Browse Articles
Discover research articles across all indexed journals
Characterizing the flow-pressure relationship for flexible and navigable suction ureteral access sheaths in retrograde intrarenal surgery
First evidence of nanoplastics in Antarctica soil
Abstract Plastic contamination has become a global concern, with evidence even in remote regions like Antarctica. While macro- and microplastics have been documented in Antarctic marine ecosystems, their presence in soils - particularly submicro- and nanoplastics - remains poorly studied. This study analyses soil samples from the McMurdo Dry Valleys collected on January 8th to 28th, 2023, and reports the first detection of nanoplastics - including polypropylene, polyethylene, polyethylene terephthalate, polystyrene, polyvinyl chloride, and tyre wear particles - using thermal desorption proton transfer reaction mass spectrometry. These plastics were detected at multiple topsoil sampling sites ( n = 13), with concentrations reaching up to 295 ng g⁻¹ with nanoplastics detected above polymer-specific method detection limits at 54% of sites (median: 26.6 ng g⁻¹). They were also detected at lower concentrations in deeper soil layers (> 20 cm; n = 4), where nanoplastics were present at 50% of the sampled sites (median: 1.95 ng g⁻¹). Lagrangian particle dispersion model FLEXPART suggested seasonal deposition patterns, with inputs from both local sources and long-range atmospheric transport. This evidence shows that soils in one of Earth’s most pristine environments are not exempt from plastic contamination, with the reported concentrations providing a crucial baseline for global pollution assessments. These findings also highlight the urgent need to study plastic fate, transport, and ecological impacts in polar regions.
Zero-shot classification of ECG signals using CLIP-based models
Abstract In this study, we present a comprehensive evaluation of Contrastive Language-Image Pre-training (CLIP) for zero-shot electrocardiogram (ECG) classification across multiple datasets and diagnostic classes. Traditional ECG classification models require significant labeled training data for every diagnostic class, limiting their adaptability to new or unseen classes. To address this limitation, we trained and evaluated 24 CLIP-based models, composed of one image encoder (CNN Base, CNN V2, CNN V3, RNN, ISIBrno) and one text encoder (BioBERT, Bio+ClinicalBERT, bert-case-uncased) on 27 seen classes on three datasets (PTB-XL, Ningbo, and Gerogia) and evaluated zero-shot classification performance on 11 unseen classes. We investigate the impact of training dataset size, encoder architectures, and pretraining effects on both in-distribution and out-of-distribution generalization performance. Our experiments included internal evaluation (Experiment A) and external validation on two independent datasets (SPH and CODE-15%, Experiment B). The top-performing models achieves a macro-averaged ROC-AUC of 0.70 for zero-shot out-of-distribution classification, 0.70 for zero-shot in-distribution classification, and 0.83 for out-of-distribution classification with classic training. These results demonstrate that CLIP-based models can meaningfully classify ECG conditions beyond the scope of their training, offering a flexible alternative to current ECG diagnostic systems where clinical needs are constantly evolving.
Serum vitamin D level in children with bronchiolitis
Mathematical modeling of peritoneal buffer transport and acidosis correction in patients on peritoneal dialysis
Abstract Assessing buffer base kinetics in dialysate using the three-pore model of peritoneal transport is not enough to predict changes to the acid-base equilibrium in blood during peritoneal dialysis (PD). We complemented the three-pore model with a model of acid-base homeostasis to evaluate differences between PD sessions using dialysis fluid buffered with lactate (PD4) or bicarbonate/lactate (B/L). Six patients underwent two 4-hour dwells with either B/L or PD4; blood and peritoneal fluid sampling was used to quantify intraperitoneal volume—using a radioiodinated albumin as volume marker—and solute concentrations. The integrated model describes CO 2 and O 2 transport across central circulation and tissues, predicting changes in acid-base equilibrium driven by peritoneal transport of bicarbonate, lactate, and dissolved CO 2 . The model accurately predicted dialysate concentrations of bicarbonate and dissolved CO 2 , with errors in the range 5–10% of measured values for both fluids, and with similar prediction errors for plasma concentrations despite tuning the model to dialysate data only. Estimated transport parameters fell into the expected ranges and were generally comparable between fluids. These results validate our understanding of transport kinetics and acid-base homeostasis during PD, demonstrating that different buffer compositions do not appear to impact small solute transport nor mechanisms of acid-base regulation.
