Search Results (505)

This study investigates the impact of data distribution and bootstrap resampling on the anomaly detection performance of the Isolation Forest (iForest) algorithm in statistical process control. Although iForest has received attention for its multivariate and ensemble-based nature, its performance under non-normal data distributions and varying bootstrap settings remains underexplored. To address this gap, a comprehensive simulation was performed across 18 scenarios involving log-normal, gamma, and t-distributions with different mean shift levels and bootstrap configurations. The results show that iForest substantially outperforms the conventional Hotelling’s T² control chart, especially in non-Gaussian settings and under small-to-medium process shifts. Enabling bootstrap resampling led to marginal improvements across classification metrics, including accuracy, precision, recall, F1-score, and average run length (ARL)₁. However, a key limitation of iForest was its reduced sensitivity to subtle process changes, such as a 1σ mean shift, highlighting an area for future enhancement. Full article

Lymphoma continues to pose a significant global health burden, highlighting the urgent need for novel therapeutic strategies. Recent advances in microbiome research have identified gut-microbiota-derived metabolites, or postbiotics, as promising candidates in cancer therapy. This study investigates the antiproliferative and mechanistic effects of two postbiotics, Nisin (N) and Urolithin B (UB), individually and in combination, against the human lymphoma cell line HKB-11. Moreover, this study evaluated cytotoxic efficacy and underlying molecular pathways using a comprehensive experimental approach, including the Alamar Blue assay, combination index (CI) analysis, flow cytometry, reactive oxygen species (ROS) quantification, and bottom-up proteomics. N and UB displayed notable antiproliferative effects, with IC₅₀ values of 1467 µM and 87.56 µM, respectively. Importantly, their combination at a 4:6 ratio demonstrated strong synergy (CI = 0.09 at IC₉₅), significantly enhancing apoptosis (p ≤ 0.0001) and modulating oxidative stress. Proteomic profiling revealed significant regulation of key proteins related to lipid metabolism, mitochondrial function, cell cycle control, and apoptosis, including upregulation of COX6C (Log₂FC = 2.07) and downregulation of CDK4 (Log₂FC = −1.26). These findings provide mechanistic insights and underscore the translational potential of postbiotics in lymphoma treatment. Further preclinical and clinical investigations are warranted to explore their role in therapeutic regimens. Full article

Emerging digital technologies are transforming how consumers participate in financial markets, yet their benefits depend critically on the speed, reliability, and transparency of the underlying platforms. Online stock trading platforms must maintain high efficiency underload to ensure a good user experience. This paper presents performance analysis under various load conditions based on the containerized stock exchange system. A comprehensive data logging pipeline was implemented, capturing metrics such as API response times, database query times, and resource utilization. We analyze the collected data to identify performance patterns, using both statistical analysis and machine learning techniques. Preliminary analysis reveals correlations between application processing time and database load, as well as the impact of user behavior on system performance. Association rule mining is applied to uncover relationships among performance metrics, and multiple classification algorithms are evaluated for their ability to predict user activity class patterns from system metrics. The insights from this work can guide optimizations in similar distributed web applications to improve scalability and reliability under a heavy load. By framing performance not merely as a technical property but as a determinant of financial decision-making and well-being, the study contributes actionable insights for designers of consumer-facing fintech services seeking to meet sustainable development goals through trustworthy, resilient digital infrastructure. Full article

