Search Results (1,787)

IoT networks now handle traffic from billions of devices, and edge nodes are under constant pressure to classify that traffic and dispatch tasks within tight latency deadlines. Most existing systems treat classification and scheduling as two separate steps that run one after the other. This sequence adds unnecessary delay and breaks the feedback between the two tasks: the scheduler never sees the traffic type, and the classifier never sees the queue state. We propose ParallelEdge-AI, a system built around a shared flow encoder that feeds two task-specific heads in parallel, one for multi-class traffic classification and one for task-urgency scoring. Both heads are trained end-to-end using a joint loss that combines cross-entropy and pairwise ranking. A load-balance controller then reads the urgency scores alongside live queue lengths to decide, every 200 ms, whether a task stays local or moves to a less-loaded edge node. No global synchronisation is needed. We test the system on three real IoT datasets: RT-IoT2022, N-BaIoT, and CICIoT2023. ParallelEdge-AI reaches 97.63% accuracy and an F1-score of 97.34%, which is 3.16 percentage points above the best baseline. Inference latency is 19.62 ms per batch, the deadline-miss rate is 2.34%, and the load-imbalance index is 0.083, all three are the best results in our comparison. These numbers show that running classification and scheduling together on a shared representation is both faster and more accurate than treating them as separate problems. Full article

Vehicular Internet of Things (V-IoT) networks require reliable scheduling for safety-critical communication, cooperative awareness, and cooperative perception under dynamic mobility and limited roadside infrastructure. This paper proposes MoReSP, a Mobility- and Reliability-aware Scheduling Policy for roadside unit (RSU)-assisted V-IoT networks. MoReSP uses mobility-regime inference, structured action scoring, safety projection, and episodic parameter adaptation to select among deny, grant, preempt, coexist, and handoff actions. Its multiobjective formulation jointly minimizes average delay, communication energy consumption, admission-adjusted reliability loss, a penalty for cooperative perception message (CPM) delivery/freshness, and RSU-load imbalance. The framework is evaluated under vehicle-load variation, Nagel–Schreckenberg (NaSch) density variation, and RSU-capacity scaling using admission-adjusted metrics that penalize excessive blocking and interruption. MoReSP is compared with five literature-grounded benchmark families: Age of Correlated Information (AoCI)-Heuristic, RSU-Coop, Handoff-Aware, vehicle-to-everything (V2X)-Priority, and Adaptive Learning-based Task Offloading multi-armed bandit (ALTO-MAB). Simulation results show that MoReSP achieves the lowest admission-adjusted system cost across all evaluated scenarios. At nominal RSU capacity, MoReSP reduces the system cost by 43.6% compared with the best baseline. Under high vehicle load, it reduces the cost by 54.4% at arrival scale 2.0 and maintains effective packet and CPM delivery ratios of 0.849 and 0.828, respectively. These results demonstrate that MoReSP provides a reliable and balanced scheduling solution for dynamic V-IoT environments. Full article

Maritime accident analysis increasingly uses machine learning to support safety management, but many existing studies focus on single-output prediction, such as accident-occurrence probability, severity class, near-miss frequency, or one specific consequence. This study proposes a data-driven decision-support framework based on a Multi-Input Multi-Output Artificial Neural Network (MIMO-ANN) for the simultaneous prediction of multiple maritime accident consequences. A dataset of 582 recorded accident cases is constructed by integrating SafePASS project records with consequence, severity, and structural-damage information from the literature. The dataset includes 15 input variables covering ship characteristics, operational context, environmental conditions, accident type, and geographical zone and 15 consequence outputs covering structural damage, casualties, emergency-response indicators, total loss, and secondary consequence/escalation mechanisms. The ANN is trained using the Scaled Conjugate Gradient (SCG) algorithm and evaluated under different network configurations and data-partitioning strategies. The best-performing model uses 30 hidden neurons with a 60/20/20 split, achieving a correlation coefficient (R) equal to 0.9249 and a mean squared error (MSE) equal to 0.0240 for testing, and a R equal to 0.9278 and a MSE equal to 0.0231 for validation. Ten-fold cross-validation further confirms internal predictive stability, with mean testing R equal to 0.8803 ± 0.0827 and MSE equal to 0.0445 ± 0.0478. Permutation-based sensitivity analysis shows that accident type, zone, flag, natural light, environment, and visibility are key drivers of predicted consequences, whereas vessel-specific parameters have a secondary, context-dependent influence. The framework should be interpreted as predicting the relative likelihood, severity, or magnitude of accident consequences in recorded or scenario-defined accident cases, not the probability of accident occurrence. Future work should address dataset imbalance, include near-miss and nonserious records, incorporate richer AIS and metocean data, integrate exposure data, and validate the framework using independent accident datasets. Full article

