Search Results (10,657)

As global agriculture shifts to intelligence and precision, crop attribute detection has become foundational for intelligent systems (harvesters, UAVs, sorters). It enables real-time monitoring of key indicators (maturity, moisture, disease) to optimize operations—reducing crop losses by 10–15% via precise cutting height adjustment—and boosts resource-use efficiency. This review targets harvesting-stage and in-field monitoring for grains, fruits, and vegetables, highlighting practical technologies: near-infrared/Raman spectroscopy (non-destructive internal attribute detection), 3D vision/LiDAR (high-precision plant height/density/fruit location measurement), and deep learning (YOLO for counting, U-Net for disease segmentation). It addresses universal field challenges (lighting variation, target occlusion, real-time demands) and actionable fixes (illumination compensation, sensor fusion, lightweight AI) to enhance stability across scenarios. Future trends prioritize real-world deployment: multi-sensor fusion (e.g., RGB + thermal imaging) for comprehensive perception, edge computing (inference delay < 100 ms) to solve rural network latency, and low-cost solutions (mobile/embedded device compatibility) to lower smallholder barriers—directly supporting scalable precision agriculture and global sustainable food production. Full article

Spiking neural networks (SNNs) provide a biologically inspired, event-driven alternative to artificial neural networks (ANNs), potentially delivering competitive accuracy at substantially lower energy. This tutorial-study offers a unified, practice-oriented assessment that combines critical review and standardized experiments. We benchmark a shallow fully connected network (FCN) on MNIST and a deeper VGG7 architecture on CIFAR-10 across multiple neuron models (leaky integrate-and-fire (LIF), sigma–delta, etc.) and input encodings (direct, rate, temporal, etc.), using supervised surrogate-gradient training implemented in Intel Lava, SLAYER, SpikingJelly, Norse, and PyTorch. Empirically, we observe a consistent but tunable trade-off between accuracy and energy. On MNIST, sigma–delta neurons with rate or sigma–delta encodings achieve 98.1% accuracy (ANN baseline: 98.23%). On CIFAR-10, sigma–delta neurons with direct input reach 83.0% accuracy at just two time steps (ANN baseline: 83.6%). A GPU-based operation-count energy proxy indicates that many SNN configurations operate below the ANN energy baseline; some frugal codes minimize energy at the cost of accuracy, whereas accuracy-oriented settings (e.g., sigma–delta with direct or rate coding) narrow the performance gap while remaining energy-conscious—yielding up to threefold efficiency compared with matched ANNs in our setup. Thresholds and the number of time steps are decisive factors: intermediate thresholds and the minimal time window that still meets accuracy targets typically maximize efficiency per joule. We distill actionable design rules—choose the neuron–encoding pair according to the application goal (accuracy-critical vs. energy-constrained) and co-tune thresholds and time steps. Finally, we outline how event-driven neuromorphic hardware can amplify these savings through sparse, local, asynchronous computation, providing a practical playbook for embedded, real-time, and sustainable AI deployments. Full article

Human immunodeficiency virus (HIV) remains a global public health challenge. Antiretroviral therapy (ART) improves outcomes by suppressing viral replication and enabling immune recovery, yet the early molecular mechanisms of immune-related transcriptional change after ART remain insufficiently characterized. We enrolled eight ART-naïve male patients with HIV aged 18–35. Peripheral blood mononuclear cells (PBMCs) were collected before and after 24 weeks of combination ART (TDF, 3TC, DTG) and underwent bulk RNA-seq (Illumina HiSeq 1500, Illumina, Inc., San Diego, CA, USA). Differential expression was assessed with DESeq2 (paired design); gene set enrichment analysis (GSEA), principal component analysis (PCA), hierarchical clustering, and protein–protein interaction (PPI) networks (STRING/NetworkX) explored functional patterns and transcriptomic shifts. We identified 87 differentially expressed genes, including 67 downregulated interferon-stimulated genes (e.g., IFI44L, ISG15, STAT1) and 20 upregulated transcripts, mostly pseudogenes related to ribosomal proteins. Functional enrichment revealed suppression of type I interferon and other antiviral signaling pathways. PCA and hierarchical clustering indicated a post-ART transcriptional shift. These findings suggest that early immune recovery following ART involves downregulation of chronic interferon-driven activation. This observation may correspond to partial restoration of T-cell functional capacity, reduced immune exhaustion, and a rebalanced antiviral immune environment. Full article

