Search Results (21,389)

Background/Objectives: Colonoscopy withdrawal time is an established quality indicator, but ongoing debate exists regarding whether its relationship with adenoma detection follows a fixed threshold or a continuous dose–response pattern. Methods: We retrospectively analyzed 31,780 colonoscopies from a multicenter endoscopy database to evaluate the relationship between withdrawal duration and lesion detection. Withdrawal time was assessed as both a continuous and categorical variable, and adenoma detection rate (ADR) and polyp detection rate (PDR) were examined across quartiles and clinically relevant intervals. Results: ADR increased progressively from 7.72% in the shortest withdrawal-time quartile to 36.47% in the longest quartile, while PDR increased from 21.22% to 67.86% (both p for trend <0.001). In the multivariable analysis adjusted for age, sex, and bowel-preparation quality, each additional minute of withdrawal time was independently associated with higher odds of adenoma detection (adjusted OR 1.14, 95% CI 1.13–1.15; p < 0.001). Although the confidence interval was narrow, reflecting the large sample size, the effect was clinically appreciable: model-predicted adenoma detection increased from 13.6% at six minutes to 21.0% at eight minutes and 28.2% at ten minutes, corresponding to approximately seven and fifteen additional adenoma-positive examinations per hundred procedures, respectively. The association remained consistent across age and sex strata. Receiver operating characteristic analysis showed moderate discrimination (area under the curve 0.701), with a Youden-optimal region of approximately 7 to 8 min. Because recorded withdrawal time incorporates interventional time, the principal analysis of inspection effort was conducted at the level of the endoscopist, using withdrawal time measured exclusively in the 14,611 examinations in which no polyp was detected and no tissue was sampled. Among 107 endoscopists contributing 26,793 procedures, inspection time was associated with adenoma detection (Spearman ρ = 0.46; p < 0.001), corresponding to an absolute increase of approximately 1.5 percentage points per additional minute, with attenuation of the gradient beyond approximately seven minutes. Conclusions: The procedure-level findings above describe the association with recorded withdrawal duration as captured in routine practice and are reported as supportive rather than as estimates of inspection effort. Together these analyses indicate that inspection time is associated with adenoma detection in a graded manner extending beyond the historical 6 min benchmark. Full article

This study examined the microbial quality of air and surfaces in six elevators at a tertiary hospital in Greece in 2022. The results highlighted hospital elevators as potentially relevant environmental reservoirs of microbial contamination, with contact surfaces (buttons) exhibiting significantly higher microbial loads than the air. Strong seasonality was also documented; the elevated microbial burden observed during the October sampling period likely reflects seasonal fluctuations in occupant traffic and environmental factors, although direct traffic and microclimatic variables were not monitored. Presumptive Staphylococcus spp. strains dominated the air, while Gram-negative bacteria recovered on MacConkey agar prevailed on surfaces. Phenotypically resistant isolates were also detected, including isolates showing reduced susceptibility or resistance to clinically relevant antimicrobial agents, highlighting the potential value of incorporating AMR surveillance into environmental monitoring. The semi-quantitative environmental exposure assessment showed that, while air quality remained within acceptable limits, the surface-contact route via elevator buttons represented the largest estimated exposure component. Healthcare workers showed the highest estimated cumulative exposure because of the greater frequency of elevator use assumed in the exposure scenario. In conclusion, hospital elevators may act as environmental reservoirs of microbial contamination and potential exposure pathways within the hospital environment, highlighting the need for enhanced hygiene measures, systematic monitoring and the integration of antimicrobial resistance into exposure assessment. Full article

Accurate detection of tomato flowering stages is important for greenhouse crop management and automated pollination, but it remains challenging because tomato flowers are small, densely distributed, frequently occluded, and visually similar across adjacent developmental stages. To address these problems, this study proposes YOLO11-FAL, an improved object detection model based on YOLO11 for tomato flowering stage detection in greenhouse environments. The original C3k2 modules are replaced with C3k2_Faster to reduce redundant spatial computation and enhance local structural feature representation. An Attentional Scale Sequence Fusion (ASF) structure is introduced into the neck network to strengthen multi-scale feature interaction, and a Localization Quality Estimation Head (LQEHead) is incorporated to recalibrate classification confidence using bounding-box distribution information. Experiments were conducted on a self-constructed tomato flower dataset containing Bud, Anthesis, and Post-anthesis stages under varying illumination conditions. YOLO11-FAL achieved 92.00% Precision, 88.88% mAP@0.5, 62.21% mAP@0.75, and 56.72% mAP@0.5:0.95. The model contained 2.384 M parameters and achieved 49.58 FPS on an NVIDIA Jetson AGX Orin (NVIDIA Corporation, Santa Clara, CA, USA) using TensorRT FP16, with a measured onboard power consumption of 11.07 W. These results indicate that YOLO11-FAL provides a practical and efficient visual perception approach for greenhouse tomato flowering stage detection. Full article