A novel dynamic risk assessment tool for evaluating cybersecurity risks in the digital industry
Adoption and heterogeneous impacts of hired mechanization services among smallholder farmers in Kenya, Zambia, and Zimbabwe
Assessing the validity of self-reported weight, height, and body mass index among hispanic breast cancer survivors
Development of a hybrid RSM–Fermatean fuzzy WASPAS framework for multi-objective drilling optimization of WAAM nickel–stainless steel bimetallics
Abstract This study develops an integrated RSM–Fermatean fuzzy WASPAS framework for multi-response drilling optimization of wire arc additively manufactured ERNiCrMo-3/316LSi nickel–stainless steel bimetallic structures. WAAM bimetallics often contain bead waviness, geometric deviation, residual stress, thermal distortion, and microstructural heterogeneity, which make drilling performance difficult to optimize through a single response. The effects of cutting speed, feed rate, tool point angle, and coolant condition were evaluated through twelve drilling trials. Four responses were considered: surface roughness, material removal rate, hole dimensional accuracy, and energy consumption. The measured results showed clear variation across the tested conditions. Surface roughness ranged from 0.666 to 2.166 μm, material removal rate from 10 to 75 mm³/min, hole dimensional accuracy from 90 to 99%, and energy consumption from 120 to 300 W. Response Surface Methodology was used to model factor effects and interaction behavior, while Fermatean fuzzy logic was applied to reduce the influence of response uncertainty in ranking. WASPAS combined the weighted criteria into a single ranking index to identify the best compromise among the tested alternatives. The selected condition was 50 m/min cutting speed, 0.1 mm/rev feed rate, 135° tool point angle, and oil-based cooling, producing Ra = 1.1667 μm, MRR = 50 mm³/min, HDA = 96%, and EC = 220 W. This condition offered a balanced trade-off among surface quality, productivity, dimensional accuracy, and energy demand. The observed performance is associated with reduced chip load, stable tool engagement, improved chip evacuation, lower friction, and controlled heat generation at the tool–workpiece interface. Statistical validation and sensitivity analysis showed that the ranking remained stable within the tested experimental range. The proposed framework provides a reproducible decision-support method for drilling the investigated WAAM bimetallic system and requires validation before use in other WAAM material combinations.
Metastable multimeric G-quadruplex 2′FY-RNA aptamers that selectively bind pyoverdines
Adipose-derived mesenchymal stem cell-conditioned medium and proliferation of intestinal stem cells in mesenteric ischemia and reperfusion
A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices
Abstract Intelligent landscape recognition sits at the heart of smart coastal tourism, yet pushing vision Transformers onto resource-constrained edge hardware remains awkward—the self-attention mechanism scales quadratically with token count, and marine atmospheric haze quietly erodes recognition accuracy long before it is noticed in the laboratory. We propose a hybrid lightweight architecture that swaps standard self-attention in the early stages for depthwise separable convolution token mixers, switches to kernel-based linear attention in the deeper stages, and follows an adaptive channel reduction policy that trims feature dimensions where redundancy is empirically highest. A Fog-Aware Feature Calibration module, motivated by the Koschmieder atmospheric scattering model, is embedded between the convolutional and attention stages so that learned dehazing happens inside the network rather than as a separate preprocessing pass. Training proceeds through a three-phase pipeline that interleaves progressive structured pruning with temperature-scheduled knowledge distillation from a Swin-Base teacher; the composite objective shrinks the model to 4.8 M parameters and 0.91 GFLOPs without abrupt accuracy collapse. On a newly collected 17,565-image dataset spanning eight coastal scene categories across six shoreline regions of Zhejiang Province, our model reaches 92.6% ± 0.4% Top-1 accuracy (mean ± std over three independent runs), the best result among all baselines below 6 M parameters, including the recently released MobileViT-v2, FastViT, TinyViT, EfficientViT and RepViT. On an NVIDIA Jetson Orin Nano with TensorRT INT8 optimisation the system delivers 18.3 ms latency (54.6 FPS) within a 7.4 W average power envelope, and a 200 m visibility heavy-fog subset that mixes synthetic and real captures shows an 85.7% accuracy retention rate—6.8 percentage points above the strongest baseline. Cross-dataset zero-shot tests on Fujian, Hainan and the Places365 coastal subset, together with deployment benchmarks on Jetson Nano 4 GB and Coral Edge TPU, suggest the design transfers reasonably—though not effortlessly—beyond the original collection sites.