Dynamic Spectrum Sharing (DSS) enables spectrum sharing between Long-Term Evolution (LTE) and New Radio (NR) systems, addressing spectrum scarcity in NR. To avoid interference when supporting NR traffic within LTE spectrum, key factors must be compatible. Effective DSS techniques are essential for coexistence. This paper discusses these issues in two parts. Part I covers LTE and NR coexistence using DSS, introducing resource grids, control signals, and channels, and explores DSS approaches for NR data traffic, including NR Synchronization Signal/Physical Broadcast Channels (SSB) transmission via LTE Multicast-Broadcast Single-Frequency Network (MBSFN) and non-MBSFN subframes with associated challenges and standardization efforts for DSS improvement. Part II presents a DSS technique using MBSFN subframes in a heterogeneous network with a macrocell and picocells running on LTE, and in-building small cells running on NR, sharing LTE spectrum via DSS. An optimization problem is formulated to manage traffic through MBSFN allocation, determining the optimal number of MBSFN subframes per LTE frame. System simulations indicate DSS improves Spectral and Energy Efficiency in small cells. The paper concludes with key lessons for LTE and NR coexistence. Full article

Background: Tissue-specific long non-coding RNAs (lncRNAs) represent potential biomarkers. The testis-enriched lncRNA41584, previously identified as downregulated in male-sterile silkworm mutants (JMS, GMS), is associated with male sterility, but its functional mechanism remained unknown. Subcellular localization, dual-luciferase reporter assays, MTT, and flow cytometry were employed to examine lncRNA41584–miR-3047-z–BmCDK20 interactions. In vivo functional validation included lncRNA41584 knockdown and miR-3047-z overexpression in Bombyx mori. lncRNA41584 localizes predominantly to the cytoplasm and acts as a competing endogenous RNA (ceRNA) by sponging miR-3047-z, thereby upregulating the cyclin-dependent kinase BmCDK20. Perturbation of this axis impaired BmN cell proliferation, causing G1 phase arrest, and led to spermatocyst malformation, reduced fertilization rates, and increased unfertilized eggs. The lncRNA41584–miR-3047-z–BmCDK20 ceRNA network is essential for testicular cell cycle progression and spermatogenesis in silkworms, offering mechanistic insights into lepidopteran male sterility and potential targets for pest fertility regulation. Full article

The photocatalytic degradation of Crystal Violet (CV) using ZnO-based nanomaterials presents a promising solution for addressing water pollution caused by synthetic dyes. This review highlights the exceptional efficiency of ZnO and its modified forms—such as doped, composite, and heterostructured variants—in degrading CV under both ultraviolet (UV) and solar irradiation. Key advancements include strategic bandgap engineering through doping (e.g., Cd, Mn, Co), innovative heterojunction designs (e.g., n-ZnO/p-Cu₂O, g-C₃N₄/ZnO), and composite formations with graphene oxide, which collectively enhance visible-light absorption and minimize charge recombination. The degradation mechanism, primarily driven by hydroxyl and superoxide radicals, leads to the complete mineralization of CV into non-toxic byproducts. Furthermore, this review emphasizes the emerging role of Artificial Neural Networks (ANNs) as superior tools for optimizing degradation parameters, demonstrating higher predictive accuracy and scalability compared to traditional methods like Response Surface Methodology (RSM). Potential operational challenges and future directions—including machine learning-driven optimization, real-effluent testing potential, and the development of solar-active catalysts—are further discussed. This work not only consolidates recent breakthroughs in ZnO-based photocatalysis but also provides a forward-looking perspective on sustainable wastewater treatment strategies. Full article

