Search Results (7,182)

β-Nicotinamide mononucleotide (NMN) is an important intermediate in nicotinamide adenine dinucleotide (NAD⁺) metabolism and has attracted increasing interest as a bioactive compound and biomanufacturing product. In this study, Pseudomonas putida KT2440 was engineered for NMN production using a “block–enhance–transport” strategy. Deletion of nicB impaired nicotinic acid degradation and resulted in the accumulation of 66.2 μM nicotinic acid in cell extracts, whereas additional deletion of the putative NMN-consuming genes pncC and ushA did not lead to detectable NMN accumulation. Coexpression of endogenous pncB and engineered Francisella tularensis nadE* enabled low-level NMN formation through a Preiss–Handler pathway-based route. By contrast, overexpression of endogenous nadA, nadB, and nadC strengthened precursor supply through the NAD⁺ de novo biosynthetic pathway and resulted in approximately 0.17 mM NMN in cell extracts. Chromosomal integration of an engineered Salmonella enterica pnuC* transporter cassette was associated with pronounced extracellular NMN accumulation. Additional overexpression of genes involved in downstream NAD⁺ metabolism or phosphoribosyl pyrophosphate supply did not considerably improve production, possibly because of metabolic competition or expression burden. The best-performing strain, LW10, produced 1.28 mM extracellular NMN, corresponding to approximately 0.43 g/L, after 96 h of shake-flask cultivation in basal salt medium containing glucose and L-aspartate. These results suggest the feasibility of NMN biosynthesis in engineered P. putida KT2440 and highlight the importance of balancing precursor supply, competing reactions, and product transport. Thus, P. putida KT2440 represents an alternative chassis for further pathway balancing and process optimization toward fermentative NMN production. Full article

Satellite precipitation products like GPM IMERG are crucial for hydrological modeling and disaster prevention; yet, their reliability in complex mountainous monsoon regions remains challenging. While the latest IMERG V07 introduces key upgrades, including a Climatological Calibration Algorithm (CCA), its cross-scale error propagation mechanisms and performance heterogeneity in complex underlying surfaces are poorly understood. This study evaluates the daily and monthly performance of IMERG V07 and V06 (Early, Late, and Final Runs) from 2014 to 2020 against 91 rain gauges in Guangxi, China—a typical mountainous monsoon region. The evaluation employs multiple statistical metrics and a multi-dimensional stratification approach based on elevation, precipitation intensity, and seasonality to quantify error propagation and climate-topography coupling effects. The results reveal that V07, particularly the Late Run, enhances daily precipitation detection capabilities, it significantly increases the proportion of systematic positive bias from 62.3 to 64.8% (V06) to 67.2–68.9% (V07). Consequently, upon temporal aggregation to the monthly scale, this systematic overestimation is severely amplified, leading to degraded performance, with the Final Run suffering the most substantial accuracy loss. Furthermore, retrieval accuracy is heavily constrained by surface heterogeneity, with systematic overestimation surging in areas where relatively dry (mean annual precipitation < 1300 mm) and complex terrain (elevation 100–500 m) coincide. The introduced CCA effectively improved dry season estimations but failed during wet season by introducing substantial positive biases. Ultimately, while V07 better captures short-term precipitation dynamics, its structural systematic biases compromise long-term cumulative reliability, highlighting the necessity for physics-based bias correction in hydrological applications and dynamic calibration in future algorithm upgrades. Full article

