Search Results (337)

Region-level weakly supervised hyperspectral target detection (HTD) using multiple-instance learning (MIL) reduces annotation costs but encounters challenges such as bag label ambiguity, boundary over-smoothing, and test-time computational latency. To address these issues, we propose Hyper-VMIL, a spatial–spectral topology-regularized variational hypergraph network. Hyper-VMIL formulates latent target localization as variational inference over dual-path hypergraphs: a boundary-aware spatial hypergraph modeling geometric patch continuity and a dynamic spectral-manifold hypergraph capturing non-local material similarity. Node-adaptive gating dynamically balances spatial and spectral evidence to mitigate over-smoothing near target boundaries. Furthermore, a confidence-aware continuous posterior refinement (CTPR) mechanism reduces the confirmation bias associated with conventional hard pseudo-label binarization. Finally, a teacher–student distillation strategy transfers contextual topology into a lightweight single-spectrum student detector. Benchmark experiments on simulated ASTER and airborne MUUFL Gulfport and Avon datasets show that Hyper-VMIL achieves competitive performance against 15 baseline methods. Notably, Hyper-VMIL supports dual inference modes: Context Mode provides improved detection accuracy (+4.6% average NAUC over VMIL-ECM on MUUFL), while Pixel Mode enables single-spectrum inference ($1.25\mathsf{\mu }\mathrm{s}$ single-instance latency and an amortized streaming throughput of $0.015\mathsf{\mu }\mathrm{s}$ per pixel) suitable for onboard real-time deployment. Full article

Macroplastic pollution in rivers is an increasing environmental concern because rivers function simultaneously as active transport pathways and temporary storage compartments for land-based plastic waste. This study investigated the spatial distribution, composition, and size characteristics of riverbank and floating macroplastic debris in the Can Tho River, a tidal tributary of the Hau River in the Mekong Delta, Vietnam, to improve understanding of macroplastic transport, selective retention, and environmental partitioning between active transport and temporary storage compartments. Riverbank debris was surveyed at twelve sites spanning urban, peri-urban, and rural sections, while floating debris was quantified using a net-based sampling system. Riverbank accumulations exhibited pronounced local spatial heterogeneity, although litter density and mass density did not differ significantly among river sections. Plastics dominated both environmental compartments, accounting for 53–60% of accumulated debris and more than 95% of floating debris by abundance. Riverbank accumulations were dominated by plastic bags, food packaging, and beverage containers, whereas floating debris was dominated by expanded polystyrene foam products. Significant differences were also observed in material composition, plastic-product composition, and size distribution. Riverbank accumulations contained proportionally larger macroplastics (100–500 mm), whereas floating debris was dominated by smaller macroplastics (50–200 mm), supporting the role of size-dependent transport and selective retention in environmental partitioning. These findings show that floating debris and riverbank accumulations represent complementary components of the riverine plastic continuum, linking active transport and temporary storage through selective environmental partitioning. Integrating floating and riverbank monitoring provides a more comprehensive framework for understanding macroplastic transport and environmental fate while informing management strategies to reduce downstream plastic transport to the Hau River and ultimately estuarine and coastal ecosystems. Full article

Background: Acute myeloid leukemia (AML) is a heterogeneous and aggressive hematologic malignancy, and LY-11h is a novel acylhydrazide-based histone deacetylase inhibitor with promising therapeutic potential for AML. However, its poor aqueous solubility, limited intestinal dissolution, and narrow therapeutic window hinder oral formulation development and motivate the development of dosage forms with flexible dose-design capabilities. Herein, an integrated hot-melt extrusion (HME)–fused deposition modeling (FDM) strategy was developed to convert LY-11h into printable amorphous solid dispersion (ASD) dosage forms. Methods: HPMC-AS was used as a pH-responsive carrier to enhance intestinal release while restricting premature gastric release, and HPC-EF was incorporated to improve filament processability. Single-factor and DoE studies identified critical formulation and process variables and established formulation–process–property relationships, while machine learning further modeled nonlinear interactions and guided optimization. In-line near-infrared spectroscopy combined with polarized light microscopy enabled real-time monitoring of LY-11h amorphization and melt homogenization during HME. Results: ExtraTrees and Bagging models showed promising predictive performance for key filament properties, and PAT-stage validation confirmed strong agreement with experimental values. The 15 DoE-designed ASD filaments were successfully fabricated into FDM-printed tablets with reproducible geometry. Equilibrium-solubility and in vitro dissolution studies demonstrated enhanced intestinal-pH solubility and reproducible pH-responsive release. Conclusions: Collectively, these findings establish a technological proof of concept for the manufacture of LY-11h dosage forms with adjustable formulation and geometric attributes. Further in vivo pharmacokinetic studies are required to determine whether these manufacturing capabilities translate into predictable dose–exposure relationships and individualized dose control. Full article

