Assessment (SOFA) score and septic shock. ΔNLR demonstrated better discriminative performance (AUC 0.74) compared to baseline markers and improved predictive accuracy when combined with SOFA score (AUC 0.81). Higher baseline NLR quartiles were associated with a stepwise increase in adverse outcomes. Conclusions: Early changes in inflammatory biomarkers, particularly ΔNLR, provide clinically relevant prognostic information beyond baseline measurements and severity scores in sepsis. Dynamic assessment of immune response may improve early risk stratification and support more individualized clinical decision-making. Full article

Low-visibility environments induced by sea fog severely constrain the navigational efficiency and safety in narrow waterways, where traditional radar and Automatic Identification Systems (AIS) frequently encounter challenges such as perception blind spots and information lag. To address this critical issue, this study proposes a UAV-based perception and decision-making methodology for main navigational routes in fog, integrating physical priors with unmanned aerial vehicle (UAV) vision. Firstly, a joint physical dehazing and fog-domain adaptive detection network is constructed. This network addresses the overcomes the interference of non-uniform fog through feature-level enhancement, generating a spatio-temporally continuous visibility field and ship probability grids under a bird’s-eye view (BEV). Subsequently, a quantified “Sailability Score” model is established, providing a scientific basis for the dynamic diversion, speed limitation, and safe distance maintenance of main navigational routes. Simulation-based verifications using real-world fog navigation scenarios in the Qiongzhou Strait, coupled with a joint analysis of Vessel Traffic Service (VTS) and AIS data, suggest that at the critical visibility threshold (≤500 m), the proposed method improves the recall rate of long-distance small target detection by approximately 16.2% and reduces the visibility estimation error by 19.3%. Furthermore, the consistency between the proposed Sailability Score and the actual VTS navigation restriction windows reaches 82.1%, exhibiting a conservative preference for safety (i.e., risk preference ratio $\gamma >1$). Additionally, by introducing a temporal anti-jitter mechanism (parameterized by a smoothing window $\Delta t$), the proposed method extends the navigable time window of the main routes by approximately 12.4% while ensuring navigational safety. The simulation results indicate the framework’s potential perception capabilities and engineering applicability, providing reliable technical support for smart shipping and intelligent VTS systems. Full article

Modern machine learning approaches for vulnerability detection achieve strong performance within specific datasets, yet their reliability degrades under domain shift, limiting their effectiveness for real-world secure software development lifecycle (SDLC) decision-making. In particular, probabilistic vulnerability predictions, while well-calibrated, exhibit instability across heterogeneous codebases, reducing their suitability as standalone risk indicators. This paper introduces Intelligent Risk-Adaptive Secure SDLC (IRAS-SDLC), a lifecycle risk aggregation framework for Secure AI-Augmented Software Assurance under the Risk Management Framework (RMF) and Zero Trust. The proposed framework integrates model-derived vulnerability likelihood with structured security metrics, specifically exploitability and impact derived from standardized Common Vulnerability Scoring System (CVSS) data, to construct a unified and interpretable risk representation. This formulation enables consistent prioritization across SDLC phases while aligning with RMF control families and Zero Trust continuous verification principles. By combining learned semantic signals with domain-independent security factors, IRAS mitigates the instability of vulnerability likelihood under distributional shifts and provides a more robust basis for cross-domain risk assessment. The framework embeds risk evaluation early in the SDLC, enabling proactive identification of vulnerabilities during the requirements and design phases rather than post-implementation detection. Empirical evaluation demonstrates that IRAS-SDLC maintains meaningful risk estimation under domain shift and significantly improves lifecycle outcomes. In particular, early risk identification yields negative detection latency relative to conventional methods and reduces simulated remediation costs by up to an order of magnitude. IRAS-SDLC bridges the gap between machine learning-based vulnerability prediction and governance-aligned security assurance by providing a stable, interpretable, and lifecycle-aware risk assessment mechanism that is directly compatible with RMF-based compliance workflows and Zero Trust architectures. Full article

