Search Results (2,004)

Background/Objectives: Large language models (LLMs) show promise for patient education, yet their safety and efficacy for chronic diseases requiring lifelong management remain unclear. This study presents the first comprehensive comparative evaluation of three leading LLMs for celiac disease patient education. Methods: We conducted a cross-sectional evaluation comparing ChatGPT-4, Claude 3.7, and Gemini 2.0 using six blinded clinical specialists (four gastroenterologists and two dietitians). Twenty questions spanning four domains (general understanding, symptoms/diagnosis, diet/nutrition, lifestyle management) were evaluated for scientific accuracy, clarity (5-point Likert scales), misinformation presence, and readability using validated computational metrics (Flesch Reading Ease, Flesch-Kincaid Grade Level, SMOG index). Results: Gemini 2.0 demonstrated superior performance across multiple dimensions. Gemini 2.0 achieved the highest scientific accuracy ratings (median 4.5 [IQR: 4.5–5.0] vs. 4.0 [IQR: 4.0–4.5] for both competitors, p = 0.015) and clarity scores (median 5.0 [IQR: 4.5–5.0] vs. 4.0 [IQR: 4.0–4.5], p = 0.011). While Gemini 2.0 showed numerically lower misinformation rates (13.3% vs. 23.3% for ChatGPT–4 and 24.2% for Claude 3.7), differences were not statistically significant (p = 0.778). Gemini 2.0 achieved significantly superior readability, requiring approximately 2–3 fewer years of education for comprehension (median Flesch-Kincaid Grade Level 9.8 [IQR: 8.8–10.3] vs. 12.5 for both competitors, p < 0.001). However, all models exceeded recommended 6th–8th grade health literacy targets. Conclusions: While Gemini 2.0 demonstrated statistically significant advantages in accuracy, clarity, and readability, misinformation rates of 13.3–24.2% across all models represent concerning risk levels for direct patient applications. AI offers valuable educational support but requires healthcare provider supervision until misinformation rates improve. Full article

Reverse transcription quantitative PCR (RT-qPCR) is extensively used to quantify gene expression under drought conditions; however, its reliability depends on the validation of the reference genes under specific conditions. In cucumber, reference genes have rarely been validated under drought conditions. This study identified stable housekeeping genes for RT-qPCR normalization in the leaves of two inbred lines with contrasting drought responses. Plants underwent a 7-day drought period, with leaf samples collected at multiple points along with watered controls. The expression stability of 13 candidate genes was evaluated using four algorithms: geNorm, NormFinder, BestKeeper, and the comparative ΔCt method, with the results integrated using RefFinder. Ten genes producing specific and efficient amplicons were analyzed for stability. CACS and UBI-1 consistently ranked among the most stable genes, with TIP41-like as an additional reliable option, whereas GAPDH and HEL were unstable. GeNorm pairwise variation analysis showed that the two reference genes were sufficient for accurate normalization. Functional validation with three drought-responsive targets (LOX, HsfC1, and CYP72A219) and comparison with RNA sequencing (RNA-seq) fold changes confirmed that normalization using CACS and UBI-1 yielded the most biologically credible expression profiles. These reference genes will facilitate robust RT-qPCR analyses of drought response in cucumber leaves and provide a starting point for validating suitable normalizers in other cucumber organs and related cucurbits. Full article

Modern target tracking systems rely on radar as a sensor to detect targets and generate raw track points. These raw track points are affected by the radar’s own noise and the asymmetric non-Gaussian noise resulting from the nonlinear transformation from polar coordinates to Cartesian coordinates. Without effective processing, such data cannot directly support highly reliable situational awareness, early warning decisions, or weapon guidance. Track filtering, as a core component of target tracking, plays an irreplaceable foundational role in achieving real-time, accurate, and stable estimation of moving target states. Traditional deep learning filtering algorithms struggle with capturing long-term dependencies in high-dimensional spaces, often exhibiting high computational complexity, slow response to transient signals, and compromised noise suppression due to their inherent architectural asymmetries. In order to address these issues and balance the model’s high accuracy, strong real-time performance, and robustness, a new trajectory filtering algorithm based on a temporal convolutional network (TCN), Residual Gated Recurrent Unit (ResGRU), and multi-head attention (MHA) is proposed. The TCN-ResGRU-MHA hybrid structure we propose combines the parallel processing advantages and detail-capturing ability of a TCN with the residual learning capability of a ResGRU, and introduces the MHA mechanism to achieve adaptive weighting of high-dimensional features. Using the root mean square error (RMSE) and Euclidean distance to evaluate the model effect, the experimental results show that the RMSE of TCN-ResGRU-MHA is 27.4621 (m) lower than CNN-GRU, which is an improvement of 15.99% in the complex scene of high latitude, and the distance is 37.906 (m) lower than CNN-GRU, which is an improvement of 18.65%. These results demonstrate its effectiveness in filtering and tracking tasks in high-latitude complex scenarios. Full article