This study evaluated the effects of placing wood chips (F-Wood chip) or pumice (F-Pumice) at the bottom of poultry feather compost on open-field ginger soilless media production. Root control bags were prepared with 10 L of wood chips or pumice overlain by 20 L of feather compost, with three ginger rhizomes planted. Crops were drip-irrigated without synthetic fertilization and replenished with compost three times. Results indicated that bottom-placed wood chips or pumice improved water infiltration. Ginger yield was significantly higher in the F-Wood chip treatment than in the F-Pumice, with fresh weights of 3.4 and 2.5 kg, and dry weights of 451.8 and 332.7 g, respectively. Furthermore, F-Wood chip significantly increased rhizome calcium levels. Although no significant differences were observed between treatments regarding leaf and post-harvest media nutrient contents, the F-Wood chip group exhibited higher microbial abundance (9.5 ± 4.5 × 10⁵ CFU ⁻¹) and greater endophytic diversity, spanning 7 genera and 11 species with potential plant growth-promoting and stress-resistance functions. Overall, this innovative integrated crop management strategy demonstrates great potential to substitute for fossil-fuel-based chemical fertilizers, this innovative production mode eliminates the need for fossil-fuel-based chemical fertilizers, offering an applicable and sustainable soilless cultivation solution for open-field ginger production under extreme weather conditions like typhoons and heavy rainfall. Full article

Reliable photovoltaic (PV) operation requires monitoring strategies that can detect performance degradation before it develops into persistent efficiency loss. This study proposes an interpretable data-driven framework for condition monitoring and early warning of low-efficiency events using only inverter-based electrical measurements. The novelty of this study lies in its focus on detecting low-efficiency operating conditions from inverter electrical data, rather than merely classifying individual PV fault types. Thirty-minute operational data from a 110 kW grid-connected PV plant in Kastamonu, Türkiye, covering January 2023–December 2025, were analyzed. Phase currents, phase voltages, total active power, and DC power were transformed into electrical health indicators, including mean current, mean voltage, current and voltage variability, phase imbalance index, and conversion efficiency. Correlation and imbalance analyses showed highly synchronized three-phase operation, with a mean phase imbalance index of 0.004865. Conversion efficiency remained stable, with an instantaneous mean of 0.964. Generalized Additive Model results explained 66.4% of efficiency variability and identified mean current as the dominant nonlinear determinant, while phase imbalance acted as a secondary but significant factor. A Random Forest classifier achieved 96.34% accuracy, 3.87% out-of-bag error, and 53.4% recall for rare low-efficiency events. Decision-tree rules indicated high risk when mean current fell below 9.4 A and very low risk above 12 A. The framework provides a practical, sensor-minimal, and interpretable approach for PV performance monitoring and proactive maintenance. Full article

