Search Results (9,562)

Background: As a widely used chemotherapeutic agent, cisplatin frequently induces acute kidney injury (AKI), which severely compromises patient survival and limits its clinical use. While the natural flavonoid diosmetin (Dio) shows promise in mitigating cisplatin-induced nephrotoxicity, its clinical translation is challenged by poor solubility, low bioavailability, and incompletely elucidated mechanisms. This study aimed to overcome these limitations by developing a novel drug delivery system using the microalgae Dunaliella salina (D. salina, Ds) to load Dio (Ds-Dio), thereby enhancing its efficacy and exploring its therapeutic potential. Methods: We first characterized the physicochemical properties of Ds and Dio, and then Ds-Dio complex was synthesized via co-incubation. Its nephroprotective efficacy and safety were systematically evaluated in a cisplatin-induced mouse AKI model by assessing renal function (serum creatinine, blood urea nitrogen), injury biomarkers, histopathology, body weight, and organ index. The underlying mechanism was predicted by network pharmacology and subsequently validated experimentally. Results: The novel Ds-Dio delivery system has been successfully established. In the AKI model, Ds-Dio significantly improved renal function and exhibited a superior protective effect over Dio alone; this benefit is attributed to the enhanced bioavailability provided by Ds carrier. In addition, Ds-Dio also demonstrated safety performance, with no evidence of toxicity to major organs. Network pharmacology analysis predicted the involvement of PI3K/AKT pathway, which was experimentally verified. Specifically, we confirmed that Ds-Dio alleviates AKI by modulating the PI3K/AKT pathway, resulting in concurrent suppression of NF-κB-mediated inflammation and activation of NRF2-dependent antioxidant responses. Conclusion: This study successfully developed a microalgae-based drug delivery system, Ds-Dio, which significantly enhances the nephroprotective efficacy of Dio against cisplatin-induced AKI. The nephroprotective mechanism is associated with modulation of the PI3K/AKT pathway, resulting in the simultaneous attenuation of oxidative stress and inflammation. Full article

The accurate prediction of human movement direction plays a crucial role in fields such as rehabilitation monitoring, sports science, and intelligent military systems. Based on plantar pressure and inertial sensor data, this study developed a hybrid deep learning model integrating a Convolutional Neural Network (CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) network to enable joint spatiotemporal feature learning. Systematic comparative experiments involving four distinct deep learning models—CNN, BiLSTM, CNN-LSTM, and CNN-BiLSTM—were conducted to evaluate their convergence performance and prediction accuracy comprehensively. Results show that the CNN-BiLSTM model outperforms the other three models, achieving the lowest RMSE (0.26) and MAE (0.14) on the test set, with an R² of 0.86, which indicates superior fitting accuracy and generalization ability. The superior performance of the CNN-BiLSTM model is attributed to its ability to effectively capture local spatial features via CNN and model bidirectional temporal dependencies via BiLSTM, thus demonstrating strong adaptability for complex motion scenarios. This work focuses on the optimization and comparison of deep learning algorithms for spatiotemporal feature extraction, providing a reliable framework for real-time human motion prediction and offering potential applications in intelligent gait analysis, wearable monitoring, and adaptive human–machine interaction. Full article

Accurate and efficient defect detection in printed circuit boards (PCBs) is critical for manufacturing quality control. Existing methods predominantly rely on manually extracted features such as surface texture, color, and shape for defect recognition and classification within small-dimensional feature datasets. A convolutional neural network (CNN) model was developed via transfer learning. Feature extraction involves diverse operations across different CNN layers. Essential features were selected, and dimensionality was reduced via either t-distributed stochastic neighbor embedding (t-SNE) or principal component analysis (PCA). Defect classification was subsequently performed by clustering the reduced features with either the K-means or K-nearest neighbors (KNN) algorithm. Compared with alternative model feature learning classifiers, the proposed small-dimensional CNN model performs significantly better. A defect recognition accuracy of 97.33% was achieved, with processing completed in approximately 60 s. This approach, which integrates transfer learning-based CNN feature extraction with dimensionality reduction and clustering techniques, provides a fast and effective method for high-precision defect detection and classification in PCBs. Full article