Focusing on the West African Craton (WAC) as a test bed, this protocol-oriented scoping review synthesizes indicators for orogenic gold and translates them into an auditable, map-first checklist that separates Fertility and Preservation, while deliberately deferring any performance estimation to a blinded, out-of-sample evaluation. There is a need for a transparent, auditable, and field-ready framework that integrates geological, structural, geophysical, and geochemical evidence. We (i) synthesize the state of knowledge into a map-first, reproducible targeting checklist, (ii) formalize an indicator decision matrix that separates Fertility from Preservation factors, and (iii) specify a deferred, out-of-sample evaluation protocol to quantify performance. We conduct a Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR)-style scoping review (2010–2025) and codify commonly used indicators (e.g., transpressional jogs, lineament density, proximity to tonalite-trondhjemite-granodiorite (TTG)/tonalite contacts, Sr/Y proxies). Indicators are operationalized as auditable pass/fail rules and assembled into a decision matrix with explicit uncertainty handling and risk logging. We further define a deferred evaluation protocol using classification and ranking metrics (receiver operating characteristic (ROC) and precision–recall (PR) curves, odds ratios), ablation/sensitivity tests, and district-level threshold calibration. We deliver (1) a unified, auditable checklist with default (tunable) thresholds; (2) an indicator decision matrix that disentangles Fertility vs. Preservation signals; and (3) a deferred evaluation protocol enabling a reproducible, out-of-sample assessment without inflating apparent performance. All numerical thresholds reported here are explicit placeholders that facilitate transparency and auditability; they are not optimized. A properly blocked train/validation/test scheme, operating-point selection criteria, null models, and uncertainty procedures are prespecified for future evaluation. By publishing the checklist, data lineage, and audit-log schema now—without performance claims—we enable reproducible adoption and stress-test the framework ahead of calibration. Full article

This research presents a solution to the problem of controlling the energy demand and carbon footprint of old buildings, with the focus being on a (heated) building located in Madrid, Spain. A framework that incorporates AI and advanced hybrid ensemble approaches to make very accurate energy consumption predictions was developed and tested using the MATLAB environment. At first, the study evaluated six individual AI models (ANN, RF, XGBoost, RBF, Autoencoder, and Decision Tree) using a dataset of 100 points that were collected from the building’s sensors. Their performance was evaluated with high-quality data, which were ensured to be free of missing values or outliers, and they were prepared using L1/L2 normalization to guarantee optimal model performance. Later, higher accuracy was achieved through combining the models by means of hybrid ensemble techniques (voting, stacking, and blending). The main contribution is the application of a Long Short-Term Memory (LSTM) model for predicting the energy consumption of the building and, very importantly, its carbon footprint over a 30-year period until 2050. Additionally, the proposed methodology provides a structured pathway for existing buildings to progress toward nearly Zero-Energy Building (nZEB) performance by enabling more effective control of their energy demand and operational emissions. The comprehensive assessment of predictive models definitively concludes that the blended ensemble method is the most powerful and accurate forecasting tool, achieving 97% accuracy. A scenario where building heating energy use jumps to 135 by 2050 (a 35% increase above 2020 levels) represents an alarming complete failure to achieve energy efficiency and decarbonization goals, which would fundamentally jeopardize climate targets, energy security, and consumer expenditure. Full article

Small object localization is one of the most challenging tasks owing to the poor visual appearance and noisy representation caused by the intrinsic structure of small targets. Recent advances in localizing small objects are mainly dependent on regression-based counting approaches, which require considerable annotations for training. As a contrast, human learners can quickly master labeling skills from only a few annotation examples. In this paper, we attempt to simulate this training mechanism and propose a novel positive-unlabeled (PU) learning based approach that can localize small objects by learning from partial point annotations. We evaluate our approach on five typical datasets of small objects involving a single cell, an animal/insect, and human crowds. Quantitative experimental results show that our approach has achieved inspiring localization performance (F1 score > 0.75) even under the supervision of less than 10% of the overall point annotations. This approach paves the way for low-annotation-cost single-cell analysis within microfluidic droplets. Full article