Small-scale and urban indoor farms typically rely on manual observation, which delays stress detection and yields inconsistent crop quality. While large language models (LLMs) and retrieval-augmented generation (RAG) have been explored for agricultural advisory systems, their integration into a single cloud-free indoor-farming platform that couples multimodal symptom interpretation with autonomous environmental control remains largely unexamined. This study presents an integrated platform built around an agentic AI advisory pipeline that runs entirely on-device on commodity hardware. The pipeline couples a RAG-grounded Mistral 7B language model with a LLaVA 7B vision-language model through condition-based routing, intent classification, multi-step reasoning, and an LLM validation gate, delivering context-aware text and image-based symptom-level guidance from a conversational interface. The advisory layer operates alongside vision-based plant monitoring and a deliberately isolated threshold-based control layer, in which an ESP32 microcontroller autonomously actuates irrigation and lighting against predefined thresholds while a Raspberry Pi 5 performs continuous plant detection and browning monitoring. On Cos lettuce, the advisory pipeline achieved 82.00% weighted accuracy across 50 queries spanning health, symptom, watering, pest, root-health, and growth-stage categories, scored against established plant pathology and postharvest references, with no incorrect responses recorded. The study contributes the design of an agentic advisory pipeline and its integration into a working, cloud-free indoor-farming platform, providing an on-device foundation for intelligent small-scale farming. Full article

Riparian land use and land cover (LULC) exerts scale- and season-dependent controls on surface water quality, yet its influence in regulated agricultural–urban rivers is poorly characterized. We combined seasonal t-tests, one-way ANOVA, and redundancy analysis (RDA) at three riparian buffer scales (500, 1000, and 2000 m) to examine discharge, dissolved oxygen (DO), temperature, turbidity, and pH at four USGS stations along the Kansas River mainstem (2019–2026). DO and temperature showed the strongest seasonal contrasts: DO was 3.1–3.7 mg/L higher in the dry season and temperature 13–15 °C higher in the wet season, a coupling central to aquatic habitat suitability. Turbidity rose significantly in the wet season, consistent with agricultural runoff and sediment mobilization, whereas discharge showed no significant seasonal difference at three of four stations, reflecting upstream reservoir regulation. Spatial ANOVA detected station-level differences only for wet-season DO ($F_{3,28}=4.91$, $p=0.007$), which was lowest at the downstream urbanized station. RDA linked agricultural cover to turbidity and urban cover to reduced wet-season DO, although permutation tests were non-significant ($p\ge 0.42$) at $n=4$ replicates. Seasonality and riparian LULC jointly shape water quality along this regulated river, and the 500 m buffer is the most spatially discriminating scale for land-cover assessment. Full article

Smartphone GNSS metadata remains an underexplored source for evaluating navigation-signal integrity in real-world conditions. We investigated GPS behavior recorded by a Samsung Galaxy A72 smartphone across two European flights (EXP03: Rome–Bucharest, n = 377; EXP10: Bucharest–Lisbon, n = 78) and 275 terrestrial reference photographs (Cabo da Roca, Portugal). A total of 730 photographs were analyzed using a seven-indicator taxonomy, conceptually inspired by Receiver Autonomous Integrity Monitoring (RAIM) principles, covering anti-spoofing, anti-sniffing, and anti-tampering checks. GPS capture rates reached 100% (Timestamp Camera) and 83.3% (native camera) up to 11,439 m WGS84, among the highest EXIF altitude profiles reported to date. Velocity spikes (1560–1875 km/h), one at cruise altitude and one during landing, were indistinguishable from GPS spoofing at the EXIF level, and their physical origin is undetermined. Phantom geolocation was absent in flight (0/442) versus 37.3% on the ground, consistent with GPS constellation visibility as the primary factor. In total, 88% (n = 920) of JPEG files lacked the standard EOI marker, generating false positives in integrity validators. Findings are device-specific, derived from non-independent observations, and require replication before generalizing to other GNSS receivers or latitudes. The framework offers a methodological basis for EXIF integrity analysis relevant to EU AI Act Article 10(3) data quality and UAV precision-agriculture georeferencing. Full article