Routine inspections of transmission lines are essential for maintaining the reliability of the power grid. Airborne LiDAR technology provides detailed 3D corridor data for automated hazard detection, such as vegetation encroachment and structural anomalies. However, manually analyzing large point clouds is inefficient, and current segmentation methods struggle with scene complexity, scale variation, and the high cost of annotation. In this study, we present a label-efficient segmentation method built on MinkUNet, a sparse voxel convolutional network enhanced with self-attention modules in its encoder–decoder for better spatial reasoning over corridor objects (e.g., trees, buildings, towers). To further handle structural diversity and class imbalance, we adopt task-specific data augmentations and focal loss. A multi-stage pseudo-labeling strategy is then employed to enable effective cross-scene generalization with minimal labeled data. We validate our method on three real-world transmission line datasets. On the Foshan dataset, it achieves a mean Intersection over Union (mIoU) of 0.740 with an inference time of 1.31 s. Cross-scene tests at two other locations, Shumuyuan and Langwang Village, yield mIoUs of 0.762 and 0.757, respectively. These results confirm robust performance even with limited annotations. Overall, our findings demonstrate the practicality of our approach for routine power line inspections, enabling reliable hazard detection with minimal annotation effort. Full article

The increasing integration of renewable energy resources in microgrids requires effective frameworks for energy management, system monitoring, and operational assessment. This study presents a simulation-based digital twin-oriented framework for a small-scale renewable energy microgrid with integrated energy management control. The framework consists of a solar photovoltaic (PV) system, a lithium-ion battery energy storage system, and a variable load implemented in a MATLAB/Simulink 2024b environment. Mathematical models are developed to represent PV generation, battery state-of-charge (SOC) dynamics, and load variations, while a rule-based energy management strategy is used to regulate power flow between generation, storage, and demand. An interactive dashboard is incorporated to provide dynamic visualization within the simulation environment of the system operation and key performance indicators. Simulation results show that the controller successfully maintains the battery SOC within the safe operating range of 30–90% and eliminates SOC constraint violations. Compared with uncontrolled operation, renewable energy utilization increases from 67.4% to 92.8%, overall system efficiency improves from 79.6% to 91.3%, and system reliability increases from 93.1% to 99.2%. The Loss of Power Supply Probability (LPSP) decreases from 0.069 to 0.008, while RMS power imbalance is reduced by 50.0%. Battery and converter losses decrease by 41.7% and 43%, respectively. These results demonstrate the effectiveness of the proposed framework in improving energy utilization, reliability, and operational stability while providing a foundation for future digital twin-enabled microgrid optimization and decision support applications. Full article

Small and severely imbalanced optical coherence tomography (OCT) datasets pose a major challenge for deep learning algorithms while reflecting the reality of smaller ophthalmology centers, where rare retinal pathologies such as retinal artery occlusion (RAO) and vitreomacular interface disease (VID) occur only sporadically. The experiments were performed on the imbalanced OCTDL dataset (seven classes, 2064 images) and on a custom imbalanced subset of OCT-C8 (eight classes) matching the OCTDL class distribution. An extended ablation of eight loss functions was carried out under 10-fold stratified cross-validation (CV). As the loss for the final Swin Transformer Base model, we adopt dual-weighted PolyLoss (DW-PolyLoss), a lightweight modification of PolyLoss in which inverse-frequency class weights are applied symmetrically to both the cross-entropy and the polynomial correction terms. Under 10-fold stratified CV, the mean OCTDL accuracy is 95.63%, and a logit-averaging ensemble of the 10-fold models reaches 96.71% accuracy on OCTDL and 96.21% on a custom imbalanced subset of OCT-C8 constructed to match the OCTDL class distribution. Score-CAM analysis suggests that the model attends to clinically interpretable retinal structures, supporting potential use as a screening-triage tool subject to prospective clinical validation. Full article