The scalability of coordination policies is a critical challenge in swarm robotics, where agent numbers may vary substantially between deployment scenarios. Reinforcement learning (RL) offers a promising avenue for learning decentralized policies from local interactions, yet a fundamental question remains: can policies trained on one swarm size transfer to different population scales without retraining? This zero-shot transfer problem is particularly challenging because the traditional RL approaches learn fixed-dimensional representations tied to specific agent counts, making them brittle to population changes at deployment time. While existing work addresses scalability through population-aware training (e.g., mean-field methods) or multi-size curricula (e.g., population transfer learning), these approaches either impose restrictive assumptions or require explicit exposure to varied team sizes during training. Graph Neural Networks (GNNs) offer a fundamentally different path. Their permutation invariance and ability to process variable-sized graphs suggest potential for zero-shot generalization across swarm sizes, where policies trained on a single population scale could deploy directly to larger or smaller teams. However, this capability remains largely unexplored in the context of swarm coordination. For this reason, we empirically investigate this question by combining GNNs with deep Q-learning in cooperative swarms. We focused on well-established 2D navigation tasks that are commonly used in the swarm robotics literature to study coordination and scalability, providing a controlled yet meaningful setting for our analysis. To address this, we introduce Deep Graph Q-Learning (DGQL), which embeds agent-neighbor graphs into Q-learning and trains on fixed-size swarms. Across two benchmarks (goal reaching and obstacle avoidance), we deploy up to three times larger teams. The DGQL preserves a functional coordination without retraining, but efficiency degrades with size. The ultimate goal distance grows monotonically (15–29 agents) and worsens beyond roughly twice the training size ($\approx 20$ agents), with task-dependent trade-offs. Our results quantify scalability limits of GNN-enhanced DQL and suggest architectural and training strategies to better sustain performance across scales. Full article

With the rapid growth of wireless devices, security has become a key research concern in beyond-5G (B5G) and sixth-generation (6G) networks. Non-orthogonal multiple access (NOMA), one of the supporting technologies, is a strong contender to enable massive connectivity, increase spectrum efficiency, and guarantee high-quality access for a sizable user base. Furthermore, the scientific community has recently paid close attention to the effects of hardware impairments (HIs). The safe transmission of NOMA in a two-user uplink relay network is examined in this paper, taking into account both hardware limitations and the existence of listening devices. Each time frame in a mobile network environment comprises two phases in which users use a relay (R) to interact with the base station ($BS$). The research focuses on scenarios where a malicious device attempts to intercept the uplink signals transmitted by users through the R. Using important performance and security metrics, such as connection outage probability (COP), secrecy outage probability (SOP), and intercept probability (IP), system behavior is evaluated. To assess the system’s security and reliability under the proposed framework, closed-form analytical expressions are derived for SOP, IP, and COP. The simulation results provide the following insights: (i) they validate the accuracy of the derived analytical expressions; (ii) the study significantly deepens the understanding of secure NOMA uplink transmission under the influence of HIs across all the network entities, paving the way for future practical implementations; and (iii) the results highlight the superior performance of secure and reliable NOMA uplink systems compared to benchmark orthogonal multiple access (OMA) counterparts when both operate under the same HI conditions. Furthermore, an extended model without a relay is considered for comparison with the proposed relay-assisted scheme. Moreover, the numerical results indicate that the proposed communication model achieves over 90% reliability (with a COP below 0.1) and provides approximately a 30% improvement in SOP compared to conventional OMA-based systems under the same HI conditions. Full article

Background: Cardiorenal syndrome (CRS) reflects bidirectional heart–kidney injury whose mechanisms extend far beyond hemodynamics. High-throughput genomics and multi-omics now illuminate the molecular circuits that couple cardiac and renal dysfunction. Methods: We narratively synthesize animal and human studies leveraging transcriptomics, proteomics, peptidomics, metabolomics, and non-coding RNA profiling to map convergent pathways in CRS and to highlight biomarker and therapeutic implications. Results: Across acute and chronic CRS models, omics consistently converge on extracellular matrix (ECM) remodeling and fibrosis (e.g., FN1, POSTN, collagens), immune–inflammatory activation (IL-6 axis, macrophage/complement signatures), renin–angiotensin–aldosterone system hyperactivity, oxidative stress, and metabolic/mitochondrial derangements in both organs. Single-nucleus and bulk transcriptomes reveal tubular dedifferentiation after cardiac arrest-induced AKI and myocardial reprogramming with early CKD, while quantitative renal proteomics in heart failure demonstrates marked upregulation of ACE/Ang II and pro-fibrotic matricellular proteins despite near-normal filtration. Human translational data corroborate these signals: urinary peptidomics detects CRS-specific collagen fragments and protease activity, and circulating FN1/POSTN and selected microRNAs (notably miR-21) show diagnostic potential. Epigenetic and microRNA networks appear to integrate these axes, nominating targets such as anti-miR-21 and anti-fibrotic strategies; pathway-directed repurposing exemplifies dual-organ benefit. Conclusions: Genomics and multi-omics recast CRS as a systems disease driven by intertwined fibrosis, inflammation, neurohormonal and metabolic programs. We propose a translational framework that advances (i) composite biomarker panels combining injury, fibrosis, and regulatory RNAs; (ii) precision, pathway-guided therapies; and (iii) integrated, longitudinal multi-omics of well-phenotyped CRS cohorts to enable prediction and personalized intervention. Full article