Accurate estimation of long-term average (LTA) discharge is fundamental for water resource assessment, infrastructure planning, and hydrological modeling, yet it remains a significant challenge, particularly in data-scarce or ungauged basins. This study introduces an advanced machine learning framework to estimate long-term average discharge using globally available hydrological station metadata from the Global Runoff Data Centre (GRDC). The methodology involved comprehensive data preprocessing, extensive feature engineering, log-transformation of the target variable, and the development of multiple predictive models, including a custom deep neural network with specialized pathways and gradient boosting machines (XGBoost, LightGBM, CatBoost). Hyperparameters were optimized using Bayesian techniques, and a weighted Meta Ensemble model, which combines predictions from the best individual models, was implemented. Performance was rigorously evaluated using R², RMSE, and MAE on an independent test set. The Meta Ensemble model demonstrated superior performance, achieving a Coefficient of Determination (R²) of 0.954 on the test data, significantly surpassing baseline and individual advanced models. Model interpretability analysis using SHAP (Shapley Additive explanations) confirmed that catchment area and geographical attributes are the most dominant predictors. The resulting model provides a robust, accurate, and scalable data-driven solution for estimating long-term average discharge, enhancing water resource assessment capabilities and offering a powerful tool for large-scale hydrological analysis. Full article

The deep carbonate reservoirs in the Tarim Basin, Xinjiang, China, are widely developed with multi-scale complex reservoir spaces such as fractures, pores, and karst caves under the coupling of abnormal high pressure, diagenesis, karst, and tectonics and have strong heterogeneity. Among them, fracture–cavity carbonate reservoirs are one of the main reservoir types. Revealing the petrophysical characteristics of fracture–cavity carbonate reservoirs can provide a theoretical basis for the log interpretation and geophysical prediction of deep reservoirs, which holds significant implications for deep hydrocarbon exploration and production. In this study, based on the mineral composition and complex pore structure of carbonate rocks in the Tarim Basin, we comprehensively applied classical petrophysical models, including Voigt–Reuss–Hill, DEM (Differential Effective Medium), Hudson, Wood, and Gassmann, to establish a fracture–cavity petrophysical model tailored to the target block. This model effectively characterizes the complex pore structure of deep carbonate rocks and addresses the applicability limitations of conventional models in heterogeneous reservoirs. The discrepancies between the model-predicted elastic moduli, longitudinal and shear wave velocities (Vp and Vs), and laboratory measurements are within 4%, validating the model’s reliability. Petrophysical template analysis demonstrates that P-wave impedance (Ip) and the Vp/Vs ratio increase with water saturation but decrease with fracture density. A higher fracture density amplifies the fluid effect on the elastic properties of reservoir samples. The Vp/Vs ratio is more sensitive to pore fluids than to fractures, whereas Ip is more sensitive to fracture density. Regions with higher fracture and pore development exhibit greater hydrocarbon storage potential. Therefore, this petrophysical model and its quantitative templates can provide theoretical and technical support for predicting geological sweet spots in deep carbonate reservoirs. Full article

Wastewater-based epidemiology (WBE) has historically proven to be a powerful surveillance tool, particularly during the SARS-CoV-2 pandemic. Effective WBE depends on the sensitive detection of pathogens in wastewater. However, determining the process limit of detection (PLOD) of WBE through a comprehensive evaluation that accounts for pathogen concentration, nucleic acid extraction, and molecular analysis has rarely been documented. We prepared dilution series with known concentrations of S. Typhi, V. cholerae, rotavirus, and SARS-CoV-2 in surface water and wastewater. Pathogen concentration was performed using Nanotrap particles with the KingFisher™ Apex robotic platform, followed by nucleic acid extraction. Quantitative real-time PCR (qPCR) and digital PCR (dPCR) were used to detect the extracted nucleic acids of the pathogens. The PLODs and recovery efficiencies for each of the four pathogens in surface water and wastewater were determined. Overall, the observed PLODs for S. Typhi, V. cholerae, and rotavirus in surface water and wastewater were approximately 3 log₁₀ loads (2.1–2.8 × 10³/10 mL) using either qPCR or dPCR as the detection method. For SARS-CoV-2, the PLOD in surface water was 2.9 × 10⁴/10 mL with both RT-qPCR and dPCR, one log₁₀ higher than the PLODs of the other three pathogens. In wastewater, the PLOD for SARS-CoV-2 was 2.9 × 10⁴/10 mL using RT-qPCR and 2.9 × 10³/10 mL using dPCR. The mean recovery rates of S. Typhi, V. cholerae, rotavirus, and SARS-CoV-2 for dPCR in both surface water and wastewater were below 10.4%, except for S. Typhi and V. cholerae in wastewater, which showed significantly higher recoveries, from 26.5% at 4.6 × 10⁵/10 mL for S. Typhi to 58.8% at 4.8 × 10⁵/10 mL for V. cholerae. Our study demonstrated that combining qPCR or dPCR analysis with automated Nanotrap particle concentration and nucleic acid extraction using the KingFisher™ platform enables the sensitive detection of S. Typhi, V. cholerae, rotavirus, and SARS-CoV-2 in surface water and wastewater. Full article