Background/Objectives: Following the LACC trial, the role of minimally invasive radical surgery for early-stage cervical cancer remains controversial. Robotic-assisted approaches have been proposed as a potential strategy to preserve the benefits of minimally invasive surgery while incorporating contemporary oncologic precautions. This study compared perioperative, pathological, and early oncologic outcomes between robotic and open radical surgical management in patients with FIGO 2018 stage IA2–IIA1 cervical cancer. Methods: Patients underwent robotic surgery (n = 20; da Vinci Xi), including robotic radical hysterectomy, compartment-based procedures, and fertility-sparing surgery when clinically indicated, or open abdominal radical hysterectomy (n = 22). Perioperative outcomes, histopathological parameters (including lymphovascular space invasion [LVSI], lymph node status, and margin status), and early oncologic outcomes were evaluated. Exploratory multivariable regression analyses were performed to adjust for baseline differences, including age and tumor size. Results: Patients in the open-surgery cohort were older (56.23 ± 15.87 vs. 45.67 ± 9.31 years; p = 0.012) and had significantly larger tumors (3.07 ± 1.10 vs. 1.4 ± 0.7 cm; p = 0.003). Robotic surgery was associated with longer operative time (178 ± 42 vs. 150 ± 38 min; p = 0.028), lower blood loss (112 ± 61 vs. 518 ± 98 mL; p < 0.001), and shorter hospital stay (4.2 ± 1.6 vs. 6.2 ± 1.4 days; p < 0.001). The robotic cohort also demonstrated a higher lymph node yield (median 18 vs. 9; p < 0.001). No statistically significant differences were observed between groups in lymph node metastasis (20.0% vs. 22.7%; p = 1.000), LVSI (33.3% vs. 63.6%; p = 0.121), or R0 resection rate (100% vs. 95.5%; p = 1.000). In exploratory adjusted analyses, surgical approach was not associated with adverse pathological features, whereas tumor size emerged as an independent predictor of both lymph node metastasis and LVSI. No recurrences were observed in the robotic cohort during the available follow-up period. Conclusions: In this exploratory comparative cohort study, robotic radical surgical management in carefully selected patients with predominantly small-volume disease was associated with favorable perioperative outcomes and no statistically significant differences in pathological parameters compared with open surgery. Tumor size, rather than surgical approach, emerged as the principal predictor of adverse pathological features. Given the limited sample size, baseline imbalances between cohorts, heterogeneous robotic procedures, and absence of mature survival data, these findings should not be interpreted as evidence of oncologic equivalence and require confirmation in larger prospective studies. Full article

Turner syndrome (TS), a disorder caused by the complete or partial absence of an X chromosome, exhibits significant clinical variability that cannot be fully explained by chromosomal anomalies alone. This narrative review highlights the crucial role of epigenetic mechanisms, particularly X-chromosome inactivation (XCI), in shaping the TS phenotype. The haploinsufficiency of genes that normally escape XCI is a primary driver of TS features. The specific epigenetic consequences depend on the chromosomal anomaly. In complete monosomy (45,X), the absence of escape-mediated dosage compensation genes from a second X chromosome amplifies haploinsufficiency across X-linked escape genes. Isochromosome Xq (i(Xq)) variants involve the loss of the short arm (Xp) and duplication of the long arm (Xq), creating a dual dosage imbalance with extreme XCI skewing. Carriers of i(Xq) also have a heightened risk for autoimmune disorders compared to those with 45,X TS. For ring-X chromosomes (r(X)), which are mitotically unstable, the functional status of the XIST gene is critical. If the ring is XIST-negative, it remains transcriptionally active, resulting in functional disomy and a more severe phenotype with pronounced neurodevelopmental and craniofacial features. Ultimately, the clinical heterogeneity in TS arises from a complex interplay of the specific chromosomal structure, tissue-specific mosaicism, XIST function, and variable escape from XCI, defining TS as a disorder of epigenetic and gene-regulatory imbalance. However, future research requires a better understanding of the complex mechanism of X-chromosome inactivation. Full article

Aero-engine blades operate under extreme conditions involving high temperature, pressure, rotational speed, and cyclic loads, making them susceptible to surface defects such as micro-cracks. Due to their small scale, weak edges, low contrast, and elongated morphology, micro-cracks are easily affected by metallic reflections, uneven illumination, and complex background textures in borescope images, resulting in high missed-detection rates for conventional detection methods. To address these challenges, this study proposes an improved YOLO11-based framework for aero-engine blade micro-crack detection. The proposed method introduces P1/P2 shallow high-resolution detection branches to enhance the perception of fine crack edges and textures, incorporates Focal Loss to alleviate foreground–background imbalance, applies object-level Tversky Loss to strengthen false-negative constraints, and adopts a hard mining strategy to improve learning for difficult crack samples. Experiments conducted on a real aero-engine borescope image dataset demonstrate that the proposed model achieves a Precision of 0.9981, Recall of 0.9606, F1-score of 0.9790, mAP50 of 0.9781, and mAP50-95 of 0.6938 on an independent test set. Compared with the YOLO11 baseline, the proposed method significantly improves crack detection accuracy, localization quality, and robustness in complex borescope inspection scenarios. Full article