Background: Breast cancer is the most common malignancy among women worldwide and a leading cause of cancer-related mortality. Despite advances in diagnosis and treatment, accurately predicting 5-year survival remains challenging because of disease heterogeneity. This study systematically evaluated the performance of machine learning (ML) models for predicting 5-year breast cancer survival and synthesised their overall discriminative performance. Methods: A systematic search of PubMed, Scopus, and Web of Science (2010–2024) identified studies developing ML models for 5-year survival prediction. Study quality was assessed using PROBAST. Logit-transformed area under the receiver operating characteristic curve (AUC) estimates were quantitatively synthesised using a random-effects meta-analysis with the restricted maximum likelihood (REML) estimator. Heterogeneity was assessed using Cochran’s Q and I² statistics, publication bias using funnel plots and Egger’s regression test, and subgroup analyses according to study characteristics. Statistical analyses were performed using IBM SPSS version 29.0 and R version 4.6.1. Results: Fifteen studies were included. A wide range of ML models, including Random Forest, Random Survival Forest, and gradient boosting methods, was evaluated. The pooled analysis demonstrated good discriminative performance, with an overall AUC of 0.83 (95% CI: 0.80–0.86). Substantial heterogeneity was observed across studies. Ensemble-based models generally showed consistent performance. Publication bias was detected, and several studies exhibited moderate-to-high risk of bias. Conclusions: ML models show strong potential for predicting 5-year breast cancer survival and may support early risk stratification and clinical decision-making. However, greater methodological standardisation, rigorous external validation, and improved reporting are required before widespread clinical implementation. 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

Electromyography (EMG)-based sensing is an important tool for assessing neuromuscular activity and monitoring athlete development; its reliability depends on electrode placement, signal quality, and accurate identification of muscle activation periods. This study proposes an intelligent EMG sensing framework integrating preliminary electrode placement assessment, muscle activity detection, feature extraction, and regression-based progression prediction. A placement assessment indicated that positioning the electrode adjacent to the innervation zone produced the highest RMS under the tested conditions. A two-stage activity detection method based on clustering and probabilistic modeling achieved an average error of 1.5% and a temporal deviation of 19 ms. Nine time-domain EMG features extracted from the detected activity segments were used to characterize athlete progression and estimate the time required to reach a reference neuromuscular profile. Among the methods, Linear Regression provided the best fit to the data, obtaining R² = 0.987 and RMSE = 4.21 and suggesting a predominantly linear relationship between the EMG-derived features and training duration within the dataset. However, these results were obtained from only six longitudinal observation periods for a single representative athlete, with each period represented by a 90-dimensional EMG feature vector derived from the ten movement classes. Therefore, the results should be interpreted as preliminary, athlete-specific goodness-of-fit findings rather than evidence of generalizable predictive performance. Validation using larger longitudinal cohorts and independent datasets is required. The proposed framework is compatible with future IoT-enabled wearable and edge-computing architectures; however, hardware-level implementation was beyond the scope of this study. Full article

Shear-wave time-of-flight (TOF) measurement enables controlled assessment of frequency-dependent wave propagation, but its feasibility in biological tissues remains insufficiently established. This study evaluated whether a custom shear-wave TOF device could detect frequency- and direction-dependent responses in ex vivo tissues. Three porcine liver samples and three chicken breast samples were examined. Chicken breast was measured with propagation parallel and perpendicular to visible muscle fibers. One-cycle sinusoidal excitations were applied at 40–160 Hz, with 50 acquisitions ensemble-averaged per sample–frequency measurement. TOF was estimated using Tx threshold detection and cumulative-energy-based Rx onset detection, and TOF-derived apparent shear-wave propagation speed was calculated from the Tx–Rx distance and the measured TOF. Frequency-dependent data were fitted using the Kelvin–Voigt fractional derivative model to obtain model-dependent KVFD fit parameters. Signal quality was assessed, and a preliminary descriptive comparison with HISKY EQTouch UD3000 (Wuxi Hisky Medical Technologies Co., Ltd., Wuxi, China) SWE was performed. All 63 averaged sample–frequency measurements satisfied the predefined primary-detection criteria. Mean apparent shear-wave speed was 3.145 m/s in porcine liver, 6.133 m/s in chicken breast measured parallel to the fibers, and 5.914 m/s in chicken breast measured perpendicular to the fibers, giving a parallel-to-perpendicular speed ratio of 1.037. Mean post-averaging, post-processing SNR ranged from 24.47 to 31.52 dB. The UD3000 comparison showed the same tissue ranking. The device detected frequency- and direction-dependent responses in averaged ex vivo signals, supporting its feasibility as a controlled research platform. Claims of absolute stiffness accuracy and intrinsic muscle anisotropy require independent calibration and validation. Full article

