Search Results (538)

In the information age, Internet of Things (IoT) devices are more susceptible to intrusion due to today’s complex network attack methods. Therefore, accurately detecting evolving network attacks from complex and ever-changing IoT environments has become a key research goal in the current intrusion detection field. Due to the spatial and temporal characteristics of IoT data, this paper proposes a Spatiotemporal Feature Weighted Fusion Approach Combining Gating Attention Transformation (STWGA). STWGA consists of three parts, namely spatial feature learning, the gated attention transformer, and the temporal feature learning module. It integrates improved convolutional neural networks (CNN), batch normalization, and Bidirectional Long Short-Term Memory Network (Bi-LSTM) to fully learn the deep spatial and temporal features of the data, achieving the goal of global deep spatiotemporal feature extraction. The gated attention transformer introduces an attention mechanism. In addition, an additional control mechanism is introduced in the self-attention module to more effectively improve detection accuracy. Finally, the experimental results show that STWGA has better spatiotemporal feature extraction ability and can effectively improve the intrusion detection effect of anomalies. Full article

Neural networks significantly outperform traditional methods in both decoding amplitude-, phase-, and polarization-encoded data pages and suppressing noise within them. However, the mechanism behind neural networks’ denoising capability remains not fully understood. We discover that zeroing channels can improve the reconstruction effect of the model. Consequently, this paper presents a method to locate the noise feature objectively from γ, the weights of the Batch Normalization (BN) layer. γ stands for the importance of the channel in the model and γ < 1 means the channel may contain noise feature. Through experiments, removing the channels that contained a higher proportion of these noisy features, the reconstructed data pages showed a ~2% improvement in Peak Signal-to-Noise Ratio (PSNR) compared to results obtained by directly outputting data without removing the noisy channels. It indicates that neural networks achieve efficient denoising of encoded data pages by adjusting the weight parameters of BN layers, thereby suppressing or enhancing specific channels. Full article

Remaining useful life (RUL) prediction is central to predictive maintenance, but acquiring sufficient run-to-failure data remains challenging. To better exploit limited labeled data, this study investigates a pipeline combining an unsupervised autoencoder (AE) and supervised LSTM regression on the NASA C-MAPSS dataset. Building on an AE-LSTM baseline, we analyze how window size and batch size affect accuracy and training efficiency. Using the FD001 and FD004 subsets with training-capped RUL labels, we perform multi-seed experiments over a wide grid of window lengths and batch sizes. The AE is pre-trained on normalized sensor streams and reused as a feature extractor, while the LSTM head is trained with early stopping. Performance was assessed using RMSE, C-MAPSS score, and training time, reporting 95% confidence intervals. Results show that fine-tuning the encoder with a batch size of 128 yielded the best mean RMSE of 13.99 (FD001) and 28.67 (FD004). We obtained stable optimal window ranges (40–70 for FD001; 60–80 for FD004) and found that batch sizes of 64–256 offer the best accuracy–efficiency trade-off. These optimal ranges were further validated using Particle Swarm Optimization (PSO). These findings offer practical recommendations for tuning AE-LSTM-based RUL prediction models and demonstrate that performance remains stable within specific hyperparameter ranges. Full article

