Search Results (1,499)

Improving both the high- and low-temperature performance of asphalt is still difficult in modern pavement applications. This performance imbalance has motivated the development of new modification strategies that can enhance temperature stability while maintaining construction workability. In this research, a low-molecular-weight elastic polyolefin (POL) with inherent compatibility was introduced as a novel asphalt modifier. POL was incorporated at five dosages (0%, 2%, 4%, 6%, and 8% by weight of asphalt) to investigate its effects on the fundamental physical, rheological, and low-temperature properties of the asphalt. The rheological behavior was characterized by dynamic shear rheometer (DSR) and bending beam rheometer (BBR), while the modification mechanism and dispersion morphology were analyzed through Fourier-transform infrared spectroscopy (FT-IR) and fluorescence microscopy (FM). The results reveal that POL markedly improves the high-temperature performance and workability of asphalt, with the rutting factor increasing by two- to eightfold. POL modification improved the thermal stability of asphalt, shifting the maximum decomposition temperature from 455.2 °C for the base binder to 461–463 °C, while the total mass loss remained nearly constant at 80–83%. Microscopic observations confirm that POL forms a physically blended network within the asphalt matrix, exhibiting a green fluorescent structure that becomes progressively continuous with increasing dosage. The most homogeneous dispersion and optimal compatibility occur at a POL dosage of 6%, beyond which phase segregation emerges and low-temperature properties deteriorate. Accordingly, a 6% POL dosage is recommended for achieving balanced performance. These findings provide theoretical and practical guidance for the development of balanced performance and thermally stable POL-modified asphalt materials. Full article

In recent years, the expansion of network scale and the diversification of attack methods pose dual challenges to intrusion detection systems in extracting effective features and addressing class imbalance. To address these issues, the Spatial–Temporal Equilibrium Graph Convolutional Network (STEGCN) is proposed. This model integrates Graph Convolutional Network (GCN) and Gated Recurrent Unit (GRU), leveraging GCN to extract high-order spatial features from network traffic data while capturing complex topological relationships and latent patterns. Meanwhile, GRU efficiently models the dynamic evolution of network traffic over time, accurately depicting temporal trends and anomaly patterns. The synergy of these two components provides a comprehensive representation of network behavior. To mitigate class imbalance in intrusion detection, the Equalization Loss v2 (EQLv2) is introduced. By dynamically adjusting gradient contributions, this function reduces the dominance of majority classes, thereby enhancing the model’s sensitivity to minority-class attacks. Experimental results demonstrate that STEGCN achieves superior detection performance on the UNSW-NB15 and CICIDS2017 datasets. Compared with traditional deep learning models, STEGCN shows significant improvements in accuracy and recall, particularly in detecting minority-class intrusions. Full article

Background and Objectives: The optimal pharmacological strategy to mitigate progressive vertebral collapse during the acute phase of osteoporotic vertebral compression fractures (OVCFs) remains a subject of debate. This initial 3-month window is the most critical period for evaluating the structural stability of the fracture, as the majority of progressive height loss occurs before solid bone union is achieved, directly influencing the decision to continue conservative management or transition to surgical intervention. Materials and Methods: In this retrospective study, 123 patients were allocated to control (n = 26), denosumab (n = 35), teriparatide (n = 30), or romosozumab (n = 32) groups. Treatment choice was non-randomized, driven by clinical pragmatism and patient preference. Serial changes in vertebral compression rate (VCR) and pain (VAS) were analyzed over 3 months using linear mixed models (LMMs) specifically adjusted for baseline imbalances in initial VCR. Results: In the unadjusted analysis, DMAB appeared to show a slower progression of compression compared to the control group. However, after adjusting for the initial VCR, no significant structural benefit was observed in any medication group (p > 0.05), with all groups showing small effect sizes (Cohen’s d < 0.4). In contrast, unstable fracture morphology was identified as the most potent driver of vertebral collapse (β = 2.758, 95% CI: 1.51–4.01, p < 0.001). Clinically, the RM group showed significantly lower overall pain levels throughout the follow-up period compared to the control group (p = 0.014). Conclusions: Short-term osteoporosis medication does not significantly mitigate vertebral collapse during the acute phase of OVCFs. Practically, these findings suggest that unstable fracture morphology and the baseline VCR—reflecting a potential ‘floor effect’ where less initially collapsed vertebrae may undergo more significant progression—are more informative predictors of acute collapse than medication choice. Consequently, early imaging-based risk stratification is crucial to identify patients at high risk for progressive deformity, regardless of their pharmacological regimen. Full article