Miniature commercial trucks constitute a critical component of urban freight systems but face elevated crash risk due to distinctive driving patterns, frequent operation, and variable loads. This study quantifies how long-term and short-term driving behaviors jointly shape crash economic loss levels and identifies factors most strongly associated with severe claims. A driver-level dataset linking multi-source running behavior indicators, vehicle attributes, and insurance claims is constructed, and an enhanced Wasserstein generative adversarial network with Euclidean distance is employed to synthesize minority crash samples and alleviate class imbalance. Crash economic loss levels are modeled using a random-effects generalized ordinal logit specification, and model performance is compared with a generalized ordered logit benchmark. Marginal effects analysis is used to evaluate the influence of pre-collision driving states (straight, turning, reversing, rolling, following closely) and key behavioral indicators. Results indicate significant effects of inter-provincial duration and count ratios, morning and empty-trip frequencies, no-claim discount coefficients, and vehicle age on crash economic loss, with prolonged speeding duration and fatigued mileage associated with major losses, whereas frequent speeding and fatigue episodes are primarily linked to minor claims. These findings clarify causal patterns for miniature commercial truck crashes with different economic losses and provide an empirical basis for targeted safety interventions and refined insurance pricing. Full article

Satellite-aided multi-unmanned aerial vehicle (UAV) collaborative localization systems combine the extensive coverage of satellites with the flexibility of UAVs, offering new opportunities for locating highly dynamic emitters across large areas. However, the openness of space-air communication links and the increasing complexity of cybersecurity threats make these systems vulnerable to false data injection attacks. Most existing detection approaches focus only on temporal dependencies in time-frequency features and lack diagnostic mechanisms for identifying malicious UAVs, which limits their ability to effectively detect and mitigate such attacks. To address this issue, this paper proposes an intelligent collaborative localization framework that safeguards localization integrity by identifying and correcting false ranging information from malicious UAVs. The framework captures spatio-temporal correlations in multidimensional ranging sequences through a graph attention network (GAT) coupled with a time-attention-based variational autoencoder (VAE) to detect anomalies through anomalous distribution patterns. Malicious UAVs are further diagnosed through an anomaly scoring mechanism based on statistical analysis and reconstruction errors, while detected anomalies are corrected via a K-nearest neighbor-based (KNN) algorithm to enhance localization performance. Simulation results show that the proposed model improves localization accuracy by 25.9%, demonstrating the effectiveness of spatial–temporal feature extraction in securing collaborative localization. Full article

Background: Coronary heart disease (CHD) remains a leading cause of morbidity and mortality worldwide. Mitochondria-associated endoplasmic reticulum membranes (MAMs) have recently emerged as critical mediators in cardiovascular pathophysiology; however, their specific contributions to CHD pathogenesis remain largely unexplored. Objective: This study aimed to identify and validate MAM-related biomarkers in CHD through integrated analysis of transcriptomic sequencing data and Mendelian randomization, and to elucidate their underlying mechanisms. Methods: We analyzed two gene expression microarray datasets (GSE113079 and GSE42148) and one genome-wide association study (GWAS) dataset (ukb-d-I9_CHD) to identify differentially expressed genes (DEGs) associated with CHD. MAM-related DEGs were filtered using weighted gene co-expression network analysis (WGCNA). Functional enrichment analysis, Mendelian randomization, and machine learning algorithms were employed to identify biomarkers with direct causal relationships to CHD. A diagnostic model was constructed to evaluate the clinical utility of the identified biomarkers. Additionally, we validated the two hub genes in peripheral blood samples from CHD patients and normal controls, as well as in aortic tissue samples from a low-density lipoprotein receptor-deficient (LDLR−/−) atherosclerosis mouse model. Results: We identified 4174 DEGs, from which 3326 MAM-related DEGs (DE-MRGs) were further filtered. Mendelian randomization analysis coupled with machine learning identified two biomarkers, DHX36 and GPR68, demonstrating direct causal relationships with CHD. These biomarkers exhibited excellent diagnostic performance with areas under the receiver operating characteristic (ROC) curve exceeding 0.9. A molecular interaction network was constructed to reveal the biological pathways and molecular mechanisms involving these biomarkers. Furthermore, validation using peripheral blood from CHD patients and aortic tissues from the Ldlr−/− atherosclerosis mouse model corroborated these findings. Conclusions: This study provides evidence supporting a mechanistic link between MAM dysfunction and CHD pathogenesis, identifying candidate biomarkers that have the potential to serve as diagnostic tools and therapeutic targets for CHD. While the validated biomarkers offer valuable insights into the molecular pathways underlying disease development, additional studies are needed to confirm their clinical relevance and therapeutic potential in larger, independent cohorts. Full article