The Lower–Middle Jurassic of the Kuqa Depression consists of terrestrial clastic deposits containing coal seams and thick lacustrine mudstones, and is of great significance for oil and gas exploration. Based on the comprehensive analysis of core, well-logging, outcrop, and seismic data, the sequence stratigraphy, depositional systems, and the controlling factors of the basin filling in the depression are systematically documented. Four primary depositional systems, including braided river delta, meandering river delta, lacustrine, and swamp deposits, are identified within the Ahe, Yangxia, and Kezilenuer Formations of the Lower–Middle Jurassic. The basin fills can be classified into two second-order and nine third-order sequences (SQ1–SQ9) confined by regional or local unconformities and their correlative conformities. This study shows that the sedimentary evolution has undergone the following three stages: Stage I (SQ1–SQ2) primarily developed braided river, braided river delta, and shallow lacustrine deposits; Stage II (SQ3–SQ5) primarily developed meandering river, meandering river delta, and extensive deep and semi-deep lacustrine deposits; Stage III (SQ6–SQ9) primarily developed swamp (SQ6–SQ7), meandering river delta, and shore–shallow lacustrine deposits (SQ8–SQ9). The uplift of the Tianshan Orogenic Belt in the Early Jurassic (Stage I) may have facilitated the development of braided fluvial–deltaic deposits. The subsequential expansion of the sedimentary area and the weakened sediment supply can be attributed to the planation of the source area and widespread basin subsidence, with the transition of the depositional environments from braided river delta deposits to meandering river delta and swamp deposits. The regional expansion or rise of the lake during Stage II was likely triggered by the hot and humid climate conditions, possibly associated with the Early Jurassic Toarcian Oceanic Anoxic Event. The thick swamp deposits formed during Stage III may be controlled by the interplay of rational accommodation, warm and humid climatic conditions, and limited sediment supply. Milankovitch cycles identified in Stage III further reveal that coal accumulation was primarily modulated by long-period eccentricity forcing. Full article

With the expansion of oil and gas exploration and development to complex oil and gas resource areas such as deep and ultra-deep formation onshore and offshore, the kick is one of the high drilling risks, and timely and accurate early kick detection is increasingly important. Based on the kick generation mechanism, kick characterization parameters are preliminarily selected. According to the characteristics of the data and previous research progress, Random Forest (RF), Support Vector Machine (SVM), Feedforward Neural Network (FNN), and Long Short-term Memory Neural Network (LSTM) are established using experimental data from Memorial University of Newfoundland. The test results show that the accuracy of the SVM-linear model was 0.968, and the missing alarm and the false alarm rate only was 0.06 and 0.11. Additionally, through the analysis of the kick response time, the lag time of the SVM-linear model was 1.3 s, and the comprehensive equivalent time was 23.13 s, which showed the best performance. The different effects of the model after data transformation are analyzed, the mechanism of the best effect of the SVM model is analyzed, and the changes in the effect of other models including RF are further revealed. The proposed early-warning model warns in advance in historical well logging data, which is expected to provide a fast, efficient, and accurate gas kick warning model for drilling sites. Full article