Background/Objectives: This study investigated the impact of feature type, feature dimensionality, and feature selection strategies on intensive care unit (ICU) mortality prediction using first-day portable frontal chest radiographs and demographic data. In addition, the study evaluated whether comparable predictive performance could be achieved using reduced, clinically interpretable feature sets. Methods: A total of 500 patients were included, comprising 400 cases for model development and 100 independent cases for testing. Two complementary experimental frameworks were evaluated. In Experiment Set-1, a clinically interpretable feature set consisting of 12 radiographic and demographic variables was analyzed. In Experiment Set-2, a high-dimensional feature space consisting of 74 radiographic, radiological, radiomic, and image-derived features was investigated using six feature selection methods: ANOVA, Chi-Square, Kruskal–Wallis, MRMR, ReliefF, and Shapley-based importance analysis. Machine learning models were evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and F1-score. Results: In Experiment Set-1, feature selection reduced the number of predictors from 12 to 4. Using the complete 12-feature set, the Bagged Trees classifier achieved the highest performance (AUC = 0.95). Following feature reduction, the Subspace KNN classifier achieved an AUC of 0.96 with an accuracy of 0.82, indicating that satisfactory predictive performance could be achieved using a substantially smaller feature set. In Experiment Set-2, comparison of six feature selection strategies demonstrated that MRMR and Kruskal–Wallis produced the highest-performing feature subsets. Cobb angle, bilateral infiltrates, bilateral pleural effusion, and unilateral pleural effusion were consistently identified across all feature selection methods as stable predictors associated with ICU mortality. These predictors were considered stable because they were consistently identified across multiple independent feature selection methods despite differences in the underlying selection algorithms. Conclusions: The findings suggest that appropriate feature selection strategies may reduce model complexity while retaining clinically relevant predictive information. The consistent identification of a small group of radiographic features across multiple feature selection methods in the 74-feature experimental framework suggests that these variables represent robust and clinically meaningful predictors of ICU mortality. Further external validation using larger multicenter datasets is required before clinical implementation. Full article

Interplanting Paeonia lactiflora Pall. under Phyllostachys edulis (Carrière) J. Houzeau (moso bamboo) forests forms an innovative agroforestry intercropping system. This practice has great potential to improve economic benefits and maintain ecosystem stability of moso bamboo forests. Nevertheless, the physiological and growth adaptation mechanisms of P. lactiflora in the understory microenvironment of moso bamboo are still poorly understood. In this study, a two-factor field experiment was conducted to examine how moso bamboo stand density (0, 1200, 1800, 2400 plants·ha⁻¹) and cultivation pattern (bag cultivation, field cultivation) shape the growth, root traits, and resource allocation of P. lactiflora. The results showed that under bag cultivation, the low bamboo stand density (1200 plants·ha⁻¹) promoted biomass accumulation in P. lactiflora leaves, stems, and roots, and facilitated root elongation and radial thickening. Under this condition, P. lactiflora adopted a root-prioritized growth strategy and preferentially allocated nitrogen and phosphorus to leaves and roots. For field cultivation, P. lactiflora growth was inhibited by dual stresses of aboveground shading and underground interspecific root competition. To cope with such stress, P. lactiflora prioritized leaf growth, allocated more nitrogen and phosphorus to leaves, and developed thinner roots. This study reveals the differential growth, root plasticity, and resource allocation strategies of P. lactiflora in response to dual stresses of aboveground shading and underground root competition from moso bamboo, filling the research gaps regarding multi-factor adaptation mechanisms of understory herbaceous plants in moso bamboo forests. The optimal undercropping regime combining low moso bamboo stand density with bag cultivation pattern is identified. These findings offer practical references for the efficient and sustainable management of bamboo–peony agroforestry systems and are critical for balancing the economic benefits of moso bamboo stands and maintaining the stability of understory ecosystems. Full article

Soil monitoring is a key component of precision agriculture and environmental sensing systems, where reliable measurements support irrigation management and crop monitoring. Although high-end sensing platforms provide accurate measurements, their cost limits widespread adoption, particularly in resource-constrained agricultural environments. Low-cost soil sensors, such as widely available 7-in-1 probes capable of measuring soil moisture, temperature, electrical conductivity, and pH, offer a scalable alternative for distributed monitoring; however, their limited accuracy raises concerns regarding their reliability for decision-support systems. This study investigates whether measurements from a single low-cost 7-in-1 soil sensor contain sufficient information to reconstruct the outputs of a commercial high-end sensing platform (CropX), specifically volumetric water content (VWC) and pore-water electrical conductivity (ECpw). Field data were collected in a tropical Hass avocado orchard in Colombia, and four machine learning models were evaluated to reconstruct CropX measurements from low-cost sensor signals at three soil depths (20, 41, and 66 cm). Random Forest achieved the highest reconstruction performance, with coefficient of determination R2 values between 0.9965 and 0.9986 and consistently low root mean square error (RMSE) and mean absolute error (MAE) across depths. Out-of-bag validation and multi-seed stability analyses confirmed the robustness of the models despite the limited dataset size. A chronological validation (80–20%) showed substantially reduced performance, indicating that the proposed approach is more suitable for reconstructing high-end sensor signals under concurrent measurement conditions than for strict temporal extrapolation. Therefore, the framework should be interpreted as a virtual sensing strategy for reconstructing simultaneous CropX measurements from low-cost sensor observations rather than as a standalone model for predicting future soil conditions without periodic recalibration. These results demonstrate that low-cost multi-parameter sensors can support high-fidelity virtual reconstruction of high-end soil measurements, contributing to the development of scalable and cost-effective soil monitoring systems for precision agriculture. Full article