Maximum power point tracking (MPPT) under partial shading is a nonconvex, rapidly varying control problem that challenges multi-agent policies deployed on photovoltaic modules. We present Fuzzy–MAT3D, a fuzzy-augmented multi-agent TD3 (Twin-Delayed Deep Deterministic Policy Gradient) controller trained under centralized training/decentralized execution (CTDE). On the theory side, we prove that differentiable fuzzy partitions of unity endow the actor–critic maps with global Lipschitz regularity, reduce temporal-difference target variance, enlarge the input-to-state stability (ISS) margin, and yield a global $L_{\infty }$$\gamma$-contraction of fixed-policy evaluation (hence, non-expansive with $\kappa =\gamma <1$). We further state a two-time-scale convergence theorem for CTDE-TD3 with fuzzy features; a PL/last-layer-linear corollary implies point convergence and uniqueness of critics. We bound the projected Bellman residual with the correct contraction factor (for $L^{\infty }$ and $L^2\left(\rho \right)$ under measure invariance) and quantified the negative bias induced by $min\{Q_1,Q_2\}$; an N-agent extension is provided. Empirically, a balanced common-random-numbers design across seven scenarios and 20 seeds, analyzed by ANOVA and CRN-paired tests, shows that Fuzzy–MAT3D attains the highest mean MPPT efficiency (92.0% ± 4.0%), outperforming MAT3D and Multi-Agent Deep Deterministic Policy Gradient controller (MADDPG). Overall, fuzzy regularization yields higher efficiency, suppresses steady-state oscillations, and stabilizes learning dynamics, supporting the use of structured, physics-compatible features in multi-agent MPPT controllers. At the level of PV plants, such gains under partial shading translate into higher effective capacity factors and smoother renewable generation without additional hardware. Full article

Right-turning vehicles at intersections permitting right turn on red (RTOR) frequently conflict with pedestrians, posing significant safety risks. Existing studies often simplify vehicle trajectories by treating vehicles as centroid points, ignoring the spatial disparities between pedestrians and vehicles. To address this gap, we propose a conflict risk assessment framework based on front and rear wheel trajectories (FRWTs), which accounts for the dynamic differences in vehicle segments during turns. First, we partition vehicles into four segments (inner/outer and front/rear wheels) and develop a trajectory prediction model to quantify risk variations across these segments. Our analysis reveals that the inner front wheel poses the highest collision risk due to its speed, trajectory curvature, and pedestrian proximity. Next, we introduce three conflict interaction modes—hard interaction, no interaction, and soft interaction—and evaluate the applicability of conflict indicators (e.g., Time to Collision (TTC) and Post-Encroachment Time (PET)) under each mode. Using a Support Vector Machine (SVM) classification algorithm, we classify risk severity with high accuracy: 96% for hard interaction, 96% for no interaction, and 97% for soft interaction modes when TTC-PET dual indicators are employed. Our findings demonstrate that FRWT-based modeling significantly improves conflict risk assessment accuracy compared to centroid-point approaches. This work provides actionable insights for proactive traffic safety management and supports the development of targeted conflict mitigation strategies at RTOR intersections. Full article

Serotonin receptors, in particular 5-HT_1A and 5-HT_2A receptors, are important molecular targets for the central nervous system (CNS) disorders, such as schizophrenia, depression, anxiety disorders, memory deficits, and many others. Here, we present structural and pharmacological evaluation of a serotonin receptor ligand, SERAAK2, identified in a structure-based virtual screening campaign. Molecular docking studies revealed that SERAAK2 binds with its molecular targets via Asp3.32 as the main anchoring point, which is typical for orthosteric ligands of aminergic GPCRs. Molecular dynamics simulations confirmed the stability of the ligand binding poses in the studied receptors. MMGBSA calculations were in accordance with the receptor in vitro binding affinity studies, which indicated that SERAAK2 is a potent ligand of 5-HT_1A and 5-HT_2A receptors. It was also found that SERAAK2 displays favorable ADMET parameters. The demonstrated anxiolytic- and antidepressant-like effects of SERAAK2 in animal models, which may involve its interaction with 5-HT_1A receptors, warrant further studies to confirm these activities and elucidate the underlying mechanisms. Full article