Parkinson’s disease (PD) is a heterogeneous neurodegenerative disorder in which oxidative stress represents a final common pathway linking diverse genetic and environmental insults to dopaminergic neuron loss. This review synthesizes evidence on how the commonly observed pathological changes in PD converge on excessive reactive oxygen species generation and redox imbalance. We present an overview on these pathways and key PD-linked genes that perturb mitochondrial quality control, lysosomal function, and inflammatory signaling, reinforcing oxidative stress. The major classes of redox-targeted therapeutic strategies under preclinical and clinical evaluation are outlined. Although many candidates show robust target engagement and neuroprotection in models, clinical trials have frequently yielded neutral or modest results, highlighting challenges related to brain delivery, off-target effects, optimal treatment window, and the fact that oxidative stress alone may be necessary but not sufficient to drive human disease progression. In the current paper, beyond cataloguing oxidative pathways, we explain the role of etiologic heterogeneity on biochemical target engagement and clinical outcomes. We outline subtype-enriched trial strategies and rational combination approaches. Targeting oxidative stress–related pathways thus remains a promising avenue for disease modification in PD, provided that future interventions are mechanistically informed and adapted to patient-specific redox vulnerabilities. Full article

Patients with hepatocellular carcinoma (HCC) often sit at the crossroads of malignancy and chronic liver disease, where impaired hepatic reserve, systemic inflammation, and treatment-related stress accelerate loss of skeletal muscle mass and function. In this narrative review, we synthesize current evidence on the two-way relationship between sarcopenia and HCC management across curative and palliative settings. We outline key biological pathways—altered energy substrate use, amino acid imbalance, hyperammonemia-related signaling, and inflammatory and hormonal perturbations—that promote progressive muscle wasting, and we summarize how sarcopenia influences tolerance, complications, and outcomes of surgery, locoregional therapies, and systemic agents. We then translate the literature into practical supportive-care principles, including adequate energy and protein delivery, optimized meal distribution (including late-evening snacks), and selected supplementation alongside hepatic rehabilitation/exercise. Potential adjuncts discussed include branched-chain amino acids, L-carnitine, vitamin D, zinc, and other micronutrients. Because the available data are heterogeneous and largely derived from observational cohorts or extrapolated from cirrhosis populations, HCC-specific randomized trials and standardized intervention protocols remain limited. Therefore, nutritional and exercise recommendations should be individualized according to tumor stage, hepatic function, comorbidities, and treatment goals, and viewed as supportive guidance that requires confirmation in well-designed prospective studies. Full article

Unplanned downtime in industrial machinery remains a major challenge, causing substantial economic losses and safety risks across sectors such as manufacturing, food processing, oil and gas, and transportation. This systematic review investigates the application of machine learning (ML) techniques for anomaly and fault detection within the broader context of predictive maintenance. Following a hybrid review methodology, relevant studies published between 2010 and 2025 were collected from major databases including IEEE Xplore, ScienceDirect, SpringerLink, Scopus, Web of Science, and arXiv. The review categorizes approaches into supervised, unsupervised, and hybrid paradigms, analyzing their pipelines from data collection and preprocessing to model deployment. Findings highlight the effectiveness of deep learning architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), autoencoders, and hybrid frameworks in detecting faults from time series and multimodal sensor data. At the same time, key limitations persist, including data scarcity, class imbalance, limited generalizability across equipment types, and a lack of interpretability in deep models. This review concludes that while ML-based predictive maintenance systems are enabling a transition from reactive to proactive strategies, future progress requires improved hybrid architectures, Explainable AI, and scalable real-time deployment. Full article