Although the Power Internet of Things (PIoT) significantly improves operational efficiency by enabling real-time monitoring, intelligent control, and predictive maintenance across the grid, its inherently open and deeply interconnected cyber-physical architecture concurrently introduces increasingly complex and severe security threats. Existing IoT security solutions are not fully adapted to the specific requirements of power systems, such as safety-critical reliability, protocol heterogeneity, physical/electrical context awareness, and the incorporation of domain-specific operational knowledge unique to the power sector. These limitations often lead to high false positives (flagging normal operations as malicious) and false negatives (failing to detect actual intrusions), ultimately compromising system stability and security response. To address these challenges, we propose a domain knowledge-driven threat source detection and localization method for the PIoT. The proposed method combines multi-source features—including electrical-layer measurements, network-layer metrics, and behavioral-layer logs—into a unified representation through a multi-level PIoT feature engineering framework. Building on advances in multimodal data integration and feature fusion, our framework employs a hybrid neural architecture combining the TabTransformer to model structured physical and network-layer features with BiLSTM to capture temporal dependencies in behavioral log sequences. This design enables comprehensive threat detection while supporting interpretable and fine-grained source localization. Experiments on a real-world Power Internet of Things (PIoT) dataset demonstrate that the proposed method achieves high detection accuracy and enables the actionable attribution of attack stages aligned with the MITRE Adversarial Tactics, Techniques, and Common Knowledge (ATT&CK) framework. The proposed approach offers a scalable and domain-adaptable foundation for security analytics in cyber-physical power systems. Full article

The increasing deployment of wind energy systems, particularly offshore wind farms, necessitates advanced monitoring and maintenance strategies to ensure optimal performance and minimize downtime. Supervisory Control And Data Acquisition (SCADA) systems have become indispensable tools for monitoring the operational health of wind turbines, generating vast quantities of time series data from various sensors. Anomaly detection techniques applied to this data offer the potential to proactively identify deviations from normal behavior, providing early warning signals of potential component failures. Traditional model-based approaches for fault detection often struggle to capture the complexity and non-linear dynamics of wind turbine systems. This has led to a growing interest in data-driven methods, particularly those leveraging machine learning and deep learning, to address anomaly detection in wind energy applications. This study focuses on the development and application of a semi-supervised, multivariate anomaly detection model for horizontal axis wind turbines. The core of this study lies in Bidirectional Long Short-Term Memory (BI-LSTM) networks, specifically a BI-LSTM autoencoder architecture, to analyze time series data from a SCADA system and automatically detect anomalous behavior that could indicate potential component failures. Moreover, the approach is reinforced by the integration of the Isolation Forest algorithm, which operates in an unsupervised manner to further refine normal behavior by identifying and excluding additional anomalous points in the training set, beyond those already labeled by the data provider. The research utilizes a real-world dataset provided by EDP Renewables, encompassing two years of comprehensive SCADA records collected from a single offshore wind turbine operating in the Gulf of Guinea. Furthermore, the dataset contains the logs of failure events and recorded alarms triggered by the SCADA system across a wide range of subsystems. The paper proposes a multi-modal anomaly detection framework orchestrating an unsupervised module (i.e., decision tree method) with a supervised one (i.e., BI-LSTM AE). The results highlight the efficacy of the BI-LSTM autoencoder in accurately identifying anomalies within the SCADA data that exhibit strong temporal correlation with logged warnings and the actual failure events. The model’s performance is rigorously evaluated using standard machine learning metrics, including precision, recall, F1 Score, and accuracy, all of which demonstrate favorable results. Further analysis is conducted using Cumulative Sum (CUSUM) control charts to gain a deeper understanding of the identified anomalies’ behavior, particularly their persistence and timing leading up to the failures. Full article