Przewalski’s gazelle (Procapra przewalskii) is an endangered ungulate endemic to the Qinghai–Tibet Plateau, with a small population size and exposure to multiple ecological pressures. Its gut microbiota may play a crucial role in host environmental adaptation. To investigate the functional divergence of gut microbial communities, we performed high-throughput metagenomic sequencing on 105 wild fecal samples collected from 10 geographic regions around Qinghai Lake. The results revealed significant regional differentiation in key functional modules related to metabolism, antibiotic resistance mechanisms, and virulence-associated pathways. All populations showed enrichment in core metabolic pathways such as carbohydrate and amino acid metabolism, with carbohydrate-active enzymes dominated by glycoside hydrolases (GHs) and glycosyltransferases (GTs), exhibiting overall functional conservation. Although populations shared many antibiotic- and virulence-related reference genetic markers, the marker composition associated with distinct resistance mechanisms and pathogenic processes exhibited clear population-specific patterns, suggesting differential microbial responses to local environmental pressures. Correlation network analysis further identified core taxa (e.g., Arthrobacter and Oscillospiraceae/Bacteroidales lineages) as key genera linking community structure with core metabolic, resistance-related, and virulence-associated marker functions. Overall, the gut microbiota of Przewalski’s gazelle exhibits a complex spatially structured functional differentiation, reflecting host–microbiome co-adaptation under region-specific ecological pressures. These findings provide critical methodological and theoretical support for microecological health assessment and regionally informed conservation management of this endangered species. Full article

Traditional composite indicators for the study of sustainability often obscure the complex network of relationships among individual indicators, functioning as black boxes that fail to diagnose the underlying structural and functional weaknesses of the system. The objective of this research is to develop and apply a complementary approach grounded in network theory to diagnose and evaluate the structural and functional cohesion of environmental indicator systems. We developed a study that combines the Principal Component Analysis (PCA) method with network theory to comprehensively analyze the indicator system. The core of this contribution is the development of the $Mo\left(G\right)$ index, designed to quantify the structural–functional cohesion of an indicator network. This approach is applied to an environmental dataset of 19 indicators for Cruz Grande, Guerrero, Mexico (2010–2023). The results reveal that although the indicator network is relatively dense ($d=0.6199$), its structural–functional cohesion is low ($Mo\left(G\right)=520.68$), placing the region in the Fair category. This result provides an explanation for the sustained decline of the system, as shown by the PCA-based Regional Environmental Sustainability Index . We conclude that this approach is a complementary tool for diagnosing and evaluating environmental systems, enabling the detection of vulnerabilities that remain invisible to conventional aggregation methods. Full article

The concentrations of atmospheric particulate matter (PM₁₀ and PM_2.5) significantly impact global environment, human health, and climate change. This study developed a particulate matter concentration retrieval method based on multi-source data, proposing a dual-branch retrieval network architecture named CSLTNet that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. The CNN branch is designed to extract spatial features, while the LSTM branch captures temporal characteristics, with attention modules incorporated into both the CNN and LSTM branches to enhance feature extraction capabilities. Notably, the model demonstrates robust spatial generalization capability across different geographical regions.Comprehensive experimental evaluations demonstrate the outstanding performance of the CSLTNet model. For the Beijing–Tianjin–Hebei region in China: in PM₁₀ retrieval, sample-based 10-fold cross-validation achieved R² = 0.9427 (RMSE = $16.47\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$), while station-based validation yielded R² = 0.9213 (RMSE = $19.50\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$); for PM_2.5 retrieval, sample-based 10-fold cross-validation resulted in R² = 0.9579 (RMSE = $6.49\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$), with station-based validation reaching R² = 0.9296 (RMSE = $8.32\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$). For Northwest China: in PM₁₀ retrieval, sample-based 10-fold cross-validation achieved R² = 0.9236 (RMSE = $34.52\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$), while station-based validation yielded R² = 0.9046 (RMSE = $37.24\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$); for PM_2.5 retrieval, sample-based 10-fold cross-validation resulted in R² = 0.9279 (RMSE = $10.56\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$), with station-based validation reaching R² = 0.8787 (RMSE = $13.71\mathsf{\mu }\mathrm{g}/{\mathrm{m}}^3$). Full article