Medical image segmentation acts as a foundational element of medical image analysis. Yet its accuracy is frequently limited by the scale fluctuations of anatomical targets and the intricate contextual traits inherent in medical images—including vaguely defined structural boundaries and irregular shape distributions. To tackle these constraints, we design a multi-scale feature attention network (MFA-Net), customized specifically for thyroid nodule, skin lesion, and breast lesion segmentation tasks. This network framework integrates three core components: a Bidirectional Feature Pyramid Network (Bi-FPN), a Slim-neck structure, and the Convolutional Block Attention Module (CBAM). CBAM steers the model to prioritize boundary regions while filtering out irrelevant information, which in turn enhances segmentation precision. Bi-FPN facilitates more robust fusion of multi-scale features via iterative integration of top-down and bottom-up feature maps, supported by lateral and vertical connection pathways. The Slim-neck design is constructed to simplify the network’s architecture while effectively merging multi-scale representations of both target and background areas, thus enhancing the model’s overall performance. Validation across four public datasets covering thyroid ultrasound (TNUI-2021, TN-SCUI 2020), dermoscopy (ISIC 2016), and breast ultrasound (BUSI) shows that our method outperforms state-of-the-art segmentation approaches, achieving Dice similarity coefficients of 0.955, 0.971, 0.976, and 0.846, respectively. Additionally, the model maintains a compact parameter count of just 3.05 million and delivers an extremely fast inference latency of 1.9 milliseconds—metrics that significantly outperform those of current leading segmentation techniques. In summary, the proposed framework demonstrates strong performance in thyroid, skin, and breast lesion segmentation, delivering an optimal trade-off between high accuracy and computational efficiency. Full article

The canopy temperature of winter jujube serves as a direct indicator of plant water status and transpiration efficiency, making its accurate prediction a critical prerequisite for effective water management and optimized growth conditions in greenhouse environments. This study developed a data-driven model to forecast canopy temperature. The model serially integrates a Long Short-Term Memory (LSTM) network and a Random Forest (RF) algorithm, leveraging their complementary strengths in capturing temporal dependencies and robust nonlinear fitting. A three-stage framework comprising temporal feature extraction, multi-source feature fusion, and direct prediction was implemented to enable reliable nowcasting. Data acquisition and preprocessing were tailored to the greenhouse environment, involving multi-sensor data and thermal imagery processed with Robust Principal Component Analysis (RPCA) for dimensionality reduction. Key environmental variables were selected through Spearman correlation analysis. Experimental results demonstrated that the proposed LSTM–RF model achieved superior performance, with a determination coefficient (R²) of 0.974, mean absolute error (MAE) of 0.844 °C, and root mean square error (RMSE) of 1.155 °C, outperforming benchmark models including standalone LSTM, RF, Transformer, and TimesNet. SHAP (SHapley Additive exPlanations)-based interpretability analysis further quantified the influence of key factors, including the “thermodynamic state of air” driver group and latent temporal features, offering actionable insights for irrigation management. The model establishes a reliable, interpretable foundation for real-time water stress monitoring and precision irrigation control in protected winter jujube production systems. Full article