Reliable perception in smoke-filled tunnels is essential for rescue robots, yet the temporal evolution of LiDAR degradation under realistic smoke conditions is not well quantified. This paper investigates the degradation of LiDAR data induced by environmental factors in a full-scale tunnel experiment involving three smoke scenarios, with real combustion smoke and theatrical smoke. To enable consistent comparison across experiments with different smoke dynamics, an RGB-based visibility reference is first used for temporal alignment across measurements. Based on this alignment, physically interpretable LiDAR indicators, such as intensity attenuation and range-dependent point density loss, are used to characterize smoke-induced changes in the LiDAR data. In addition, an established Deep semi-supervised anomaly detection (DeepSAD) model is employed to derive a continuous data-driven degradation score that indicates deviations from nominal LiDAR range image patterns. The learned score remains stable under clear-air conditions, despite geometric variations caused by object and sensor movement. During smoke exposure, the score increases in all smoke scenarios, although the temporal evolution differs between scenarios. The results show that the degradation score derived from DeepSAD provides a continuous data-driven description of changes in LiDAR range images under smoke exposure. Overall, the study presents an experimental analysis of LiDAR degradation using full-scale tunnel experiments with smoke, reporting physically interpretable LiDAR indicators and a continuous data-driven degradation score. Full article

Distribution drift can substantially degrade the performance of deployed machine learning models; for example, accuracy on SST-2 can fall from $88\%$ to $58\%$. Detecting such degradation is fundamentally challenging because deployment provides inputs but not labels, so the detection itself must be reference-free. We introduce DecayBench, the first reference-free, calibrated benchmark for evaluating drift detectors. DecayBench measures detector trustworthiness along five axes (calibrated, valid, timely, no-regret, adaptive), and compares ten existing detectors across ten NLP, vision, and multimodal datasets using paired-bootstrap significance testing. Evaluation on DecayBench shows that no existing detector is uniformly optimal. Motivated by this observation, we propose Alert, a label-free aggregation rule for drift detection. Unlike all competing combiners, it uses a label-free self-configuring selection rule with a no-regret guarantee. Alert has three contributions: (i) a dilution analysis yielding a self-configuring detector selection rule; (ii) a finite-sample conformal guarantee that controls the false-alarm probability on clean data at any prescribed level (e.g., $5\%$) for arbitrary score distributions; and (iii) a no-regret result: when no single detector dominates (constituents of comparable effect size, a condition checkable offline), Alert matches or beats the best constituent, being never significantly worse and sometimes better by a large margin; this holds across NLP, NLI, and vision (ResNet), with the largest gains under multimodal drift, and the proof identifies a dominant single detector (MMD on CLIP) as the only dilution exception. We prove the no-regret property and, across the benchmark, report its empirical counterpart, non-dominance under a paired bootstrap (Alert is never significantly worse than the best constituent), which at some operating points is statistically inconclusive rather than a strict win. Because Alert combines only embedding- and logit-based detector scores, it directly transfers across NLP, vision, and multimodal models. Empirically, Alert strictly improves over single-modality monitoring, increasing AUC by up to 25 points under mixed-modality drift and by approximately 50 points under cross-modal mismatch, where individual modality-specific detectors perform near chance. Alert also matches or outperforms the Fisher, Simes, Bonferroni, and median combiners, performs best under low-severity drift, and matches or surpasses early fusion (Concat-MMD) in both multimodal settings. Full article

Availability-oriented monitoring may require low-latency anomaly inference, but deploying a lightweight model as a separate service adds process, serialization, and network-path overhead. This article evaluates the inference-path architecture used by a prototype availability-response loop; it does not evaluate the effectiveness of a new DoS/EDoS detector or claim production attack mitigation. The same IsolationForest model and telemetry vectors were executed through an external Python/FastAPI sidecar and through ONNX Runtime embedded in a Spring Boot JVM. The prototype also included a LSTM Autoencoder and a bounded rule layer producing SCALE_UP, RETRY, FALLBACK, or NONE. The main Kubernetes benchmark used Docker Desktop 4.73.1 (Docker Engine 29.4.3) with Kubernetes v1.34.3 on a single-node cluster. Embedded ONNX reduced request-response latency from 26.090 ms to 4.839 ms on average, from 62.065 ms to 5.367 ms at P95, and from 76.487 ms to 6.244 ms at P99; calculated sequential throughput increased from 38.3 to 206.7 requests/s. Additional checks covered concurrent load, gRPC transport, cold start, memory footprint, and limited Kubernetes actions. In an extended comparison using 15 runs of 950 s for each active configuration, CPU-based HPA, rules-only, and AI+rules produced overlapping aggregate SLA-violation rates; no aggregate benefit of the ML gate over rules alone was observed under the tested degradation proxies. The evidence supports a narrow architectural conclusion: in-process ONNX is a lower-latency and lower-footprint execution path for the tested lightweight model. Detection quality, superiority over rule-only or established autoscaling mechanisms, and effectiveness against real adversarial traffic remain open validation tasks. Full article