The Bactrian camel exemplifies mammalian adaptation to deserts, but the spatial organization of its innate immune system remains uncharacterized. This study integrated transcriptomes from 110 samples across 11 major tissues and organs to resolve tissue-specific gene expression and innate immune modularity. Through differential expression analysis, Tau specificity index (τ > 0.8), and machine learning validation (Random Forest F1-score = 0.86 ± 0.11), we identified 4242 high-confidence tissue-specific genes (e.g., LIPE/PLIN1 in adipose). Weighted gene co-expression network analysis (WGCNA) of 1522 innate immune genes revealed 11 co-expression modules, with six exhibiting significant tissue associations (FDR < 0.01): liver-specific (r = 0.96), spleen-adipose-enriched (r = 0.88), muscle-associated (r = 0.82), and blood-specific (r = 0.80) modules. These networks demonstrated multifunctional coordination of immune pathways—including Pattern Recognition, Cytokine Signaling, and Phagocytosis—rather than isolated functions. Our results establish that camel innate immunity is organized into spatially modular networks tailored to tissue microenvironments, providing the first systems-level framework for understanding immune resilience in desert-adapted mammals and may inform strategies for enhancing livestock resilience in arid regions. Full article

The ODYSEA (Ocean DYnamics and Surface Exchange with the Atmosphere) mission will provide simultaneous two-dimensional measurements of currents and winds for the first time. According to the ODYSEA radar concept, with a high incidence angle, current noise is primarily driven by backscattered power, which is triggered by wind speed. Therefore, random noise will affect the quality of observations. In low wind conditions, the absence of surface roughness increases the noise level considerably, to the point where the measurement becomes unusable, as the error can exceed 3 m/s at 5 km posting compared to mean current amplitudes of tens of cm/s. Winds higher than 7.5 m/s enable current measurements at 5 km posting with an RMS accuracy below 50 cm/s, but derivatives of currents will amplify noise, hampering the understanding of ocean dynamics and the interaction between the ocean and the atmosphere. In this context, this study shows the advantages and limitations of using noise-reduction algorithms. A convolutional neural network, a UNet inspired by the work of the SWOT (Surface Water and Ocean Topography) mission, is trained and tested on simulated radial velocities that are representative of the global ocean. The results are compared with those of classical smoothing: an Adaptive Gaussian Smoother whose filtering transfer function is optimized based on local wind speed (e.g., more smoothing in regions of low wind). The UNet outperforms the kernel smoother everywhere with our simulated dataset, especially in low wind conditions (SNR << 1) where the smoother essentially removes all velocities whereas the UNet mitigates random noise while preserving most of the signal of interest. Error is reduced by a factor of 30 and structures down to 30 km are reconstructed accurately. The UNet also enables the reconstruction of the main eddies and fronts in the relative vorticity field. It shows good robustness and stability in new scenarios. Full article

The assessment of forest health and terrain usability is closely tied to slope accessibility. Current methods for evaluating terrain accessibility based solely on slope characteristics often lack precision and fail to capture the combined effects of topography and vegetation. This study introduces succolarity, together with succolarity reservoir and delta (Δ) succolarity, as fractal-based measures for assessing undeveloped land accessibility. The analysis focused on two test areas: the Ceahlău Mountains and the Blaj–Vulpăr Hills. Results revealed lower accessibility values for the Ceahlău Mountains (0.01 to 0.23 for slopes of 0–5° and 0–30°) compared to the Blaj–Vulpăr Hills (0.035 to 0.598 for the same ranges). These significant contrasts demonstrate that terrain fragmentation and compact forests act as decisive constraints, with slope predominating in mountains and vegetation in hilly areas. The findings are valuable for environmental agencies, emergency services, and research groups studying land morphology and mobility. Practical applications include infrastructure planning, sustainable land-use management, and strategic operations in remote terrains. Incorporating additional datasets (e.g., hydrographic networks, seasonal vegetation) and refining methodologies will further enhance succolarity-based assessments, supporting sustainable development in challenging environments. Full article