Liver fibrosis induced by Schistosoma japonicum Katsurada, 1904 (S. japonicum) infection lacks effective diagnostic markers and specific anti-fibrotic therapies. Although dysregulated iron homeostasis and ferroptosis pathways may contribute to its pathogenesis, the core regulatory mechanisms remain elusive. To unravel the ferroptosis-related molecular features, this study integrated transcriptomic datasets (GSE25713 and GSE59276) from S. japonicum-infected mouse livers. Following batch effect correction and normalization, ferroptosis-related differentially expressed genes (FRDEGs) were identified. Subsequently, core hub genes were screened through the construction of a protein–protein interaction (PPI) network, functional enrichment analysis, immune infiltration evaluation, and receiver operating characteristic (ROC) analysis. The expression patterns of these hub genes were further validated in an S. japonicum-infected mouse model using reverse transcription-quantitative polymerase chain reaction (RT-qPCR). The study identified 7 hub genes (Lcn2, Timp1, Cth, Cp, Hmox1, Cbs, and Gclc) as key regulatory molecules. Functional enrichment analysis revealed that these hub genes are closely associated with sulfur amino acid metabolism and oxidative stress responses. Specifically, key enzymes involved in cysteine and glutathione (GSH) synthesis (Cth, Cbs, Gclc) were consistently downregulated, suggesting a severe impairment of the host antioxidant defense capacity. Conversely, pro-fibrotic and pro-inflammatory markers (Timp1, Lcn2, Hmox1) were upregulated. This molecular pattern was significantly associated with a remodeled immune microenvironment, characterized by increased infiltration of neutrophils and eosinophils. In vivo validation confirmed the expression trends of 6 hub genes, corroborating the bioinformatics predictions, while the discrepancy in Cp expression highlighted the complexity of post-transcriptional regulation in vivo. The identified hub genes demonstrated excellent diagnostic potential, with Timp1 achieving an area under the curve (AUC) of 1.000. This study elucidates the molecular link between S. japonicum infection and the ferroptosis pathway, suggesting that these hub genes may drive liver fibrosis progression by regulating sulfur metabolism and the immune microenvironment. These findings offer potential diagnostic biomarkers and novel therapeutic targets for schistosomal liver fibrosis. Full article

In order to achieve better accuracy, modern models have become increasingly large, leading to an exponential increase in computational load, making it challenging to apply them to edge computing. Binary neural networks (BNNs) are models that quantize the filter weights and activations to 1-bit. These models are highly suitable for small chips like advanced RISC machines (ARMs), field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), system-on-chips (SoCs) and other edge computing devices. To design a model that is more friendly to edge computing devices, it is crucial to reduce the floating-point operations (FLOPs). Batch normalization (BN) is an essential tool for binary neural networks; however, when convolution layers are quantized to 1-bit, the floating-point computation cost of BN layers becomes significantly high. This paper aims to reduce the floating-point operations by removing the BN layers from the model and introducing the scaled weight standardization convolution (WS-Conv) method to avoid the significant accuracy drop caused by the absence of BN layers, and to enhance the model performance through a series of optimizations, adaptive gradient clipping (AGC) and knowledge distillation (KD). Specifically, our model maintains a competitive computational cost and accuracy, even without BN layers. Furthermore, by incorporating a series of training methods, the model’s accuracy on CIFAR-100 is 0.6% higher than the baseline model, fractional activation BNN (FracBNN), while the total computational load is only 46% of the baseline model. With unchanged binary operations (BOPs), the FLOPs are reduced to nearly zero, making it more suitable for embedded platforms like FPGAs or other edge computers. Full article

Automatic sleep-stage classification is a key component of long-term sleep monitoring and digital health applications. Although deep learning models trained on centralized datasets have achieved strong performance, their deployment in real-world healthcare settings is constrained by privacy, data-governance, and regulatory requirements. Federated learning (FL) addresses these issues by enabling decentralized training in which raw data remain local and only model parameters are exchanged; however, its effectiveness under realistic physiological heterogeneity remains insufficiently understood. In this work, we investigate a subject-level federated deep learning framework for sleep-stage classification using polysomnography data from the ISRUC-Sleep dataset. We adopt a realistic one subject = one client setting spanning three clinically distinct subgroups and evaluate a lightweight one-dimensional convolutional neural network (1D-CNN) under four training regimes: a centralized baseline and three federated strategies (FedAvg, FedProx, and FedBN), all sharing identical architecture and preprocessing. The centralized model, trained on a cohort with regular sleep architecture, achieves stable performance (accuracy 69.65%, macro-F1 0.6537). In contrast, naive FedAvg fails to converge under subject-level non-IID data (accuracy 14.21%, macro-F1 0.0601), with minority stages such as N1 and REM largely lost. FedProx yields only marginal improvement, while FedBN—by preserving client-specific batch-normalization statistics—achieves the best federated performance (accuracy 26.04%, macro-F1 0.1732) and greater stability across clients. These findings indicate that the main limitation of FL for sleep staging lies in physiological heterogeneity rather than model capacity, highlighting the need for heterogeneity-aware strategies in privacy-preserving sleep analytics. Full article