Despite developing on other fungi and forming complex ecological interactions, fungicolous fungi are among the understudied groups. This study examined the fungicolous fungi Hypomyces aurantius and its host species, Lyophyllum littoralis, in detail using morphological and molecular data. Both species are new records for the Turkish mycobiota. The macroscopic and microscopic characteristics of both species were evaluated in depth, and morphological descriptions were corroborated by molecular analyses based on the nuclear ITS region. It was observed that H. aurantius heavily colonised L. littoralis, a previously unreported host, causing significant softening and degradation of the host tissue. We investigated extracellular enzyme activities to determine the mycoparasitic activity of H. aurantius on the host fungus and its environmental survival strategies. We detected positive enzyme activities. These results suggest that H. aurantius is a potential mycoparasite. This study contributes to our understanding of fungal biodiversity in Türkiye and provides a foundation for understanding the ecology of supra-fungal fungi and fungicolous/mycoparasitic interactions. conclusions. Full article

Continuous, non-invasive fish monitoring supports aquatic animal management, biodiversity assessment, and sustainable aquaculture, but embedded deployment requires a careful balance among accuracy, speed, memory, and computation under visually degraded underwater conditions. We developed ULFD-YOLO, an ultra-lightweight detector derived from YOLOv11n through coordinated redesign of the backbone, neck, and detection head. The model combines a custom convolutional MobileNetV4-tiny backbone, a hypergraph-based multi-scale fusion neck, and a lightweight MBConv head with channel attention. Experiments were conducted on Fish-BJ, an in-house dataset of 3402 images covering 21 species-informed aquarium-fish detection categories, and on a deliberately difficult 1180-image WildFish subset after dataset-specific training. On Fish-BJ, ULFD-YOLO achieved 0.960 mAP@0.5 and 0.732 mAP@0.5:0.95 with 1.3 M parameters, 2.6 GFLOPs, and a 3.0 MB model file, reducing parameters and computation by 50.0% and 58.7% relative to YOLOv11n. Bootstrap resampling yielded 95% confidence intervals of 0.946–0.973 and 0.638–0.821 for the two metrics, respectively. The model achieved 0.803 mAP@0.5 on WildFish and 19–24 FPS at 448 × 640 on a Jetson Orin Nano under its 15 W nvpmodel power mode. These results establish a practical accuracy–efficiency trade-off for embedded fish monitoring rather than peak localization accuracy. Full article

In large-scale multiple-input multiple-output (MIMO) systems, inadequate pilots can necessitate pilot reuse, reducing channel-estimation accuracy, while too few received pilot observations can also lead to inaccurate interference-plus-noise covariance estimates. These estimation errors can further degrade the performance of downstream interference suppression and data detection. Learning-based methods have been developed for pilot assignment, channel estimation, and receiver processing, but these methods are often studied separately. This survey organizes recent studies according to where learning-based methods are applied in the signal-processing chain: pilot-domain mitigation, intelligent channel estimation with contaminated or limited pilots, and intelligent receiver processing with contaminated or limited pilots. We also classify the studies by learning method and compare them using the same set of evaluation criteria. Across the surveyed papers, performance is evaluated using different metrics. Many studies also lack evaluations under changing channel or system conditions and do not fully report implementation costs such as computational complexity, memory usage, and latency. Among the studies that satisfy our selection criteria, none directly investigates learning-based estimation of the interference-plus-noise covariance matrix for interference rejection combining (IRC) receivers when only limited pilot observations are available. Based on these findings, we propose a minimum set of benchmarking requirements and identify lightweight online adaptation, joint processing, learning-based covariance estimation for IRC receivers, and robust processing for large-array architectures as future research directions for emerging sixth-generation (6G) systems. Full article