With the advancement of the carbon peaking and carbon neutrality goals, large-scale grid integration of photovoltaic (PV) systems has become a core trend in the development of distribution networks (DNs). However, this high penetration breaks the inherent spatiotemporal symmetry of power flow in traditional DNs, leading to severe spatiotemporal imbalance issues, including voltage violations, reverse power flow, and a sharp increase in network power loss. To address these challenges, an optimized flexible control method for PV inverters in DNs considering cluster partitioning is proposed in this paper. First, a comprehensive performance index system integrating improved modularity, source-load matching degree, and voltage sensitivity is constructed, which quantifies the electrical coupling symmetry and source-load power symmetry within clusters, providing a rigorous quantitative basis for dynamic cluster partitioning. Moreover, based on a dynamic monitoring mechanism, an improved Particle Swarm Optimization algorithm for cluster partitioning is proposed to achieve the optimal cluster partitioning of DN nodes and the selection of key control nodes. Finally, a Q-V flexible control model of the inverter adapted to cluster control is established; thus, an optimization model with the objectives of minimizing voltage deviation, PV curtailment loss, and PV reactive power output is constructed. The distributed and efficient solution is performed using the Alternating Direction Method of Multipliers algorithm and the GUROBI solver. Simulation results based on the modified IEEE 123-node test feeder show that, compared with traditional methods, the proposed method improves the cluster partitioning effectiveness, ensures that the operating voltage deviation of the control system is within 5%, and reduces the PV curtailment loss of the system. Full article

Accurate pixel-level segmentation of weld defects in radiographic images is essential for automated non-destructive testing (NDT) and quantitative weld-quality assessment. However, this task remains challenging because weld defects often exhibit severe foreground–background imbalance, ambiguous boundaries, weak grayscale contrast, and defect-like weld background structures, which can lead to boundary over-expansion and false-positive predictions. To address these issues, this paper proposes a boundary-guided feature modulation framework with false-positive suppression for weld radiographic defect segmentation. The method constructs boundary bands from training annotations and uses them only as label-derived training-time regularizers for attention-driven feature modulation and region-aware optimization; during validation and testing, no ground-truth mask, boundary band, or predicted boundary map is provided to the model. Multi-scale feature fusion is used to recover weak defect responses, boundary-guided dual attention enhances boundary-sensitive feature representation, and a false-positive suppression loss penalizes foreground leakage above a tolerated confidence margin in stable non-boundary background regions. Experiments on a real-world pipeline weld radiographic dataset containing 10,079 images show that the proposed method achieves a Dice score of $0.810\pm 0.003$, a Precision of $0.809\pm 0.004$, and a Surface Dice at 3 pixels of $0.394\pm 0.008$, outperforming representative CNN-based and Transformer-based segmentation baselines. Ablation studies, qualitative visualization, and distance-based false-positive analysis further demonstrate that the proposed framework improves contour reliability and reduces background false positives. Full article

We propose a theoretical framework based on an affine-eigenstate transformation for analyzing ultracold atom–molecule conversion dynamics with particle loss. The transformation maps the mean-field dynamics to an effective two-mode representation in which fixed points, Bloch-sphere trajectories, and linear stability can be examined in a common set of variables. We give the derivation of the transformed Hamiltonian and specify the invertibility and conjugate-condition requirements under which the mapping is used. Within this representation, we distinguish ordinary, pseudo, and strict self-trapping regimes. The strict regime is associated with the balanced condition $S=0$ in the transformed variables; in the corresponding linearized dissipative flow, the leading attractor/repeller bifurcation term controlled by $S{\mathsf{\Gamma }}_{-}$ vanishes, explaining the observed robustness against atom- and molecule-loss imbalance. We also introduce von Neumann and linear-entropy diagnostics for future mixed-state or ensemble descriptions in the transformed two-level representation, and we provide an inverse reconstruction procedure for preparing initial states that realize strict self-trapping. Finally, we discuss the limits of the mean-field and Markovian approximations and outline how finite-particle simulations and phase-modulated control protocols could connect this mechanism to decoherence-resilient quantum simulations and information-processing architectures. Full article