Long-term monitoring of tree crown condition is essential for assessing forest resilience under increasing climatic variability. This study presents a comprehensive evaluation of oak (Quercus spp.) defoliation trends in Poland from 2015 to 2024, based on national forest health monitoring data. Mean defoliation remained relatively stable until 2018, followed by a significant increase in 2019 (+5.1 percentage points; p < 0.001), coinciding with a major drought event across Central Europe. In subsequent years, defoliation gradually decreased and stabilised, indicating partial canopy recovery. Segmented regression and spline models revealed a consistent breakpoint in 2019 across all age classes, with the most severe crown damage recorded in stands older than 100 years. Younger stands showed lower defoliation levels and higher regenerative capacity. A nonlinear relationship between defoliation and growing-season precipitation was also identified, showing that when rainfall fell below 40 mm, canopy loss exceeded 30%. The results confirm that oak defoliation reflects both short-term climatic stress and long-term structural changes. Integrating monitoring data with climatic analyses and statistical modelling improves the detection of stress-related drivers and the assessment of recovery processes. The combined use of these approaches supports adaptive forest management strategies, including the promotion of mixed-species and multi-aged stands, improvement of soil nutrient conditions, and targeted monitoring of drought-sensitive age classes, thereby enhancing the resilience of oak ecosystems to climate change. Full article

This study evaluates Limassol’s readiness to adopt the 15-minute city model through a spatial accessibility and urban-form analysis. Using openly available geo-referenced Points of Interest (POIs), road network data, land-use records, and census information, we generated 15-minute walking and cycling isochrones for eight essential urban functions: Education, Food, Green Areas, Health, Services, Shopping, Tourism, and Transport. Residential coverage within each isochrone was calculated to assess accessibility equity across the city. Urban sprawl was quantified using size, density, and fragmentation metrics derived from historical planning zones. Results show that while cycling accessibility is high for most categories (85–95% of residential areas), walking accessibility is considerably lower and unevenly distributed, with several critical functions, particularly Green Areas, Education, and Transport, serving less than half of the residential zones. The analysis also reveals increasing spatial fragmentation and outward population shifts consistent with low-density sprawl, driven by planning policies and development pressures. These findings indicate that Limassol is only partially aligned with the principles of the 15-minute city, with significant gaps in walkable access and decentralized service provision. The study concludes that targeted planning reforms, improved active-mobility infrastructure, and polycentric redistribution of amenities are necessary for enhancing accessibility equity and advancing the city’s transition toward a more sustainable and human-scaled urban model. Full article

Background: Obesity represents a multisystemic disorder that extends beyond metabolic dysfunction, involving hepatic, renal, and cardiovascular axes. This study introduces the Cardio–Reno–Metabolic–Liver–Obesity (CaRaMeL-O) framework as an integrated tool to assess multi-organ metabolic stress in adults with obesity. Methods: In this cross-sectional study, 287 adults with obesity (mean BMI 35.1 ± 4.6 kg/m²) were evaluated. The CaRaMeL-O score (0–13 points) incorporated metabolic (TyG index), hepatic (FIB-4, transaminases), and renal (eGFR, UACR) parameters, as well as classical and lifestyle risk factors. Participants were stratified into low, moderate, and high risk categories. Group comparisons were conducted using ANOVA and Kruskal–Wallis tests, while multivariate regressions identified independent predictors of FIB-4 and eGFR. Distributional characteristics were further analyzed using Weibull modeling. Results: Higher CaRaMeL-O scores were associated with a progressive increase in TyG (p < 0.001) and FIB-4 (p < 0.001), while eGFR showed a mild, nonsignificant downward trend. In multivariate models, age was the strongest predictor of FIB-4 (β_std = 0.33), whereas age, FIB-4, BMI, blood pressure, and UACR independently predicted eGFR. TyG did not remain significant after full adjustment. Weibull analysis revealed distinct distributional profiles, with TyG showing a narrow, homogeneous curve and FIB-4 and eGFR broader, right-skewed patterns. Conclusions: The CaRaMeL-O framework effectively captures inter-organ metabolic stress, demonstrating that hepatic and metabolic alterations precede overt renal decline. This integrated score may support early stratification and targeted prevention in obesity-related cardio-metabolic risk. Full article