Convolutional Neural Networks (CNNs) are fundamental tools for image analysis and recognition in monitoring systems, particularly in photovoltaic (PV) installations, where visual inspection and thermal imaging play crucial roles in anomaly detection and energy optimization. This publication presents a Hybrid Fuzzy Convolutional Neural Network (HFCNN) that integrates a fuzzy dense layer utilizing Ordered Fuzzy Numbers (OFNs) into the CNN architecture. The architecture is additionally validated on the public ELPV benchmark of 2624 electroluminescence images of photovoltaic cells, where the HFCNN with Mean of Maxima defuzzification attains classification quality statistically indistinguishable from a CNN baseline while using a four times smaller dense layer and training two to three times faster. The methodology combines the feature extraction capabilities of traditional CNNs with the uncertainty handling properties of fuzzy logic. Experiments using the MNIST dataset demonstrate that HFCNN with Mean of Maxima (MOM) defuzzification achieves comparable accuracy to standard CNNs while using significantly fewer parameters (75% reduction in the dense layer). This efficiency gain is advantageous for deployment on edge computing devices. This work constitutes a methodological contribution—establishing, for the first time, the feasibility of integrating Ordered Fuzzy Numbers into CNN architectures without requiring expert membership function design. While the current study validates this approach on MNIST, actual photovoltaic applications require dedicated future research on real PV thermal imagery. Nevertheless, the proposed HFCNN framework could potentially support practical photovoltaic energy system applications in detecting panel degradation, performance anomalies, and autonomous decision-making in large-scale PV installations. Full article

Historical art materials have always been a starting point for better understanding artists’ artwork composition, the long-term stability of used formulations and for designing appropriate strategies for conservation. It is important to consider how artists have always been involved over the centuries in the selection and preparation of the materials they used. This changed with the advent of industrial formulations, which were produced and marketed on a large scale, leading to significant changes from the 19th century onwards. In this study, the study of a set of 30 green paint tubes from the Munch Museum collection (Oslo) is presented. The selected investigation method allows for a progressive acquisition of complementary and increasingly detailed information through three multi-analytical steps, involving elemental, spectroscopic and chromatographic techniques, with the aim of characterising these materials in their various inorganic and organic components. Results achieved allow the identification of green pigments’ chemical composition and classification of them in distinct groups, presence of inert additives and characterisation of organic binding media. Degradation processes were also highlighted through the detection of metal soap formations. Full article

Conventional anaerobic digestion relies on diverse inocula present in sludge-based systems. A defined consortium approach was investigated as an alternative. Acetivibrio cellulolyticus was chosen as the cellulose degrader, and two strains of Methanosarcina barkeri were selected as methane producers. Initial cultivation following manufacturer protocols highlighted significant challenges in maintaining strict anaerobic conditions, particularly in the absence of specialized infrastructure. A simplified anaerobic cultivation workflow was therefore evaluated for pure cultures of the selected anaerobes and subsequently used to evaluate a defined consortium using microcrystalline cellulose (MCC) and industrial lignocellulosic biomass (LB) residue as growth substrates. The workflow enabled successful cultivation of pure cultures in their recommended nutrient media without detectable contamination, as assessed by microscopy, aerobic contamination checks, and gas chromatography analysis. Growth-associated observations were obtained on MCC after prolonged incubation (~30 days); however, a metabolically active cellulolytic–methanogenic consortium was not established, as methane was not detected and no activity was detected on the LB substrate. This study demonstrates that anaerobic cultivation of fastidious microorganisms is feasible using a simplified method without fully controlled anaerobic environments and highlights inherent challenges associated with the defined consortium on substrates such as MCC and complex LB residue. The simplified workflow may provide an accessible approach to anaerobic cultivation for sustainable biomethane research in laboratories lacking specialized anaerobic infrastructure. Further work is required to determine conditions supporting methane production using the defined consortium. Full article