The objective of this study was to evaluate the effects of partial loss of function (LOF) of MSTN on carcass composition and pork quality and determine whether the phenotype was influenced by the IGF2 genotype (IGF2-G3072A). Pigs (N = 44) segregating for the paternally expressed IGF2 G or A alleles were either wild-type (WT) or heterozygous (HET) for a targeted MSTN LOF allele and were harvested at 175 ± 5 days of age. Carcass composition and loin and belly quality traits were analyzed using linear models. Few significant MSTN × IGF2 genotype interactions were detected, indicating that the heterozygous MSTN phenotype was largely independent of IGF2 status. Relative to WT contemporaries, MSTN HET pigs exhibited greater dressing percentage and loin muscle area (p < 0.01) without differences in hot carcass weight or backfat depth, resulting in increased fat-free lean percentage. However, loins from HET pigs were approximately 5 L* units lighter in color (p < 0.01) and exhibited greater drip loss. Marbling scores and extractable lipid content were reduced in both loin and belly, and belly firmness decreased. The IGF2 genotype had similar muscle-deposition effects. These results indicate that partial MSTN LOF improves lean yield but introduces tradeoffs in pork quality, largely independent of IGF2 genotype. Full article

Earth Observation (EO) analysis increasingly relies on large and heterogeneous satellite datasets, yet developing EO workflows often requires specialized expertise in data selection, geospatial programming, and cloud-based processing. Recent advances in Large Language Models (LLMs) offer new opportunities for natural-language interaction with EO systems, although challenges related to transparency, reproducibility, and domain-specific reasoning remain. This study presents SeaScope, an explainable AI framework that integrates LLMs, Retrieval-Augmented Generation (RAG), scientific knowledge retrieval, and Google Earth Engine (GEE) to transform natural-language requests into transparent and executable EO workflows. The framework combines knowledge retrieval, code generation, cloud execution, provenance tracking, and interactive visualization within a unified environment. A pilot implementation is demonstrated through maritime and coastal monitoring applications, including oil spill detection, vessel monitoring, water quality assessment, floating debris detection, and air quality analysis. Multiple state-of-the-art LLMs are evaluated under both RAG and non-RAG configurations using representative EO case studies. The results indicate substantial differences among model families and show that retrieval augmentation can significantly improve workflow generation quality and reliability for capable models, while providing more limited benefits for smaller models. The proposed framework demonstrates the potential of explainable AI agents to support transparent, reproducible, and scalable EO analysis. Full article

Szechuan pepper pungency is traditionally assessed by subjective sensory panels or laboratory instruments, hindering on-site detection. Perceived numbing sensation does not always correspond directly to alkylamide content because multiple electroactive constituents, including polyphenols and sanshools, may contribute to the overall sensory response. Chromatographic methods quantify individual compounds but may not fully reflect integrated human pungency perception. Electrochemical detection bypasses separation, as the voltammetric response integrates oxidative signals from multiple electroactive species. We developed a portable electrochemical system with a custom programmable-gain potentiostat, three-electrode detector, and STM32-controlled software for differential pulse voltammetry (DPV) measurement. Coupled with machine learning, it assessed Zanthoxylum bungeanum (red peppers) and Zanthoxylum schinifolium (green peppers). Fifteen replicate scans from each of 22 origins yielded 330 DPV curves calibrated against general Labeled Magnitude Scale (gLMS) scores from a trained panel. An artificial neural network (ANN) achieved R² = 0.937 for red peppers, while principal component analysis–support vector regression (PCA–SVR) achieved R² = 0.860 for green peppers. Competitive adaptive reweighted sampling (CARS) identified three characteristic potential intervals for each type: 0.17–0.21, 0.57–0.62, and 0.69–0.80 V for red peppers, and 0.24–0.27, 0.56–0.68, and 0.69–0.77 V for green peppers. These intervals indicate that pungency-related electrochemical information is distributed across multiple potential regions. The system shows potential for rapid and objective quality assessment of Szechuan pepper. 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