An algorithm named ASR-BL-SA is proposed to solve the impact of a rectangular-part nesting sequence on final material utilization. Based on the Bottom Left principle, a coefficient, k, is defined as the ratio of the shape factor to 0.785 plus the square root of the min–max-normalized area. Parts are sorted in descending order of k. To tackle the flexible adaptation of part width and height via 90° rotation for sheet size and irregular leftover space, the Bottom Left algorithm initially compares utilization of original and rotated placements, selecting the option with higher utilization at each step. Finally, simulated annealing is applied for optimization. Experiments show that in the small-batch test, the proposed algorithm improves utilization by 5.51%, 3.75%, 8.84%, 5.51%, and 3.75% compared to the three baselines; in the mass production test, the improvements are 1.74%, 7.98%, 2.6%, 1.74%, and 7.89% within an acceptable time; in general applicability Test 3, its utilization is basically higher than the five comparative algorithms, achieving certain improvements in utilization. Full article

Background/Objectives: Acute health events in institutionalized older adults often lead to avoidable hospital referrals, requiring rapid, accurate remote decision-making. Telemedicine has become a key tool to improve assessment and care continuity in nursing homes. This study aimed to evaluate outcomes associated with telemedicine-supported management of acute events in residential care facilities for older adults and to develop a deep learning model to classify episodes and predict hospital referrals. Methods: A quasi-experimental study analyzed 5202 acute events managed via a 24/7 telemedicine system in Vitalia nursing homes (January–October 2024). The dataset included demographics, comorbidities, vital signs, event characteristics, and outcomes. Data preprocessing involved imputation, normalization, encoding, and dimensionality reduction via Truncated SVD (200 components). Given the imbalance in referral outcomes (~10%), several resampling techniques (SMOTE, SMOTEENN, SMOTETomek) were applied. A deep feedforward neural network (256–128–64 units with Batch Normalization, LeakyReLU, Dropout, AdamW) was trained using stratified splits (70/10/20) and optimized via cross-validation. Results: Telemedicine enabled the resolution of approximately 90% of acute events within the residential setting, reducing reliance on emergency services. The deep learning model outperformed traditional algorithms, achieving its best performance with SMOTEENN preprocessing (AUC = 0.91, accuracy = 0.88). The proposed model achieved higher overall performance than baseline classifiers, providing a more balanced precision–specificity trade-off for hospital referral prediction, with an F1-score of 0.63. Conclusions: Telemedicine-enabled acute care, strengthened by a robust deep learning classifier, offers a reliable strategy to enhance triage accuracy, reduce unnecessary transfers, and optimize clinical decision-making in nursing homes. These findings support the integration of AI-assisted telemedicine systems into long-term care workflows. Full article

Convolutional Neural Networks (CNNs) generally suffer from excessive computational overhead, high resource consumption, and complex network structures, which severely restrict the deployment on microprocessor chips. Existing related accelerators only have an energy efficiency ratio of 2.32–6.5925 GOPs/W, making it difficult to meet the low-power requirements of embedded application scenarios. To address these issues, this paper proposes a low-power and high-energy-efficiency CNN accelerator architecture based on a central processing unit (CPU) and an Application-Specific Integrated Circuit (ASIC) heterogeneous computing architecture, adopting an operator-fused systolic array algorithm with the YOLOv5n target detection network as the application benchmark. It integrates a 2D systolic array with Conv-BN fusion technology to achieve deep operator fusion of convolution, batch normalization and activation functions; optimizes the RISC-V core to reduce resource usage; and adopts a locking mechanism and a prefetching strategy for the asynchronous platform to ensure operational stability. Experiments on the Nexys Video development board show that the architecture achieves 20.6 GFLOPs of computational performance, 1.96 W of power consumption, and 10.46 GOPs/W of energy efficiency ratio, which is 58–350% higher than existing mainstream accelerators, thus demonstrating excellent potential for embedded deployment. Full article

