Search Results (60,201)

Background: The growth of digital patient feedback presents a new opportunity for healthcare quality monitoring. This study addresses the need to automatically classify the content of patient reviews to identify primary sources of dissatisfaction. Objective: The purpose of this study is to develop a machine learning algorithm for classifying negative patient reviews into two core categories: medical content (M—pertaining to diagnosis, treatment, and outcomes) and organizational support (O—pertaining to logistics, cost, and communication). We aim to identify which type of concern prevails and to analyze variations across cities, patient gender, and medical specialties. Methods: A database of 18,680 negative patient reviews (rated 1 star) was compiled from the Russian aggregator infodoctor.ru for the period from July 2012 to August 2023. A training set was created using an independent annotation procedure with three experts. A logistic regression model was trained to classify reviews into M and O categories, demonstrating an accuracy of 88.5%. Results: The analysis revealed a significant structural shift in Moscow, where since 2021, medical (M) complaints began to prevail over organizational (O) ones. This trend was not observed in St. Petersburg or other major Russian cities. Notably, in St. Petersburg, M-type reviews were more common within the most represented medical specialties, whereas O-type reviews consistently dominated in other cities. Gender differences were most pronounced in St. Petersburg, where women were more frequently authors of M reviews and men of O reviews. Conclusions: The developed algorithm provides a valuable tool for the automated monitoring of patient feedback. It enables healthcare managers to distinguish between clinical and service-related issues, facilitating targeted improvements in medical service quality and patient satisfaction. Full article

Turbidity is a crucial and reliable indicator that is extensively utilized in water quality monitoring through remote sensing technology. The development of accurate and applicable models for turbidity estimation is essential. While many existing studies rely on uniform models based on statistical regression or traditional machine learning techniques, the application of deep learning models for turbidity estimation remains limited. This study proposed deep learning models for turbidity estimation based on optical classification of inland waters using Sentinel-2 data. Specifically, the fuzzy c-means (FCM) clustering method was employed to classify optical water types (OWTs) based on their spectral reflectance characteristics. A weighted sum of the turbidity prediction results was generated by the OWT-based convolutional neural network-random forest (CNN-RF) model, with weights derived from the FCM membership degrees. Turbidity for four typical waters was mapped by the proposed method using Sentinel-2 images. The FCM method efficiently classified waters into three OWTs. The OWT-based weighted CNN-RF model demonstrated strong robustness and generalization performance, achieving a high prediction accuracy (R² = 0.900, RMSE = 11.698 NTU). The turbidity maps preserved the spatial continuity of the turbidity distribution and accurately reflected water quality conditions. These findings facilitate the application of deep learning models based on optical classification in turbidity estimation and enhance the capabilities of remote sensing for water quality monitoring. Full article

Text recognition on coffee bean package labels is of great importance for product tracking and brand verification, but it poses a challenge due to variations in image quality, packaging materials, and environmental conditions. In this paper, we propose a pipeline that combines several image enhancement techniques and is followed by an Optical Character Recognition (OCR) model based on vision–language (VL) Qwen VL variants, conditioned by structured prompts. To facilitate the evaluation, we construct a coffee bean package image set containing two subsets, namely low-resolution (LRCB) and high-resolution coffee bean image sets (HRCB), enclosing multiple real-world challenges. These cases involve various packaging types (bottles and bags), label sides (front and back), rotation, and different illumination. To address the image quality problem, we design a dedicated preprocessing pipeline for package label situations. We develop and evaluate four Qwen-VL OCR variants with prompt engineering, which are compared against four baselines: DocTR, PaddleOCR, EasyOCR, and Tesseract. Extensive comparison using various metrics, including the Levenshtein distance, Cosine similarity, Jaccard index, Exact Match, BLEU score, and ROUGE scores (ROUGE-1, ROUGE-2, and ROUGE-L), proves significant improvements upon the baselines. In addition, the public POIE dataset validation test proves how well the framework can generalize, thus demonstrating its practicality and reliability for label recognition. Full article