In response to the challenges of low computational efficiency, insufficient detail restoration, and dependence on multiple GPUs in 3D Gaussian Splatting for large-scale UAV scene reconstruction, this study introduces an improved 3D Gaussian Splatting framework. It primarily targets two aspects: optimization of the partitioning strategy and enhancement of adaptive density control. Specifically, an adaptive partitioning strategy guided by scene complexity is designed to ensure more balanced computational workloads across spatial blocks. To preserve scene integrity, auxiliary point clouds are integrated during partition optimization. Furthermore, a pixel weight-scaling mechanism is employed to regulate the average gradient in adaptive density control, thereby mitigating excessive densification of Gaussians. This design accelerates the training process while maintaining high-fidelity rendering quality. Additionally, a task-scheduling algorithm based on frequency-domain analysis is incorporated to further improve computational resource utilization. Extensive experiments on multiple large-scale UAV datasets demonstrate that the proposed framework can be trained efficiently on a single RTX 3090 GPU, achieving more than a 50% reduction in average optimization time while maintaining PSNR, SSIM and LPIPS values that are comparable to or better than representative 3DGS-based methods; on the MatrixCity-S dataset (>6000 images), it attains the highest PSNR among 3DGS-based approaches and completes training on a single 24 GB GPU in less than 60% of the training time of DOGS. Nevertheless, the current framework still requires several hours of optimization for city-scale scenes and has so far only been evaluated on static UAV imagery with a fixed camera model, which may limit its applicability to dynamic scenes or heterogeneous sensor configurations. Full article

Background: Chronic rhinosinusitis (CRS) exhibits marked symptom heterogeneity that is not fully explained by anatomy or endotypes. Although allergen types shape symptom patterns in allergic rhinitis, largescale systematic analyses linking allergen sensitization profiles to patient-reported outcome measures in patients with CRS are limited. Methods: We conducted a multicenter, retrospective surgical cohort study (n = 1880) including patients with CRS who underwent preoperative specific IgE testing for 35 inhalant allergens and completed the 22-item Sino-Nasal Outcome Test (SNOT-22) questionnaire within 1 year. Using a previously validated nonnegative matrix factorization model, we deconvolved each patient’s IgE profile into four allergen signatures (Mite, Grass/Weed, Pet, and Tree) and defined a dominant group. Associations between signature contributions and SNOT-22 items, domain subscores, and total score were estimated by ordinary least squares, adjusting for age, sex, nasal polyps, and asthma, with coefficients scaled per 10-percentage-point increase. Item-level multiplicity was controlled for using the false discovery rate. Seasonality was assessed using monthly means and the coefficient of variation of the dominant group. Results: Dominant groups were nonallergic (50%), mite (26%), grass/weed (9%), pet (9%), and tree (5%). Symptoms varied by age and sex, characterized by notably low nasal scores with aging and a high female burden for several items, motivating covariate adjustment. Signature–symptom associations were domain-specific: the pet signature showed the strongest and most consistent associations with nasal domain (such as rhinorrhea and nasal obstruction) and emotion domain (feelings of embarrassment); mite and grass/weed signatures were linked to the function domain (daytime fatigue/productivity); whereas the tree signature showed no significant associations. Seasonal patterns aligned with exposure ecology: grass/weed and tree groups had the largest relative variation (high coefficient of variations), the pet group showed the highest absolute burden year-round, and the mite group varied modestly with winter–spring predominance. Conclusions: Allergen signatures distilled from routine IgE panels explained meaningful variations in CRS patient-reported outcome measures, mapping to distinct symptom domains and seasonal profiles. Incorporating signature information into clinical assessments may support personalized counseling, anticipatory management around exposure windows, and targeted evaluation of environmental or immunologic interventions. Full article

Large-scale river degradation constitutes a global challenge, rendering the ecological restoration of impaired rivers ever more crucial. While ecological restoration projects have enhanced the quality of river habitats, given the dynamic nature and complexity of river and lake ecosystems, the achievement of sustainable restoration of fish habitats and the assurance of its effectiveness continue to face numerous challenges. Consequently, this study proposes an improved approach to determine the ecological flow requirements of fish habitats in impaired rivers. In relation to the screening of key species, a bespoke evaluation index system has been developed specifically for impaired rivers lacking rare and endemic fish species. Primary data were collected via field surveys, ecological monitoring, and a review of the literature, while the analytic hierarchy process (AHP) was utilized to quantitatively identify key species. In the development of the assessment framework, three core indicators were integrated: habitat-weighted usable area (WUA), habitat connectivity index (HCI), and microhabitat heterogeneity index (RMH). Incorporating the ecological requirements of key fish species across different life stages, a systematic analysis was undertaken to explore the ecological response effects of different indicator combinations under varying flow regimes. The results revealed that a flow rate of 160 m³/s gives rise to an inflection point in the RMH diversity index at 1.618, whereas a flow rate of 240 m³/s results in a significant inflection point in the HCI at 0.652. At a flow rate of 260 m³/s, the WUA attains 2,007,928 m². The optimal ecological flow range was determined to be 160–240 m³/s for the breeding period (March–June), 240–260 m³/s for the foraging period (July–October), and 120 m³/s for the winter period. These findings provide a theoretical framework for the restoration of target fish populations in similarly degraded rivers. Full article