Predicting neoadjuvant chemotherapy response in breast cancer remains critical for optimizing treatment strategies, yet robust predictive biomarkers are lacking. This study implemented an ensemble machine learning approach to identify a gene expression signature predicting pathological complete response (pCR) versus residual disease (RD) using bulk RNA-sequencing data from GSE163882 (138 RD, 80 pCR). We employed TMM normalization with differential expression analysis (250 genes, FDR < 0.05, |log2FC| ≥ 1), ensemble feature selection across five classifiers (Random Forest, Gradient Boosting, SVM, k-NN, and Neural Network) with 10-fold repeated cross-validation, and stacked ensemble development. Consensus selection identified a 17-gene signature consistently ranked across algorithms. The stacked ensemble achieved 0.97 AUC post-testing on hold-out test data. External validation on the independent GSE240671 cohort (37 pCR, 25 RD) following ComBat batch correction achieved ROC AUC of 0.78 and PR AUC of 0.85 with isotonic calibration, demonstrating balanced accuracy of 0.71 and 0.86 sensitivity for pCR detection. Pathway enrichment revealed associations with cell cycle regulation (E2F3, MKI67), DNA repair (BRCA2), and transcriptional control (MED1), with six priority genes (MED1, BRCA2, E2F3, PITPNB, H1-1, and FARP2) showing established breast cancer relevance. This externally validated 17-gene signature provides a biologically grounded tool for NAC response prediction in precision oncology. Full article

During batch soldering, assembly of micro image sensor modules, initial random pose, and feature partially occlude target micro-component image, leading to issues of missed and erroneous detection, and low 3D spatial positioning accuracy due to cross-depth-of-field detection errors in microscopic vision. This paper proposes Small object-YOLO11-Cross-Depth-of-field Positioning (SO-YOLO11-CDP), an instance segmentation-based approach for precision cross-depth-of-field positioning micro-component. First, an improved Small object-YOLO11 (SO-YOLO11) image segmentation algorithm is designed. By incorporating a coordinate attention mechanism (CA) into segmentation head to enhance localization of micro-targets, the backbone uses non-stride convolution to preserve fine-grained feature, while target regression performance is boosted via Efficient-IoU (EIoU) loss combined with normalized Wasserstein distance (NWD). Subsequently, to further improve spatial position detection accuracy in cross-depth-of-field detection, a calibration error compensation model for image Jacobian matrix is established based on pinhole imaging principles. Experimental results indicate that SO-YOLO11 achieves 16.1% increase in precision, 4.0% increase in recall, and 9.9% increase in mean average precision (mAP0.5) over baseline YOLO11. Furthermore, it accomplishes spatial detection accuracy superior to 6.5 μm for target micro-components. The method presented in this paper holds significant engineering application value for high-precision spatial position detection of micro image sensor components. Full article

In modern e-learning environments, ensuring both student identity verification and concentration monitoring during online examinations has become increasingly important. This paper introduces a robust sequential framework that integrates Generative Adversarial Networks (GANs), Convolutional Neural Networks (CNNs) and fuzzy logic to achieve reliable face recognition and interpretable attentiveness assessment. Images from the Extended Yale B (cropped) dataset are preprocessed through grayscale normalization and resizing, while GANs generate synthetic variations in pose, illumination, and occlusion to enrich the training set and improve generalization. The CNN extracts discriminative facial features for identity recognition, and a fuzzy inference system transforms the CNN’s confidence scores into human-interpretable concentration levels. To stabilize learning and prevent overfitting, the model incorporates dropout regularization, batch normalization, and extensive data augmentation. Comprehensive evaluations using confusion matrices, ROC–AUC, and precision–recall analyses demonstrate an accuracy of 98.42%. The proposed framework offers a scalable and interpretable solution for secure and reliable online exam proctoring. Full article

Radiographically visible carotid artery calcifications are typically seen at the level of the C3–C4 cervical vertebrae and can be detected on panoramic dental radiographs. Their early identification is clinically relevant, as they represent a potential marker for increased risk of stroke. In this context, the present study proposes a deep learning method for automatic identification of carotid atheromas using MobileNetV2. From a publicly available dataset, 378 region-of-interest (ROI) images (640 × 320) were prepared and split into train/val/test = 264/57/57 with class counts train 157/107, val 34/23, test 34/23 (negatives/positives). Images underwent standardized preprocessing and on-the-fly augmentation; training used a two-stage scheme (backbone frozen “head” training followed by partial fine-tuning of the top layers), class-weighting, dropout = 0.3, batch normalization (BN) head, early stopping, and partial unfreezing (~70% of the backbone). The decision threshold was selected on validation by Youden’s J. On the independent test set, the model achieved an accuracy (ACC) of 94.7%, sensitivity (SEN) of 95,7%, specificity (SPE) of 0.941, area under the receiver operating characteristic curve (AUC) 0.963, and area under the precision–recall curve (AUPRC) of 0.968. Using a sensitivity-targeted threshold (SEN ≈ 0.80), the model yielded ACC = 91.2%, SEN = 82.6%, and SPE = 97.1%. These results support panoramic radiographs as an opportunistic screening modality for systemic vascular risk and highlight the potential of artificial intelligence (AI)-assisted methods to enable earlier identification within preventive healthcare. Full article