Population mobility links cities, propelling the spatiotemporal spread of urban pandemics and adding complexity to disease dynamics. It also closely shapes, and is shaped by, the selection and intensity of intervention measures. Revealing the multistage spatial-temporal dynamics of cross-regional epidemic continuity under this interaction is often overlooked but critically important. This study innovatively applies a self-organizing map (SOM) neural network to classify cities into six distinct types based on population mobility characteristics: high-inflow core (HIC), low-inflow core (LIC), low-inflow sub-core (LISC), high-outflow semi-peripheral (HOSP), equilibrious semi-peripheral (ESP), and low-outflow peripheral (LOP). Building on this, we propose a novel SEIR-AHQ theoretical framework and construct an epidemiological model using network-coupled ordinary differential equations (ODEs). This model captures the dynamic interplay between inter-city population mobility and intervention measures, and quantifies how heterogeneous city types shape the evolution of epidemic transmission across the coupled mobility network. The results show that: (1) Cities with stronger population mobility face significantly higher infection risks and longer epidemic durations, characterized by “higher peaks and longer tails” in infection curves. HIC cities experience the greatest challenges, and LOP cities experience the least. (2) Both higher transmission rates and delayed intervention timings lead to exponential growth in infections, with nonlinear effects amplifying small changes disproportionately. (3) Intervention efficacy follows a “diminishing marginal returns” pattern, where the incremental benefits of increasing intervention intensity gradually decrease. This study offers a novel perspective on managing interregional epidemics, providing actionable insights for crafting tailored and effective epidemic response strategies. Full article

In the context of industrial transitions and tariff frictions, financial markets are experiencing frequent defaults, emphasizing the urgency of upgrading credit scoring methodologies. A novel credit risk identification model integrating generative adversarial networks (GAN) and the minimax probability machine (MPM) is proposed. GAN generates realistic augmented samples to alleviate class imbalance in the credit score dataset, while the MPM optimizes the classification hyperplane by reformulating probability constraints into second-order cone problems via the multivariate Chebyshev inequality. Numerical experiments conducted on the South German Credit dataset, which represents individual (consumer) credit risk, demonstrate that the proposed generative adversarial network’s minimax probability machine (GAN-MPM) model achieves 76.13%, 60.93%, 71.78%, and 72.03% for accuracy, F1-score, sensitivity, and AUC, respectively, significantly outperforming support vector machines, random forests, and XGBoost. Furthermore, SHAP analysis reveals that the installment rate in percentage of disposable income, housing type, duration in month, and status of existing checking accounts are the most influential features. These findings demonstrate the effectiveness and interpretability of the GAN-MPM model, offering a more accurate and reliable tool for credit risk management. Full article

To improve the accuracy and stability of anomaly detection in lithium-ion batteries for electric bicycles, in this study, we propose a hybrid deep learning model that integrates a convolutional neural network (CNN), long short-term memory (LSTM) network, and attention mechanism to extract local temporal features, capture long-term dependencies, and adaptively focus on key time segments around anomaly occurrences, respectively, thereby achieving a balance between local and global feature modeling. In terms of data preprocessing, separate feature sets are constructed for charging and discharging conditions, and sliding windows combined with min–max normalization are applied to generate model inputs. The model was trained and validated on large-scale real-world battery operation data. The experimental results demonstrate that the proposed method achieves high detection accuracy and robustness in terms of reconstruction error distribution, alarm rate stability, and Top-K anomaly consistency. The method can effectively identify various types of abnormal operating conditions in unlabeled datasets based on unsupervised learning. This study provides a transferable deep learning solution for enhancing the safety monitoring of electric bicycle batteries. Full article

Coastal profile models are frequently used for the computation of storm-induced erosion at (nourished) beaches. Attention is focused on new developments and new validation exercises for the detailed process-based CROSMOR-model for the computation of storm-induced morphological changes in sand and gravel coasts. The following new model improvements are studied: (1) improved runup equations based on the available field data; (2) the inclusion of the uniformity coefficient (C_u = d₆₀/d₁₀) of the bed material affecting the settling velocity of the suspended sediment and thus the suspended sediment transport; (3) the inclusion of hard bottom layers, so that the effect of a submerged breakwater on the beach–dune morphology can be assessed; and (4) the determination of adequate model settings for the accretive and erosive conditions of coarse gravel–shingle types of coasts (sediment range of 2 to 40 mm). The improved model has been extensively validated for sand and gravel coasts using the available field data sets. Furthermore, a series of sensitivity computations have been made to study the numerical parameters (time step, grid size and bed-smoothing) and key physical parameters (sediment size, wave height, wave incidence angle, wave asymmetry and wave-induced undertow), conditions affecting the beach morphodynamic processes. Finally, the model has been used to study various alternative methods of reducing beach erosion. Full article

This study presented a comprehensive finite element (FE) investigation into the flexural behavior of RC T-beams strengthened in the negative moment region using near-surface mounted (NSM) carbon-fiber-reinforced polymers (CFRP) rods. A three-dimensional nonlinear FE model was developed and validated against experimental data, achieving close agreement with normalized mean square error values as low as 0.006 and experimental-to-numerical ratios ranging from 0.95 to 1.04. The validated model was then employed to conduct a systematic parametric analysis considering CFRP rod diameter, concrete compressive strength, longitudinal reinforcement ratio, and FRP material type. The results showed that increasing CFRP diameter from 6 to 10 mm enhanced ultimate load by up to 47.51% and improved stiffness by 1.48 times. Higher concrete compressive strength contributed to stiffness gains exceeding 50.00%, although this improvement was accompanied by reductions in ductility. Beams with reinforcement ratios up to 2.90% achieved peak loads of 309.61 kN, but ductility declined. Comparison among FRP materials indicated that CFRP and AFRP offered superior strength and stiffness, whereas BFRP provided a more balanced combination of strength and deformation capacity. Full article