Sepsis is responsible for approximately 270,000 deaths annually in the United States. Conventional scoring systems, such as SOFA and qSOFA, are largely reactive and do not effectively leverage longitudinal ICU data for early prediction. This study aims to develop a deep learning framework capable of predicting sepsis onset up to 6 h before Sepsis-3 criteria are met while also providing clinically interpretable temporal explanations. The PhysioNet/CinC 2019 Challenge dataset, comprising 1,552,210 patient hours from 40,336 ICU patients, was utilized. A temporal transformer encoder (TTE) was trained using 12-h look-back windows with 92 engineered features. Severe class imbalance (2.6% positive rate) was addressed through weighted random sampling and focal loss. Fivefold patient-level cross-validation was employed to prevent temporal leakage. Platt scaling was applied for probability calibration. Grad-CAM was adapted for temporal explainability, while SHAP was used for feature-level attribution. BiLSTM-Attention and XGBoost models served as baseline comparators. The TTE model achieved a cross-validated AUROC of $0.8320\pm 0.0032$ and an AUPRC of $0.1505\pm 0.0148$, significantly outperforming BiLSTM-Attention (AUROC: 0.7859) and XGBoost (AUROC: 0.7731; DeLong $p<0.0001$). Platt scaling reduced the expected calibration error from 0.3154 to 0.0017. The median alert lead time was 46.5 h (IQR: 21–84 h), with 95.3% of septic patients receiving alerts at least 3 h before onset. Grad-CAM analysis identified time steps $t-10$ and $t-9$ as the most predictive. However, high-severity patients (SOFA proxy $\ge 3$) demonstrated substantially reduced performance (AUROC: 0.257). The proposed TTE framework demonstrated strong and well-calibrated early sepsis prediction with substantial clinical lead time. The concentration of predictive signals 10–11 h prior to alert generation supports the feasibility of continuous automated ICU monitoring from admission onward. Reduced performance in high-severity patients highlights the need for severity-stratified modelling in future research. Full article

Parkinson’s disease (PD) is characterized by progressive dopaminergic neurodegeneration driven by oxidative stress, mitochondrial dysfunction, synaptic loss, and impaired neurotrophic signaling; however, the role of integrin α5β1 in neuronal vulnerability remains unclear. Here, the data show that rotenone-induced stress reduces integrin α5 expression in a dose- and time-dependent manner, leading to increased ROS accumulation, glutathione imbalance, synaptic degeneration, senescence-like β-gal activity, and apoptosis, whereas integrin α5 knockdown further exacerbates these deficits, supporting a protective role of α5β1. In contrast, treatment with the fibronectin-derived α5β1-activating peptide Ac-PHSRN-NH₂ restores integrin signaling by engaging the FAK–PI3K–AKT/ERK cascade and NRF2-mediated antioxidant responses, thereby reducing oxidative stress, suppressing cell death, and improving redox homeostasis. Moreover, PHSRN enhances NGF and BDNF levels, preserves synaptic integrity, and promotes dopaminergic neuronal activity and dopamine release. Consistently, in MPTP-lesioned mice, PHSRN preserves nigral TH-positive neurons, reduces apoptosis, restores neurotrophic support, and improves motor function. Collectively, these findings identify integrin α5β1 as a critical protective axis and support PHSRN as a potential disease-modifying therapeutic strategy for PD. Full article

Resistance to 5-fluorouracil (5-FU) remains a major limitation in colorectal cancer therapy, prompting the development of combination strategies aimed at improving its efficacy. Bakuchiol (BAK), a natural compound with reported antioxidant and pro-oxidant properties, may modulate redox balance and enhance chemotherapy response. This study compared the effects of 5-FU and BAK, applied as monotherapies and in combination, in DLD-1 and HT-29 colorectal cancer cells. Cytotoxicity assays showed that co-treatment significantly reduced the IC₅₀ of 5-FU, particularly in DLD-1 cells, and revealed an enhanced anticancer effect of the combination treatment compared with either monotherapy. Flow cytometric analyses demonstrated enhanced apoptosis via extrinsic and intrinsic pathways, including increased caspase 8 activity, loss of mitochondrial membrane potential (ΔΨ_m), activation of caspase 9, and subsequent activation of caspases 3/7. These effects were associated with a pronounced redox imbalance, reflected by increased intracellular reactive oxygen species (ROS) levels, suggesting a central role of oxidative stress in mediating cytotoxicity. Antioxidant pre-treatment attenuated ROS accumulation and reduced apoptosis, confirming a causal relationship. Additionally, autophagy was induced selectively in DLD-1 cells, indicating cell-line-specific differences in redox adaptation. Taken together, BAK enhances 5-FU efficacy through ROS-dependent activation of mitochondrial and caspase-dependent pathways, with stronger effects observed in DLD-1 cells. Full article