Prostate cancer (PCa) is a common malignancy among men worldwide. After radical prostatectomy (RP) and radical radiotherapy (RT), patients may experience biochemical recurrence (BCR) of prostate cancer, indicating disease progression. Therefore, it is meaningful to predict and accurately assess the risk of BCR, and a machine-learning-based-model for BCR prediction in PCa based on fatty-acid metabolism and cancer-cell stemness was developed. A stemness prediction model and ssGSEA (single-sample gene set enrichment analysis) empirical cumulative distribution function algorithm were used to score the stemness scoring (mRNAsi) and fatty-acid metabolism of prostate-cancer samples, respectively, and further analysis showed that the two scores of the samples were positively correlated. Based on WGCNA (weighted correlation network analysis), we discovered modules significantly associated with both stemness and fatty-acid metabolism and obtained the genes within them. Then, based on this gene set, 101 algorithm combinations of 10 machine-learning methods were used for training and prediction BCR of PCa, and the model with the best prediction effect was named fat_stemness_BCR. Compared with 23 published PCa BCR models, the fat_stemness_BCR model performs better in TCGA and CPGEA data. To facilitate the use of the model, the trained model was encapsulated into an R package and an online service tool (PCaMLmodel, Version 1.0) was built. The newly developed fat_stemness_SCR model enriches the prognostic research of biochemical recurrence in PCa and provides a new reference for the study of other diseases. Full article

Flash flood events are some of the most life-threatening natural disasters, so it is important to predict extreme rainfall events effectively. This study introduces an LSTM model that utilizes a customized loss function to effectively predict extreme rainfall events. The proposed model incorporates dynamic environmental variables, such as rainfall, LST, and NDVI, and incorporates additional static variables such as soil type and proximity to infrastructure. Wavelet transformation decomposes the time series into low- and high-frequency components to isolate long-term trends and short-term events. Model performance was compared against Random Forest (RF), Support Vector Machines (SVMs), Artificial Neural Networks (ANNs), and an LSTM-RF ensemble. The custom loss LSTM achieved the best performance (MAE = 0.022 mm/day, RMSE = 0.110 mm/day, R² = 0.807, SMAPE = 7.62%), with statistical validation via a Kruskal–Wallis ANOVA, confirming that the improvement is significant. Model uncertainty is quantified using a Bayesian MCMC framework, yielding posterior estimates and credible intervals that explicitly characterize predictive uncertainty under extreme rainfall conditions. The sensitivity analysis highlights rainfall and LST as the most influential predictors, while wavelet decomposition provides multi-scale insights into environmental dynamics. The study concludes that customized loss functions can be highly effective in extreme rainfall event prediction and thus useful in managing flash flood events. Full article

LoRa-based IoT systems are increasingly used in smart farming, greenhouse monitoring, and large-scale agricultural sensing, where long-range, energy-efficient communication is essential. However, estimating link quality metrics such as PRR, RSSI, and SNR typically requires continuous packet transmission and sequence logging, an impractical approach for power-constrained field nodes. This study proposes a deep learning-driven framework for real-time prediction of link- and network-level performance in multihop LoRa networks, targeting the LoRaDisC protocol commonly deployed in agricultural environments. By integrating Bayesian surrogate modeling with Random Forest-guided hyperparameter optimization, the system accurately predicts PRR, RSSI, and SNR using multivariate time series features. Experiments on a large-scale outdoor LoRa testbed (ChirpBox) show that aggregated link layer metrics strongly correlate with PRR, with performance influenced by environmental variables such as humidity, temperature, and field topology. The optimized model achieves a mean absolute error (MAE) of 8.83 and adapts effectively to dynamic environmental conditions. This work enables energy-efficient, autonomous communication in agricultural IoT deployments, supporting reliable field sensing, crop monitoring, livestock tracking, and other smart farming applications that depend on resilient low-power wireless connectivity. Full article