The spatial relationships between lightning-induced fire ignition and topography, vegetation, climate, and weather were analyzed in the region of East Macedonia and Thrace, northeastern Greece. The study was based on reported lightning-induced ignitions during the 2009 fire period. Lightning data for the same period was provided by the ZEUS lightning detection network operated by the National Observatory of Athens, while fire statistics were obtained from the official records of the Greek Fire Service. A total of 198 lightning strike events (66 fire ignitions and 132 non-fire events) were used for model development. Statistical models based on Logistic Regression (LR) and random forests (RF) were developed to estimate the probability of lightning-induced fire using topography, climate, weather, and vegetation indices as predictor variables. According to the analysis results, the probability of an area being affected by lightning-induced fire is primarily determined by the Normalized Difference Vegetation Index (NDVI) and the accumulated precipitation in 24 h equal to or less than 2.5 mm expressed by Dry Thunderstorm (DT) day occurrence in this dataset. The logistic regression model achieved an area under the ROC curve of 0.94 and an overall classification accuracy of 91.9%, while the random forest model produced an Out-Of-Bag (OOB) error rate of 3.0%. Although the models have not been subjected to independent validation and include a single year’s data, the results demonstrate high internal classification performance and provide valuable insights into the primary drivers of fire ignition following lightning strikes in the study region. The outcomes of the present study will be useful in assessing spatially explicit fire risk, the planning and coordination of efforts to identify high-fire-risk areas, and designing long-term fire management and climate change adaptation strategies. Full article

Glacier boundary extraction on the Tibetan Plateau (TP) faces persistent challenges due to rugged terrain, seasonal snow, extensive debris cover, and topographic shadows. Traditional methods utilizing single-source or single-temporal data often yield limited accuracy. Thus, we propose an automated Double Random Forest (Double-RF) framework integrating single- and multi-temporal features from Sentinel-1 (SAR) and Sentinel-2 (Optical) data within the Google Earth Engine. We established a multidimensional feature space comprising spectral, textural, polarimetric, and topographic attributes. Feature optimization was performed using importance metrics and out-of-bag (OOB) error. A hierarchical classification strategy was employed: the first RF identifies clean glaciers and glaciers in shadow, while the second RF executes refined boundary extraction of debris-covered glaciers to mitigate spectral confusion. The results indicate that the Double-RF method significantly achieves an overall accuracy exceeding 0.84 across all sub-basins and reaching above 0.95 at best. The derived glacier inventory reveals a distinct spatial pattern: higher concentrations in the western and peripheral regions compared to the eastern and interior TP. Glaciers are predominantly distributed on shaded aspects with gentle-to-moderate slopes, highlighting the combined influence of climatic gradients and topographic controls. This multi-source, multi-temporal fusion strategy provides a robust methodological foundation for long-term glacier monitoring over the TP. Full article