Network intrusion detection has challenges that fundamentally differ from language and vision tasks typically addressed by Transformer models. In particular, network traffic features lack inherent ordering, datasets are extremely class-imbalanced (with benign traffic often exceeding 80%), and reported accuracies in the literature vary widely (57–95%) without systematic explanation. To address these challenges, we propose a controlled experimental study that isolates and quantifies the impact of tokenization strategies on Transformer-based intrusion detection systems. Specifically, we introduce and compare three tokenization approaches—feature-wise tokenization (78 tokens) based on CICIDS2017, a sample-wise single-token baseline, and an optimized sample-wise tokenization—under identical training and evaluation protocols on a highly imbalanced intrusion detection dataset. We demonstrate that tokenization choice alone accounts for an accuracy gap of 37.43 percentage points, improving performance from 57.09% to 94.52% (100 K data). Furthermore, we show that architectural mechanisms for handling class imbalance—namely Batch Normalization and capped loss weights—yield an additional 15.05% improvement, making them approximately 21× more effective than increasing the training data by 50%. We achieve a macro-average AUC of 0.98, improve minority-class recall by 7–12%, and maintain strong discrimination even for classes with as few as four samples (AUC 0.9811). These results highlight tokenization and imbalance-aware architectural design as primary drivers of performance in Transformer-based intrusion detection and contribute practical guidance for deploying such models in modern network infrastructures, including IoT and cloud environments where extreme class imbalance is inherent. This study also presents practical implementation scheme recommending sample-wise tokenization, constrained class weighting, and Batch Normalization after embedding and classification layers to improve stability and performance in highly unstable table-based IDS problems. Full article

We design a photonic terahertz (THz) antenna operating in the 275–296 GHz band, integrating two identical modified uni-traveling-carrier photodiodes (MUTC-PDs), two impedance matching networks, a Wilkinson power combiner and a Vivaldi antenna. Simulation results show a saturated output power of 1.58 dBm at 280 GHz, achieving a 4.13× power enhancement through optimized impedance and power combining compared to a standalone device. The integrated antenna achieves a peak gain of 7.93 dBi, with reflection coefficients below −10 dB. The system demonstrates a power combining efficiency exceeding 95% for phase imbalances up to 20°, and the Wilkinson combiner exhibits only 0.76 dB insertion loss at 285 GHz, demonstrating high radiation and combining efficiency in the 275–296 GHz band. Full article

The detection of exposed subsea pipelines is a key task in current marine remote sensing, and multibeam echosounders (MBESs) are a primary instrument for detecting exposed pipelines. However, complex seabed environments interfere with acoustic echoes, introducing substantial noise points into MBES point-cloud data and substantially degrading its quality. Conventional point-cloud denoising methods struggle to suppress noise while simultaneously preserving pipeline integrity, whereas point-cloud noise-segmentation methods can better address this challenge. Nevertheless, noise-segmentation methods remain constrained by the lack of geometric priors and the presence of class imbalance. To address these issues, this paper proposes a geometry-prior and rational-activation Transformer for the MBES point-cloud denoising of exposed subsea pipelines (GPRAformer). The method comprises the following three core designs: a pipeline-informed prior encoder (PIPE) sampling module to enhance the separability between pipeline points and noise points; a rational-activated Kolmogorov–Arnold network transformer (RaKANsformer) feature extraction module that couples gated self-attention with KAN structures using rational-function activations for joint feature extraction, thereby strengthening global dependency modeling and nonlinear expressivity; and class-adaptive loss (CAL)-constrained noise-segmentation module that introduces intra-class consistency and inter-class separation constraints to mitigate false detections and miss detections arising from class imbalance. Evaluations on actual measured MBES point-cloud datasets show that, compared with the suboptimal model under each metric, GPRAformer achieves improvements of 6.83%, 1.78%, 5.12%, and 6.20% in mean intersection over union (mIoU), Accuracy, F1-score, and Recall, respectively. These results indicate a significant enhancement in overall segmentation performance. Therefore, GPRAformer can achieve high-precision and robust MBES point-cloud noise segmentation in complex seabed environments. Full article