Early identification of behavioral risk patterns in large student populations remains a challenge in higher education, particularly when support systems depend on voluntary help-seeking. This study presents YuumCare, a structured and scalable framework that operationalizes population-level digital screening through a reproducible data pipeline for early detection and decision support. The framework was implemented during the first weeks of the academic term in a public higher education institution in Latin America, where 466 first-year students (38.9% coverage) completed a structured questionnaire capturing indicators of emotional well-being, academic pressure, and help-seeking attitudes. Responses were processed through a structured data pipeline comprising data ingestion, preparation, feature construction, and rule-based classification, transforming distributed self-reported data into standardized features and interpretable institutional signals for consistent analysis at scale. Results show that emotional strain, evaluation-related anxiety, and adaptation difficulties emerge early and frequently co-occur, while most students report low willingness to seek professional support. The classification process indicates that approximately one third of the cohort presents moderate to critical levels of need, providing a structured representation of vulnerability. The proposed approach connects digital screening with institutional decision-making through an interpretable and operational workflow that does not rely on complex infrastructure. Beyond descriptive findings, the study contributes a lightweight and reproducible data framework that supports scalable monitoring and coordinated response under real-world constraints, demonstrating the feasibility of transforming self-reported behavioral data into actionable decision-support signals for population-level monitoring in higher education. Full article

Traditional irrigation management for tomatoes in solar greenhouses relies heavily on empirical manual experience and single soil moisture indicators, often leading to irrigation scheduling that lacks crop-specific physiological evidence and results in suboptimal water-use efficiency. To address these challenges, this study developed an intelligent, plant-centric irrigation decision-making framework for greenhouse tomatoes in the arid region of Xinjiang. Central to this framework is the precise identification of irrigation timing—the most critical first step and a fundamental prerequisite for achieving true on-demand irrigation. By monitoring the high-frequency dynamics of stem diameter (SD) and integrating soil moisture data, the physiological responsiveness of tomatoes to water stress was systematically analyzed. A hybrid predictive model, STL-LSTM, was constructed by coupling Seasonal-Trend decomposition using Loess (STL) with Long Short-Term Memory (LSTM) networks to forecast 24-h SD trends. Furthermore, an innovative dual-threshold irrigation mechanism was established, utilizing a physiological trigger (Maximum Daily Shrinkage, MDS > 70 μm) and a soil moisture constraint (Volumetric Water Content, VWC ≤ 17%). Results demonstrated that tomato SD exhibited distinct diurnal rhythms, with MDS and Daily Increment (DI) identified as highly sensitive indicators of plant water status. The proposed STL-LSTM model achieved superior predictive performance during the peak fruiting stage, with a coefficient of determination (R²) of 0.9184, representing an improvement of 14.8% and 27.56% over standalone LSTM and ARIMA models, respectively. The validation of the dual-threshold mechanism confirms its ability to balance real-time crop water demand with conservation requirements, effectively mitigating the risks of premature or delayed irrigation inherent in traditional methods. This research provides scientific rationale and technical support for the transition of greenhouse agriculture in arid regions towards precision irrigation and optimised water resource management. Full article

Urban water-ecological governance in the Yellow River Basin is shifting from a single administratively dominated model toward a polycentric collaborative system. However, ambiguous responsibilities and overlapping tasks among governments, enterprises, and society often lead to governance conflicts, reduced coordination efficiency, and growing risks to regional ecological security. To address this challenge, this study develops a multi-actor governance analysis framework integrating deep learning, knowledge graphs, and complex network optimization. Stakeholder demands are extracted from multi-source data using a BERT-BiLSTM-CRF model, including policy documents, enterprise reports, and public discourse, and are then organized into a knowledge graph for water-ecological governance. A Relational Graph Attention Network (R-GAT) is subsequently used to transform the knowledge graph into a signed weighted network, enabling the measurement of conflict intensity and the identification of key conflict nodes across governance scenarios. Based on multi-objective optimization, a Pareto frontier is constructed to balance conflict tension, fairness, and governance efficiency, from which a compromise solution for responsibility weighting is identified. An empirical case study of a typical city in the Yellow River Basin shows that the proposed framework can identify core conflict nodes and provide quantitative support for conflict mitigation and coordination adjustment. The findings offer a quantitative reference for institutional innovation and evidence-based decision-making in urban water-ecological governance. Full article