Accurate and robust localization based on semiconductor optoelectronic LiDAR sensing is a fundamental prerequisite for autonomous navigation of mobile robots in complex scenarios. Traditional positioning methods based on semiconductor optoelectronic LiDAR sensors suffer from cumulative drift during long-term operation, while prior map-based positioning techniques often lack adaptability to dynamic environments. To address these challenges, this paper proposes a high-performance LiDAR-IMU tightly coupled positioning system integrating prior map constraints, LIO optimization, and an adaptive failure detection-relocalization mechanism. A high-precision global map constructed by LIO-SAM serves as the prior constraint to ensure global consistency of pose estimation, while the tightly coupled LIO framework maintains high accuracy and low latency in high-dynamic scenarios. The proposed dual-index failure detection strategy identifies localization anomalies in real time, and a Bag-of-Words (BoW)-based relocalization module rapidly restores precise positioning. Extensive simulations on MARSIM and physical experiments in structured, semi-structured, and feature-sparse environments demonstrate that the proposed system outperforms state-of-the-art (SOTA) methods including LIO-SAM, Ada-LIO, and Map-ICP. Specifically, the system achieves an Absolute Trajectory Error (ATE) root mean square (RMSE) of ≤0.06 m in physical experiments, a cumulative drift of ≤0.1 m per 100 m, and a relocalization success rate of ≥90% in feature-sparse scenes. These results validate the system’s superiority in accuracy, robustness, and real-time performance, providing a reliable signal-processing solution for semiconductor optoelectronic LiDAR-based sensing systems in complex practical applications. Full article

Ring finger protein 126 (RNF126) is a RING-type E3 ubiquitin ligase that has recently emerged as a multifaceted regulator of cellular homeostasis, stress adaptation, and disease progression. Through its structurally distinct zinc-finger and catalytic RING domains, RNF126 orchestrates substrate recognition and ubiquitin transfer, generating diverse ubiquitin linkages with both proteolytic and nonproteolytic functions. Initially characterized as a component of the protein quality control (PQC) machinery, RNF126 cooperates with chaperones such as BAG6 and UBQLN1 to eliminate mislocalized and misfolded proteins, thereby maintaining proteostasis. Beyond PQC, RNF126 plays pivotal roles in DNA damage response pathways by regulating homologous recombination, non-homologous end joining, checkpoint signaling, and genome stability through substrates, including MRE11, Ku80, RNF168, and 14-3-3σ. Genetic studies have further demonstrated its importance in embryogenesis and male fertility, and accumulating evidence has identified RNF126 as a critical driver of malignancy in multiple cancers. RNF126 promotes tumor progression by degrading or modulating key regulators, such as p21, PTEN, p53, PDKs, and LKB1, thereby enhancing proliferation, metabolic reprogramming, anoikis resistance, metastasis, and chemo/radioresistance. Intriguingly, RNF126 exhibits context-dependent functions, acting as an oncogene or tumor suppressor depending on the tissue type and substrate selection. In addition to cancer, RNF126 has been implicated in neurodegeneration, cardiac pathology, antiviral immunity and adaptive immune regulation. This review summarizes the current knowledge of RNF126 structure, ubiquitin signaling mechanisms, physiological functions, and pathological roles, while discussing emerging therapeutic strategies and future challenges for targeting RNF126 in precision medicine. Full article

Fresh broccoli is highly perishable, exhibiting rapid yellowing and quality deterioration with a short shelf life. In this study, we investigated the effects of nanomaterial-modified atmosphere packaging (MAP) bags with different thicknesses, designated as 2.5C (25 μm) and 4C (40 μm), on the physiological and biochemical changes in broccoli were evaluated during storage at 20 ± 1 °C for 8 days. Results showed that both MAP treatments remarkably delayed floret senescence by inhibiting the rapid color transition from green to yellow, as indicated by alterations in L*, a*, b*, and hue angle values, as well as by suppressing chlorophyll degradation. The 2.5C treatment exhibited a more pronounced effect during storage. MAP treatments helped maintain commercial quality by preserving total phenols and vitamin C (Vc) content, retaining stem firmness and surface glossiness, regulating post-opening respiration rate and reducing water loss. MAP treatments also effectively suppressed the accumulation of superoxide anion (O₂•⁻) and hydrogen peroxide (H₂O₂). Furthermore, MAP treatments enhanced free radical scavenging capacity, as demonstrated by DPPH and ABTS assays and the O₂•⁻ scavenging rate in broccoli. These results indicate that MAP treatment with an appropriate thickness (e.g., 2.5C) effectively inhibits excessive ROS production and enhances antioxidant capacity, thereby delaying floret chlorophyll degradation and senescence. This study provides a foundation for developing effective and green preservation strategies using physical MAP treatments for fresh broccoli. Full article