Traditional systems in real life lack transparency and ease of use due to their reliance on centralization and large infrastructure. Furthermore, many sectors that rely on information technology face major challenges related to data integrity, trust, and counterfeiting, limiting scalability and acceptance in the community. With the decentralization and digitization of energy transactions in smart grids, security, integrity, and fraud prevention concerns have increased. The main problem addressed in this study is the lack of a secure, tamper-resistant, and decentralized mechanism to facilitate direct consumer-to-prosumer energy transactions. Thus, this is a major challenge in the smart grid. In the blockchain, current consensus algorithms may limit the scalability of smart grids, especially when depending on popular algorithms such as Proof of Work, due to their high energy consumption, which is incompatible with the characteristics of the smart grid. Meanwhile, Proof of Stake algorithms rely on energy or cryptocurrency stake ownership, which may make the smart grid environment in blockchain technology vulnerable to control by the many owning nodes, which is incompatible with the purpose and objective of this study. This study addresses these issues by proposing and implementing a hybrid framework that combines the features of private and public blockchains across three integrated layers: user interface, application, and blockchain. A key contribution of the system is the design of a novel consensus algorithm, Proof of Energy, which selects validators based on node roles and randomized assignment, rather than computational power or stake ownership. This makes it more suitable for smart grid environments. The entire framework was developed without relying on existing decentralized platforms such as Ethereum. The system was evaluated through comprehensive experiments on performance and security. Performance results show a throughput of up to 60.86 transactions per second and an average latency of 3.40 s under a load of 10,000 transactions. Security validation confirmed resistance against digital signature forgery, invalid smart contracts, race conditions, and double-spending attacks. Despite the promising performance, several limitations remain. The current system was developed and tested on a single machine as a simulation-based study using transaction logs without integration of real smart meters or actual energy tokenization in real-time scenarios. In future work, we will focus on integrating real-time smart meters and implementing full energy tokenization to achieve a complete and autonomous smart grid platform. Overall, the proposed system significantly enhances data integrity, trust, and resistance to counterfeiting in smart grids. Full article

Background: Cholangiocarcinoma (CCA) of the liver is a highly aggressive cancer that arises from malignant cells in the bile ducts. Radical surgery remains the only curative option, but major liver resection carries high perioperative risks. This study investigates the predictive value of preoperative bone mineral density (BMD), measured via CT, for perioperative complications, mortality, and long-term outcomes. Methods: The analysis included 202 patients who underwent curative-intent surgery for intrahepatic cholangiocarcinoma (iCCA; n = 97) or perihilar cholangiocarcinoma (pCCA; n = 105) between 2010 and 2019. Preoperative bone mineral density (BMD) was assessed using computed tomography segmentation at the level of the 12th thoracic vertebra. Osteopenia was defined according to established cutoffs. Results: Osteopenia was highly prevalent in both iCCA (53/97, 54%) and pCCA (54/105, 51%) subcohorts. Patients suffering from osteopenia were significantly older than those without (71.1 [62–76.6] years vs. 61.3 [52.9–69.2] years; p < 0.001). Alteration in BMD did not demonstrate a significant prognostic effect in terms of perioperative morbidity (Mann–Whitney U; comprehensive complication index—CCI: 34 [9–56] vs. 40 [21–72] p = 0.185; iCCA: p = 0.803; pCCA: p = 0.165). The median overall survival in our cohort was 19 [14–25] months. Patients with osteopenia did not exhibit a significantly different overall survival compared to those with normal bone mineral density (log-rank p = 0.234). Conclusions: In contrast to our previous observations in other oncological patient cohorts, osteopenia alone had no significant negative impact on clinical outcomes in our large European cohort of patients undergoing curative-intent surgery for CCA. To validate these findings, further prospective studies are warranted. Full article