Marine oil spill detection using Synthetic Aperture Radar (SAR) is crucial but challenged by dynamic marine conditions, diverse spill scales, and limitations in existing algorithms regarding model size and real-time performance. To address these challenges, we propose LSFE-YOLO, a YOLOv8s-optimized (You Only Look Once version 8) lightweight model with an original, domain-tailored synergistic integration of FasterNet, GN-LSC Head (GroupNorm Lightweight Shared Convolution Head), and C2f_MBE (C2f Mobile Bottleneck Enhanced). FasterNet serves as the backbone (25% neck width reduction), leveraging partial convolution (PConv) to minimize memory access and redundant computations—overcoming traditional lightweight backbones’ high memory overhead—laying the foundation for real-time deployment while preserving feature extraction. The proposed GN-LSC Head replaces YOLOv8’s decoupled head: its shared convolutions reduce parameter redundancy by approximately 40%, and GroupNorm (Group Normalization) ensures stable accuracy under edge computing’s small-batch constraints, outperforming BatchNorm (Batch Normalization) in resource-limited scenarios. The C2f_MBE module integrates EffectiveSE (Effective Squeeze and Excitation)-optimized MBConv (Mobile Inverted Bottleneck Convolution) into C2f: MBConv’s inverted-residual design enhances multi-scale feature capture, while lightweight EffectiveSE strengthens discriminative oil spill features without extra computation, addressing the original C2f’s scale variability insufficiency. Additionally, an SE (Squeeze and Excitation) attention mechanism embedded upstream of SPPF (Spatial Pyramid Pooling Fast) suppresses background interference (e.g., waves, biological oil films), synergizing with FasterNet and C2f_MBE to form a cascaded feature optimization pipeline that refines representations throughout the model. Experimental results show that LSFE-YOLO improves mAP (mean Average Precision) by 1.3% and F1 score by 1.7% over YOLOv8s, while achieving substantial reductions in model size (81.9%), parameter count (82.9%), and computational cost (84.2%), alongside a 20 FPS (Frames Per Second) increase in detection speed. LSFE-YOLO offers an efficient and effective solution for real-time marine oil spill detection. Full article

The performance of smartphone-based pedestrian positioning strongly depends on the GNSS signal quality, the motion dynamics that influence PDR accuracy, and the way both sources of information are fused. While recent studies have shown the benefits of Factor Graph Optimization (FGO) for Pedestrian Dead Reckoning (PDR) Global Navigation Satellite Systems (GNSS) fusion, the interaction between human motion, PDR errors, and FGO window configuration has not been systematically examined. This work investigates how walking dynamics affect the optimal configuration of sliding-window FGO, and to what extent FGO mitigates motion-dependent PDR errors compared with the Kalman Filter (KF). Using data collected from ten pedestrians performing four motion types (slow walking, normal walking, jogging, and running), we analyze: (1) the relationship between walking speed and the FGO window size required to achieve stable positioning accuracy, and (2) the ability of FGO to suppress PDR outliers arising from motion irregularities across different users. The results show that a window size of around 10 poses offers the best overall balance between accuracy and computational load, providing substantial improvement over SWFGO with a 1-pose window and approaching the accuracy of batch FGO at a fraction of its cost. Increasing the window further to 30 poses yields only marginal accuracy gains while increasing computation, and this trend is consistent across all motion types. Additionally, FGO and SWFGO reduce PDR-induced outliers more effectively than KF across all users and motions, demonstrating improved robustness under gait variability and transient disturbances. Full article