Electric vehicles (EVs) rely on a battery pack as their primary energy source, making it a critical component for their operation. To guarantee safe and correct functioning, a Battery Management System (BMS) is employed, which uses variables such as State of Charge (SOC) to set charge/discharge limits and to monitor pack health. In this article, we propose a Multilayer Perceptron (MLP) network to estimate the SOC of a 14.8 V battery pack installed in a robotic vacuum cleaner. Both offline and online (real-time) tests were conducted under continuous load and with rest intervals. The MLP’s output is compared against two commonly used approaches: NARX (Nonlinear Autoregressive Exogenous) and CNN (Convolutional Neural Network). Performance is evaluated via statistical metrics, Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE), and we also assess computational cost using Operational Intensity. Finally, we map these results onto a Roofline Model to predict how the MLP would perform on an automotive-grade microcontroller unit (MCU). A generalization analysis is performed using Transfer Learning and optimization using MLP–Kalman. The best performers are the MLP–Kalman network, which achieved an RMSE of approximately 13% relative to the true SOC, and NARX, which achieved approximately 12%. The computational cost of both is very close, making it particularly suitable for use in BMS. Full article

Accurate plant leaf image segmentation plays a crucial role in species recognition, phenotypic analysis, and disease detection. However, most segmentation models perform poorly in complex field environments due to challenges such as overlapping leaves and uneven sunlight. This research proposes an Edge-Aware High-Frequency Preservation Network (EHPNet) for leaf segmentation in complex field environments. Specifically, a High-Frequency Edge Fusion Module (HEFM) is introduced into the skip connections to preserve high-frequency edge information during feature extraction and enhance boundary localization. In addition, a Structural Recalibration Attention Module (SRAM) is incorporated into the decoder to refine edge structural features across multiple scales and retain spatial continuity, which leads to more accurate reconstruction of leaf boundaries. Experimental results on a composite dataset constructed from Pl@ntLeaves and ATLDSD show that EHPNet achieves 98.25%, 99.25%, 99.03%, 98.51%, and 98.77% in mean Intersection over Union (mIoU), accuracy, precision, recall, and F1 score, respectively. Compared with state-of-the-art methods, EHPNet achieves superior overall performance, which demonstrates its effectiveness for leaf segmentation in complex field environments. Full article

The present study examined the safety of 86 veterinary antiparasitic drugs in mammals based on their mobility in the soil–water compartment, bioconcentration factor in fish, and blood–brain barrier permeability. An in silico analysis was performed based on biomembrane permeability descriptors, using novel multiple linear regression, boosted tree, and artificial neural network models. Additionally, intestinal absorption in humans was predicted quantitatively using pkCSM software and qualitatively using SwissADME. It was established that the majority of studied drugs are at least slightly mobile in soil, are unlikely to bioaccumulate in fish, and may be absorbed from the human gastro-intestinal tract; in addition, some of them have high potential to enter the mammalian brain. Full article

Private Set Intersection (PSI) is a fundamental cryptographic primitive in privacy-preserving computation and has been widely applied in federated learning, secure data sharing, and privacy-aware data analytics. However, most existing PSI protocols rely on RSA or standard elliptic curve cryptography, which limits their applicability in scenarios requiring domestic cryptographic standards and often leads to high computational and communication overhead when processing large-scale datasets. In this paper, we propose a novel PSI protocol based on the Chinese commercial cryptographic standard SM2, referred to as SM2-OT-PSI. The proposed scheme constructs an oblivious transfer-based Oblivious Pseudorandom Function (OPRF) using SM2 public-key cryptography and the SM3 hash function, enabling efficient multi-point OPRF evaluation under the semi-honest adversary model. A formal security analysis demonstrates that the protocol satisfies privacy and correctness guarantees assuming the hardness of the Elliptic Curve Discrete Logarithm Problem. To further improve practical performance, we design a software–hardware co-design architecture that offloads SM2 scalar multiplication and SM3 hashing operations to a domestic reconfigurable cryptographic accelerator (RSP S20G). Experimental results show that, for datasets with up to millions of elements, the presented protocol significantly outperforms several representative PSI schemes in terms of execution time and communication efficiency, especially in medium and high-bandwidth network environments. The proposed SM2-OT-PSI protocol provides a practical and efficient solution for large-scale privacy-preserving set intersection under national cryptographic standards, making it suitable for deployment in real-world secure computing systems. Full article