Manganese (Mn) toxicity, commonly triggered by soil acidification, poses a significant threat to citrus production. Arbuscular mycorrhizal (AM) fungi can alleviate heavy metal stress, while their specific function and quantitative effectiveness in conferring Mn tolerance to citrus remain unclear. This study investigated the physiological regulation conferred by four AM fungal species, Rhizophagus intraradices (Ri), Funneliformis mosseae (Fm), Paraglomus occultum (Po), and Diversispora epigaea (De), on trifoliate orange (Poncirus trifoliata L. Raf.) under Mn stress. Mn toxicity reduced root colonization in a species-dependent manner, significantly lowering colonization by all AM fungal isolates except Fm. It also severely inhibited plant growth and induced pronounced oxidative damage, accompanied by metabolic imbalance. Under Mn-stressed conditions, AM fungal inoculation, especially Ri, significantly enhanced plant biomass relative to the non-AM control, with respective increases of 148% in leaves, 33% in stems, and 64% in roots, demonstrating a marked species-specific efficacy. Furthermore, AM symbiosis effectively promoted chlorophyll index and limited Mn translocation to the leaves under both non-stress and Mn-stress conditions, with Ri being the most effective in reducing leaf Mn content. Symbiosis with AM fungi, particularly Ri, fine-tuned the antioxidant enzyme defense under Mn stress by selectively suppressing superoxide dismutase and peroxidase activities while further boosting catalase activity. Concurrently, AM fungi alleviated Mn-induced oxidative damage, with the magnitude of mitigation varying by species: Ri delivered the most comprehensive protection, most effectively reducing hydrogen peroxide and malondialdehyde levels in both leaves and roots, whereas Po was particularly effective in suppressing root superoxide anion radical and malondialdehyde levels in roots. Furthermore, AM fungi reversed Mn-induced shifts in organic osmolytes: they significantly reduced the excessive accumulation of soluble sugars and proline while mitigating the loss of soluble proteins, thereby assisting in restoring metabolic homeostasis. The alleviative effects varied significantly among AM fungal species, with Ri identified as the most efficient and Mn-tolerant strain. These findings highlight the potential of utilizing specific AM fungi, particularly Ri, as a sustainable biological strategy to enhance citrus productivity in acidified, Mn-contaminated soils. Full article

Lung cancer remains a leading cause of cancer-related mortality. Although reliable multiclass classification of lung lesions from CT imaging is essential for early diagnosis, it remains challenging due to subtle inter-class differences, limited sample sizes, and class imbalance. We propose an Adaptive Attention-Augmented Convolutional Neural Network with Vision Transformer (AACNN-ViT), a hybrid framework that integrates local convolutional representations with global transformer embeddings through an adaptive attention-based fusion module. The CNN branch captures fine-grained spatial patterns, the ViT branch encodes long-range contextual dependencies, and the adaptive fusion mechanism learns to weight cross-representation interactions to improve discriminability. To reduce the impact of imbalance, a hybrid objective that combines focal loss with categorical cross-entropy is incorporated during training. Experiments on the IQ-OTH/NCCD dataset (benign, malignant, and normal) show consistent performance progression in an ablation-style evaluation: CNN-only, ViT-only, CNN-ViT concatenation, and AACNN-ViT. The proposed AACNN-ViT achieved 96.97% accuracy on the validation set with macro-averaged precision/recall/F1 of 0.9588/0.9352/0.9458 and weighted F1 of 0.9693, substantially improving minority-class recognition (Benign recall 0.8333) compared with CNN-ViT (accuracy 89.09%, macro-F1 0.7680). One-vs.-rest ROC analysis further indicates strong separability across all classes (micro-average AUC 0.992). These results suggest that adaptive attention-based fusion offers a robust and clinically relevant approach for computer-aided lung cancer screening and decision support. Full article

Accurate and timely agricultural mapping is essential for supporting sustainable agricultural development, resource management, and food security. Despite its importance, Vietnam lacks detailed and consistent large-scale agricultural maps. In this study, we produced the first national-scale agricultural map of Vietnam for 2024 using a UNet++ deep learning architecture that integrates multi-temporal Sentinel-1 and Sentinel-2 imagery with Global-30 DEM data. The resulting product includes 15 land-cover categories, eight of which represent the most popular agricultural types in Vietnam. We further evaluate the model’s transferability by applying the 2024 trained model to generate a corresponding map for 2020. The approach achieves overall classification accuracies of $83.01\%\pm 1.37\%$ (2020) and $80.09\%\pm 0.76\%$ (2024). To address class imbalance within the training dataset, we introduced an adaptive weight combined loss function that automatically adjusts the weight of dice loss and cross-entropy loss within a combined loss function during the model training process. Full article