Experimental and probabilistic reliability of fiber-reinforced concrete (FRC) under drop-weight impact
Design and Manufacturing Principles of Smart Fibers
ABSTRACT The long‐standing role of fibers as basic structural components has given way to a new paradigm of smart fibers that integrate environmental sensing with programmable responses. Despite recent progress in materials chemistry and device engineering, fabrication choices are often guided by cost and convenience at the expense of advanced capabilities. This review reframes manufacturing as a central design axis that governs not only morphological properties but also molecular orientation, hierarchical organization, and interfacial properties, all of which determine performance. In this review, we first discuss the design principles of activation triggers and response mechanisms, highlighting how they interplay with the unique geometry of fibers. We then provide a detailed comparison of manufacturing strategies, including continuous single fiber spinning, non‐woven deposition, and direct synthesis, emphasizing their distinct trade‐offs in scalability, resolution, and material compatibility. We further examine post‐processing and multiscale integration as steps that preserve, refine, and extend fiber‐level properties into system‐level functions. Finally, we discuss emerging directions toward fiber electronics and intelligent systems that merge physical in‐fiber processing with data‐driven interpretation. Together, these principles establish a task‐dependent framework that aligns design, manufacturing, and function, advancing smart fibers from isolated demonstrations to integrated next‐generation devices.
Surface SO <sub>x</sub> Species Stabilized Metal–Oxygen Bonds in PtNi Nanoalloy for Highly Efficient and Durable Seawater Hydrogen Production
ABSTRACT Regulation of metal–oxygen (M–O) bonds and incorporation of sulfur are promising approaches for designing electrocatalysts for efficient hydrogen production from seawater. However, the durability of these catalysts is limited by the instability of M–O bonds under cathodic conditions and poisoning caused by the uncontrolled introduction of sulfur. In this study, a “killing two birds with one stone” strategy was developed for creating efficient hydrogen production electrocatalysts that involves anchoring PtNi nanoparticles on hydroxyl‐functionalized carbon nanotubes and modifying the surface with SO x species (S‐PtNi/CNTs) via a one‐pot hot‐injection‐combined wet‐chemical synthesis protocol. In this material, hydroxyl‐functionalized CNTs stabilize PtNi through M–O bonding, while SO x species tune the electronic structure of Pt and enhance M–O bond stability. Consequently, S‐PtNi/CNTs exhibits outstanding hydrogen production performance in alkaline seawater, delivering a 5.4‐fold increase in mass activity and a 21‐fold increase in specific activity compared to commercial Pt/C, along with remarkable stability over a 1000‐h operation period. Furthermore, S‐PtNi/CNTs significantly outperform commercial Pt/C in both a photovoltaic‐electrocatalysis electrolyzer and an anion‐exchange‐membrane water electrolysis system. The results of in situ spectroscopy and theoretical calculations confirm that SO x species stabilization of M–O bonds and improved chloride ions resistance are responsible for the superior performance of S‐PtNi/CNTs.
Risk factors for acquisition of SARS-CoV-2 Omicron variant among a prospective cohort of French Healthcare Workers: impact of hybrid immunity
Stacked graph attention network hypertuned by Heteroscedastic and Evolutionary Bayesian Optimization for fault identification on nuclear power plants robust to sensor drift
Abstract Fault identification in Nuclear Power Plants (NPPs) is critical for ensuring operational safety, reliability, and efficiency. Traditional diagnostic methods often rely on physical models and expert systems, which may struggle to capture the complex dynamics of transient events. To overcome these limitations, this paper proposes an optimized stacked Graph Attention Network (GAT) for fault detection in NPPs by modeling the complex interdependencies among system components as graphs. Transient operational data are transformed into graph representations, where nodes correspond to system variables, and edges capture physical relationships. The architecture of the proposed model is optimized using a Heteroscedastic and Evolutionary Bayesian Optimization (HEPO), ensuring the use of the best configuration. The proposed GAT-based model, hypertuned by HEPO, is trained to recognize patterns associated with both normal and faulty transient conditions, including sensor anomalies and actuator failures. Based on synthetic data generated from the Personal Computer Transient Analyzer (PCTRAN), the proposed model achieved results above 0.96 for accuracy, precision, recall, and F1-score in a statistical analysis.