Background: Acute kidney injury (AKI) in critically ill patients is often complicated by arrhythmias, potentially affecting outcomes. This study aimed to develop predictive models for arrhythmias in AKI patients and assess the impact of antiarrhythmic drugs on in-hospital mortality. Methods: We conducted a multi-database retrospective cohort study using MIMIC-IV and eICU databases. XGBoost and Bayesian Information Criterion (BIC) models were employed to identify key predictors of arrhythmias. Weighted log-rank and Cox analysis evaluated the effect of amiodarone and metoprolol on in-hospital mortality. Results: Among 14,035 critically ill AKI patients, 5614 individuals (40%) developed arrhythmias. Both XGBoost and BIC showed predictive power for arrhythmias. The XGBoost model identified HR_max, HR_min, and heart failure as the most important features, while the BIC model highlighted heart failure had the highest odds ratio (OR 1.18, 95% CI 1.16–1.20) as a significant predictor. Patients experiencing arrhythmia is associated with in-hospital mortality (arrhythmia group: 636 (11.3%) vs. non-arrhythmia group: 587 (7.0%), p < 0.01). Antiarrhythmic medications showed a statistically significant effect on in-hospital mortality (amiodarone: HR 0.28, 95% CI 0.19–0.41, p < 0.01). Conclusions: Our predictive models demonstrated a robust discriminatory ability for identifying arrhythmia occurrence in critically ill AKI patients, with identified risk factors showing strong clinical relevance. The significant association between arrhythmia occurrence and increased in-hospital mortality underscores the clinical importance of early identification and management. Furthermore, amiodarone therapy effectively reduced the risk of in-hospital mortality in these patients, even after accounting for time-dependent biases. The findings highlight the necessity of precise arrhythmia definition, careful consideration of time-dependent covariates, and comprehensive model validation for clinically actionable insights. Full article

As urban air mobility moves rapidly toward real-world deployment, accurate vehicle trajectory prediction and early collision risk detection are vital for safe low-altitude operations. This study presents a deep learning framework based on an LSTM–Attention network that captures both short-term flight dynamics and long-range dependencies in trajectory data. The model is trained on fifty-six routes generated from a UAM planned commercialization network, sampled at 0.1 s intervals. To unify spatial dimensions, the model uses Earth-Centered Earth-Fixed (ECEF) coordinates, enabling efficient Euclidean distance calculations. The trajectory prediction component achieves an RMSE of 0.2172, MAE of 0.1668, and MSE of 0.0524. The collision classification module built on the LSTM–Attention prediction backbone delivers an accuracy of 0.9881. Analysis of attention weight distributions reveals which temporal segments most influence model outputs, enhancing interpretability and guiding future refinements. Moreover, this model is embedded within the Short-Term Conflict Alert component of the Safety Nets module in the UAM traffic management system to provide continuous trajectory prediction and collision risk assessment, supporting proactive traffic control. The system exhibits robust generalizability on unseen scenarios and offers a scalable foundation for enhancing operational safety. Validation currently excludes environmental disturbances such as wind, physical obstacles, and real-world flight logs. Future work will incorporate atmospheric variability, sensor and communication uncertainties, and obstacle detection inputs to advance toward a fully integrated traffic management solution with comprehensive situational awareness. Full article

Salt stress is a significant abiotic factor that adversely impacts the yield of rapeseed (Brassica napus L.). Under salt stress conditions, the growth of rapeseed is markedly inhibited. This study integrates transcriptomic and metabolomic analyses to elucidate the molecular and physiological mechanisms underlying the salt stress response during the germination of the rapeseed variety ZS11. Metabolomic analysis revealed 175 differentially expressed metabolites, predominantly comprising amino acids, carbohydrates, and organic acids. Transcriptomic analysis highlighted the crucial roles of plant hormones and phenylpropanoid biosynthesis in enhancing the salt stress resistance of rapeseed. Comprehensive multi-omics analysis identified phenylpropanoid metabolism (p < 0.001), amino acid metabolism (FDR < 0.01), and carbohydrate metabolism (|log₂FC| ≥ 2) as the most significantly affected pathways. Crucially, we demonstrate that early-stage phenylpropanoid activation in hypocotyls dominates salt adaptation during germination. These findings provide actionable targets for molecular breeding and novel insights for optimizing crop establishment in salinized agroecosystems. Full article