Fetal health assessment is a crucial aspect of prenatal care, aimed at the early detection of potential complications to ensure optimal outcomes for both mother and child. Traditional methods, such as the visual analysis of cardiotocography (CTG) data by healthcare professionals, are valuable but often subjective and time-consuming. This work investigates the application of machine learning techniques, with a focus on ensemble learning, to enhance the accuracy and efficiency of fetal health classification based on CTG data. Genetic Algorithm (GA) is employed for optimal feature selection, identifying the most discriminative subset of CTG attributes to improve model performance and reduce computational complexity. We employ a combination of advanced machine learning models, including AdaBoost, Gaussian Naïve Bayes, Decision Tree, k-nearest neighbors (KNN), and Logistic Regression. The top two models were selected based on comprehensive performance metrics using the TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method. These models were then integrated through ensemble learning approaches, such as stacking, Particle Swarm Optimization (PSO) weighted averaging, and soft voting, to improve prediction reliability. Our proposed stacking ensemble model achieves a remarkable accuracy of 97.9%, demonstrating its potential as a robust, data-driven tool for fetal health monitoring and the early identification of at-risk pregnancies. The results indicate that machine learning can effectively complement traditional fetal health assessment methods by providing an objective framework to support clinical decision-making. Full article

The indoor microclimate of museums plays an essential role in preserving priceless cultural heritage for future generations and in ensuring visitors’ comfort and health. In this context, the present study aimed to evaluate indoor air quality, the degree of fungal contamination, and visitors’ perceptions in a museum environment through an integrated, interdependent approach. Measurements of the physicochemical parameters of air quality (temperature, relative humidity, CO₂, TVOC, HCHO, PM_2.5 and PM₁₀, negative and positive ions and brightness) were carried out in three exhibition halls within a museum in Oradea, Romania, during the period January–August 2024. Fungal contamination was assessed using surface and air samples, with classical isolation and microscopic identification methods. Visitors’ perceptions were analysed using a standardised questionnaire that focused on perceived comfort and visit duration. The results showed that the parameters defining indoor air quality generally fell within the limits set by the international standards in force, with occasional exceedances. These conditions are associated with the presence of fungi of the genera Cladosporium, Penicillium, and Aspergillus in the air and on museum exhibits, which pose risks to human health and the deterioration of the exhibited materials. The statistical decision-making model determined the critical thresholds above which visitor behaviour changed visibly. The results highlighted the importance of maintaining a stable microclimate in museum spaces, not only for the protection of exhibits, but also for optimising the cultural experience. Indoor air quality indicators and fungal microflora can only affect vulnerable people or those with pre-existing conditions. Occasional visitors do not present a significant risk of developing new conditions, considering the limited duration of exposure. Full article

The rising prevalence of chronic diseases, driven by population ageing, emerging pathogens, and evolving lifestyles, necessitates stronger healthcare systems that integrate effective prevention with timely intervention. Sepsis remains one of the most critical and life-threatening conditions, associated with high incidence, mortality, and morbidity, and frequently progressing to multiple organ dysfunction and septic shock. Early identification is therefore essential to improve patient outcomes. In this work, we propose a rapid and accurate data-driven framework for early sepsis prediction. The framework comprises four stages: data collection, preprocessing, preparation, and classification. Real-world clinical data from 1000 patients are utilized for early risk assessment. Data preprocessing focuses on cleaning and extracting clinically relevant features, followed by data preparation steps including labeling, dataset splitting, class balancing, and feature scaling. Multiple machine learning and neural network models are then implemented, with optimized parameter selection to enhance predictive performance. Finally, a deployment module enables healthcare professionals to leverage the trained models for real-time patient status assessment, supporting timely clinical decision-making. Extensive experimental results demonstrate that the proposed framework achieves fast and accurate discrimination between septic and non-septic patients, outperforming existing state-of-the-art approaches. Full article

Acute pancreatitis is a common inflammatory digestive disease with an unpredictable clinical course, ranging from self-limited forms to severe forms, associated with complications and increased mortality. Early identification of patients at risk of severe disease is particularly important from a surgical perspective, as it has a significant impact on subsequent management. Traditional severity scores, such as APACHE (Acute Physiology And Chronic Health Evaluation) II and BISAP (Bedside Index for Severity in Acute Pancreatitis), remain widely used, but their rigid structure and delayed applicability may limit initial risk assessment. In this review we highlight the evolving role of artificial intelligence in predicting the severity of acute pancreatitis and supporting clinical decision-making, with a focus on surgical management. Recent advances show that data-driven models could improve early risk assessment compared to traditional methods. Although their potential clinical benefits are becoming increasingly clear, real-world implementation remains limited. Initial results are encouraging, but important questions regarding reliability, safety, and integration into clinical practice still need to be addressed. Full article