The maintenance of bone homeostasis requires the coordinated activity of specialized cells (osteoblasts, osteoclasts and osteocytes), soluble factors and hormones with regulatory functions. Disruption of this tightly controlled balance contributes to several skeletal pathological conditions, among which osteoporosis is one of the most prevalent. Iron, an essential element for the basic cellular functions of both osteoblasts and osteoclasts, plays a pivotal role in preserving bone homeostasis and skeletal integrity. Both iron deficiency and iron overload impair bone remodeling through distinct but converging mechanisms. Iron deficiency compromises collagen synthesis, alters hypoxia-dependent signaling, and may affect vitamin D metabolism, collectively predisposing the individual to reduced bone mineral density and increased fracture risk. Conversely, excess iron enhances oxidative stress, promotes osteoclastogenesis, and suppresses osteoblast differentiation and function, thereby favoring bone loss, particularly in the aging population and postmenopausal individuals. Hepcidin, the master regulator of systemic iron availability, has emerged as a key modulator of bone turnover, whereas the bone-derived hormone fibroblast growth factor 23 (FGF23) links iron imbalance to phosphate homeostasis, vitamin D metabolism, and inflammation. Beyond metabolic bone diseases, dysregulated iron handling is increasingly recognized as a hallmark of osteosarcoma biology, influencing tumor growth, metabolic reprogramming, and an individual’s susceptibility to ferroptosis. The emerging, albeit only preclinical, evidence of the roles of iron and ferroptosis in osteosarcoma is therefore also covered. This review summarizes the current understanding of the interactions between iron metabolism and bone biology and addresses how an imbalance in iron metabolism may lead to major skeletal disorders. Overall, iron homeostasis could represent a potential target for preventing and treating osteoporosis and for improving therapeutic strategies for osteosarcoma. Full article

Climate change reshapes crop–weed interactions and challenges the cultivation of oilseed rape (Brassica napus L.). Common ragweed (Ambrosia artemisiifolia L.) strongly suppresses early crop development, increases stress sensitivity and leads to yield loss. The stress–physiological responses of oilseed rape to ragweed competition were investigated using a combination of conventional and non-invasive methods. A pot experiment was conducted with increasing ragweed densities (0, 1, 3, 5 and 10 plants). Plant height and biomass were evaluated via non-destructive indicators (SPAD, NDVI) and different stages (1–15 and 16–30 min) of delayed fluorescence (DF) alongside ferric reducing antioxidant power (FRAP). Increasing ragweed density caused changes in growth, altered DF magnitude and decay kinetics, indicating photosynthetic imbalance. Moderate weed competition (1–5) induced an adaptive, eustress-like response characterised by enhanced non-enzymatic antioxidant capacity, whereas higher ragweed densities overwhelmed this compensatory mechanism, resulting in oxidative stress-like responses. Among all measured traits, DF₁–₁₅ proved to be the earliest and most sensitive indicator of the transition from adaptive to disruptive stress: T1: 0 ragweed: 213.07 ± 10.36 cps/mm² and 92.66 ± 6.67 cps/mm². These results demonstrate that delayed fluorescence, combined with conventional physiological and antioxidant-based parameters, enables the early detection of competitive stress in oilseed rape well before visible symptoms appear. Full article

Tomato leaf diseases cause substantial yield and quality losses worldwide, yet reliable detection in real fields remains challenging. Two practical bottlenecks dominate current research: (i) limited data, including small samples for rare diseases, class imbalance, and noisy field images, and (ii) multimodal heterogeneity, where RGB images, textual symptom descriptions, spectral cues, and optional molecular assays provide complementary but hard-to-align evidence. This review summarizes recent advances in tomato leaf disease detection under these constraints. We first formalize the problem settings of limited and multimodal data and analyze their impacts on model generalization. We then survey representative solutions for limited data (transfer learning, data augmentation, few-/zero-shot learning, self-supervised learning, and knowledge distillation) and multimodal fusion (feature-, decision-, and hybrid-level strategies, with attention-based alignment). Typical model–dataset pairs are compared, with emphasis on cross-domain robustness and deployment cost. Finally, we outline open challenges—including weak generalization in complex field environments, limited interpretability of multimodal models, and the absence of unified multimodal benchmarks—and discuss future opportunities toward lightweight, edge-ready, and scalable multimodal systems for precision agriculture. Full article