Accurate fault classification of electro-hydrostatic actuators (EHAs) remains challenging because multivariate fault signals contain local transient variations, inter-variable coupling, and dynamic temporal dependencies that are difficult to capture simultaneously using a single model. To address this problem, this paper proposes a multi-view temporal feature collaborative heterogeneous ensemble learning model (MTF-HEM) for EHA multivariate time series fault classification. MTF-HEM integrates a representative subsequence-guided time series forest (RSG-TSF), XGBoost, and a lightweight LSTM to extract local morphological, global statistical, and temporal dependency features, respectively. The outputs of these heterogeneous base learners are fused using a bootstrap-driven out-of-bag probability binning stacking (BOPB-stacking) strategy. The proposed method was evaluated on an AMESim-based simulated EHA plunger pump fault dataset containing one normal condition and six fault conditions. Under the present simulation setting, MTF-HEM achieved an accuracy of 99.52% and outperformed the tested deep time series classification models, ensemble models, and individual base learners. These results suggest that multi-view heterogeneous feature fusion can improve the classification of simulated EHA fault time series and provide a methodological reference for intelligent actuator fault diagnosis. However, the current validation is based on data generated from a single AMESim simulation model, and further evaluation on real EHA systems is needed to assess the practical applicability and generalizability of the proposed approach. Full article

Maize, as a major cereal crop in China, is vital for national food security, and appropriate nitrogen fertilization is essential for its growth and yield. Avoiding excessive nitrogen fertilizer application while maintaining productivity remains a critical challenge for sustainable agriculture. Although straw returning is widely adopted to reduce chemical fertilizer inputs, its effectiveness is often regionally constrained. In the West Liaohe Plain, low temperature and spring drought limit straw decomposition and nutrient release, making it difficult to reduce nitrogen fertilizer input and improve fertilizer use efficiency. Therefore, this study examined the effects of different nitrogen management modes on straw decomposition, nutrient release, mineral fertilizer substitution potential, soil available nutrients, and maize yield under shallow buried drip irrigation with integrated water and fertilizer management. A field experiment was conducted with five nitrogen (N) fertilizer management treatments: a conventional fertilization treatment (CK), in which 15% of total N was applied as starter fertilizer; two increased starter N treatments, in which 30% (30%N) and 45% (45%N) of total N were applied as starter fertilizer; and two organic substitution treatments, in which 30% (30%ON) and 45% (45%ON) of mineral N fertilizer were substituted with decomposed sheep manure based on equivalent total N input. Straw decomposition and nutrient release were measured using the nylon mesh bag method and fitted with an exponential decay model. The mineral fertilizer substitution potential was estimated based on straw nutrient release, while soil available nutrient dynamics in the 0–40 cm soil layer were analyzed, and the Mantel test and PCA were used to assess their relationships. Organic substitution promoted straw decomposition. The 30%ON treatment showed the highest rate at 70.91%, which was 19.2% higher than that of CK, and it exhibited a higher theoretical maximum decomposition rate (a), higher decomposition rate constant (k), and a shorter half-life. All treatments increased nutrient release and soil available nutrients, and organic substitution demonstrated stronger temporal persistence and more uniform vertical distribution among soil layers. The 30%ON treatment increased straw nutrient release by 4.8% to 18.2% and enhanced mineral fertilizer substitution potential. Although the 30%ON treatment did not increase yield in the first experimental year, it showed a significant yield advantage in the second year, which coincided with greater straw nutrient release and higher soil available nutrient levels under this treatment. Substituting 30% of mineral N fertilizer with organic fertilizer under shallow buried drip irrigation (300 kg N ha⁻¹) optimized the C/N balance of the input system and facilitated straw decomposition and nutrient release. The continuous accumulation of soil available nutrients under this treatment, together with sustained straw nutrient release, was associated with a significant yield advantage in the second experimental year. Therefore, the 30%ON treatment may represent an appropriate management strategy for coordinating straw resource utilization, soil fertility maintenance, and stable maize production in the West Liaohe Plain. Full article