Background/Objectives: Microplastics (MPs) and nanoplastics (NPs) have emerged as ubiquitous environmental contaminants resulting from the extensive production, use, and degradation of plastic materials. Human exposure occurs primarily through contaminated food and drinking water, with inhalation representing an additional important route. Growing concern has focused on the ability of these particles, particularly NPs, to cross biological barriers, enter the systemic circulation, and reach human tissues. The aim of this review was to summarize current evidence on dietary exposure to MPs and NPs, their gastrointestinal bioavailability and toxicokinetics, and their potential systemic health effects, with particular emphasis on organ-specific responses, underlying biological mechanisms, and the strength and limitations of the available evidence. Methods: A comprehensive narrative review of the scientific literature published between 2000 and 2026 was conducted using PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar. Original research articles and review papers addressing dietary exposure, occurrence in food and drinking water, migration from food-contact materials, gastrointestinal absorption, translocation, biodistribution, bioaccumulation, elimination, molecular mechanisms, and potential organ-specific or systemic health effects were included. Publications without full-text availability, conference proceedings, editorials, commentaries, duplicate publications, and studies without relevance to human exposure or health were excluded. Results: Food, drinking water, beverages, and food-contact materials represent important sources of human exposure to MPs and NPs. Following ingestion, most larger particles are eliminated through the gastrointestinal tract, whereas smaller MPs and particularly NPs may cross biological barriers and potentially reach the systemic circulation and distant tissues. Experimental studies consistently identify interconnected biological responses involving oxidative stress, inflammation, mitochondrial dysfunction, barrier impairment, immune dysregulation, genotoxicity, apoptosis, and endocrine disruption. These mechanisms have been associated with alterations in the gastrointestinal, respiratory, cardiovascular, nervous, urinary, reproductive, endocrine, and skeletal systems and with biological processes relevant to carcinogenesis. However, most mechanistic evidence derives from in vitro and animal models, whereas human evidence remains limited and predominantly observational. Consequently, the extent to which these experimental findings translate into clinically significant effects in humans remains uncertain. Conclusions: Current evidence supports the biological plausibility of systemic effects associated with MNP exposure but is insufficient to establish causal relationships between chronic dietary exposure and specific human diseases. The detection of MNPs in human tissues and reported associations with pathological conditions should therefore be interpreted cautiously. Standardized analytical methods, improved characterization of realistic human exposure, and well-designed longitudinal epidemiological studies integrating quantitative exposure assessment with validated clinical outcomes are required to clarify dose–response relationships, long-term health effects, and the clinical significance of MNP exposure. Full article

Real-time detection of chemical warfare agents (CWAs) and toxic industrial chemicals underpins military protection, counter-terrorism, and emergency response. Yet field instruments usually fail for a reason unrelated to sensitivity: they cannot identify agents that are not already in their reference libraries, such as novel analogs, mixtures, and degradation products. We argue that this unknown-agent problem is a structural limitation of single-modality sensing, because any one class of information (molecular bonds, ion mobility, elemental composition, or chemical reactivity) is rarely sufficient to resolve an unfamiliar threat. We review the dominant field modalities, including FTIR, Raman/SERS, ion mobility and field-asymmetric ion mobility spectrometry, laser- and spark-induced plasma spectroscopy, metal-oxide sensor arrays, and portable mass spectrometry, and show that their weaknesses are largely complementary. We then set out the principle of orthogonal multimodal sensing, in which complementary information axes are combined by machine learning with anomaly and open-set detection so that unfamiliar agents are recognized as such rather than misidentified. Four hybrid architectures are critically compared, and we examine spark-induced decomposition diagnostics, consumable-free self-decontaminating field systems with edge AI, and the open challenges of standardized datasets, calibration transfer, and validation, before outlining a roadmap toward field-relevant recognition of unidentified chemical threats. Full article

Retrieval-augmented fine-tuning (RAFT) can improve small language models (sLMs) on retrieval-grounded question answering, but the synthetic training data produced by commercial large language models (LLMs) vary in quality. This paper contributes a quality-aware selection protocol—rather than a new RAFT or QLoRA method—that scores LLM-generated alternatives along four embedding-based dimensions (question relevance, answer faithfulness, QA coherence, and semantic similarity) and selects one alternative per task before parameter-efficient fine-tuning. Under pre-specified paired-bootstrap contrasts with Holm correction, the parameter-free faithfulness-based selector only-AF significantly exceeds random selection on Gemma-2-9B-IT (ΔF1 = +0.106, 95% CI [+0.043, +0.174], Holm-corrected p = 0.019), and its pre-specified weighted companion af-70 (w_AF = 0.70) shows the same confirmed pattern (Holm-corrected p = 0.002). Both effects persist under a Korean character-level F1 that removes particles and punctuation (Holm-corrected p = 0.004 and p = 0.042), indicating robustness to the choice of lexical metric. Relative to training on the full 150-row augmented pool, the quality-selected 50-row sets are statistically indistinguishable while using one third of the training data, which we interpret as data efficiency rather than superiority. Across six instruction-tuned models (2B–27B), a significant selector-by-model interaction indicates that the optimal quality axis is model-dependent, and the two smallest models show no benefit from selection. The study’s confirmatory contrasts use a small controlled Korean corpus under a transductive design; two pre-registered validation experiments probe external validity. On an independent five-fold larger corpus with a passage-level train/test split, fine-tuning transfers strongly and the selected one-third subsets show no significant difference from the full pool, while the advantage over random selection is directionally positive but small and not significant; under controlled corruption of 35% of the pool, the metrics detect the damaged rows, and for the score-sum selector the selection-versus-random benefit is significantly larger than on the clean pool (difference-in-differences p = 0.0014; directionally consistent but not significant for the faithfulness selectors). Within this scope, quality-aware selection is a promising, data-efficient safeguard for synthetic RAFT data—performing comparably to full-pool training at one third of the cost, with growing value as pool quality degrades—and larger-scale external validation remains future work. Full article

Manual fruit thinning is labor-intensive and inefficient, making the development of intelligent visual detection systems a crucial approach for improving the quality and production efficiency of the pear industry. However, in natural orchard environments, young pear fruits are small in size with slender fruit stalks, and their texture and color characteristics are highly similar to those of tender branches. Furthermore, variations in illumination and occlusions caused by branches and leaves make it difficult for existing detection models to simultaneously and accurately identify fruits and fruit stalks, limiting their application in automated thinning equipment. In this study, Yuluxiang pear was selected as the research object, and image data were collected under diverse field conditions, including forward-lighting, backlighting, close-range shooting, long-range shooting and fruit overlapping. A dedicated dataset containing 3057 images was established. Based on the RT-DETR-R18 network, a lightweight and high-precision fruit–stalk synchronous detection model was proposed. Specifically, the backbone network was reconstructed by integrating GCConv with C2f modules to enhance global feature extraction for slender fruit stalks. The bottleneck structure was optimized using GCConvC3 to reduce feature degradation under occlusion conditions, and an additional 4× down-sampling P2 detection head was introduced to improve the detection capability for small targets. To fully validate the model performance and stability, three types of experiments were conducted in this study: ablation experiments, repeated experiments with different random seeds, and comparative experiments. Ablation experiments verified the cumulative performance improvements brought by the introduced modules. Repeated experiments with different random seeds were performed to explore training randomness-induced performance fluctuations, and the results demonstrated that the proposed model maintains stable overall detection accuracy with minor metric fluctuations. Comparative experiments demonstrated that the proposed model achieved a compact parameter size of only 15.97 M, with a precision of 95.0% for young pear fruit detection and an mAP50 of 83.0% for fruit stalk detection, outperforming all comparative models in overall mAP50. The training convergence curves and Grad-CAM++ visualization results further confirmed the stable optimization process and enhanced feature attention capability of the proposed model. By achieving a favorable balance between detection accuracy and model lightweightness, this approach provides effective technical support for the development of intelligent fruit thinning equipment and vision-based systems for smart pear orchards. Full article