Aiming at the problem that existing equipment importance evaluation methods fail to consider interconnectivity between pieces of equipment, variability after maintenance, and the impact of dynamically changing situations on importance, and focusing on the dynamic support needs of equipment in a conflict environment, this paper proposes a batch allocation method for equipment maintenance tasks considering dynamic importance. The purpose of this study is to determine the batch priority of equipment maintenance based on the dynamically changing importance of pieces of equipment. First, a dynamic importance index system is constructed: a real-time CRITIC-AHP combined weighting method is used to calculate team importance, a dynamic Bayesian network (DBN)-influenced method is used to calculate relative importance, an attention–LSTM time-series prediction method is used to calculate future importance, and then a dynamic entropy weight method is adopted to objectively integrate the three types of importance. Second, a dual-objective optimization model with the maximum equipment importance and the minimum total maintenance time is built, with mobile distance, maintenance time, and maintenance capacity as constraints. The Dynamic Particle Swarm Optimization (DPSO) algorithm is used to solve this model, and its dynamic adaptability is improved through environmental change detection and adaptive adjustment of inertia weight. Finally, the batch allocation of maintenance tasks is realized. Example verification shows that compared with the expert scoring method, the errors of the three importance calculation methods are all reduced by more than 60%, the optimization speed of the dynamic PSO algorithm is 47% faster than that of the static algorithm, and the constructed model has good stability. This method can provide a reference for maintenance support command decisions. Full article

Background: Oncolytic virus (OV) therapy couples direct tumor lysis with systemic immune priming, yet clinical benefit remains heterogeneous and the predictive biomarker landscape is poorly defined. We undertook a systematic review and meta-analysis to quantify the efficacy and safety of OV therapy in solid tumors and to synthesize current evidence on response-modulating biomarkers. Methods: Following PRISMA 2020 guidelines, MEDLINE, Embase, Cochrane CENTRAL, ProQuest and Scopus were searched from inception to May 2025. Phase II–III randomized trials of genetically engineered or naturally occurring OV reporting objective response rate (ORR), progression-free survival (PFS), overall survival (OS) or biomarker data were eligible. Hazard ratios (HRs) or odds ratios (OR) were pooled with random-effects models; heterogeneity was assessed with I² statistics. Qualitative synthesis integrated genomic, immunologic and microbiome biomarkers. Results: Thirty-six trials encompassing around 4190 patients across different tumor types met inclusion criteria. Compared with standard therapy, OV-based regimens significantly improved ORR nearly three-fold (pooled OR = 2.77, 95% CI 1.85–4.16), prolonged PFS by 11% (HR = 0.89, 95% CI 0.80–0.99) and reduced mortality by 16% (OS HR = 0.84, 95% CI 0.72–0.97; I² = 59%). Benefits were most pronounced in melanoma (ORR 26–49%; OS HR 0.57–0.79) and in high-dose vaccinia virus for hepatocellular carcinoma (HR = 0.39). Grade ≥ 3 adverse events were not increased versus control (risk ratio 1.05, 95% CI 0.89–1.24); common toxicities were transient flu-like symptoms and injection-site reactions. Biomarker synthesis revealed that high tumor mutational burden, interferon-pathway loss-of-function mutations, baseline CD8⁺ T-cell infiltration, post-OV upregulation of IFN-γ/PD-L1, and favorable gut microbial signatures correlated with response, whereas intact antiviral signaling, immune-excluded microenvironments and myeloid dominance predicted resistance. Conclusions: OV therapy confers clinically meaningful improvements in tumor response, PFS and OS with a favorable safety profile. Integrating composite genomic–immune–microbiome biomarkers into trial design is critical to refine patient selection and realize precision viro-immunotherapy. Future research should prioritize biomarker-enriched, rational combination strategies to overcome resistance and extend benefit beyond melanoma. Full article

Heart failure with preserved ejection fraction (HFpEF) is a complex syndrome characterized by left ventricular diastolic dysfunction, exercise intolerance, low-grade chronic inflammation, and comorbidities such as hypertension, obesity, and glucose intolerance. Myocardial infiltration by activated macrophages has been proposed as a mechanism linking low-grade inflammation to increased diastolic LV stiffness in HFpEF. Changes in the relative abundance of cardiac macrophage populations may precede and promote the development of HFpEF in the aged heart. This study aimed to characterize the cardiac macrophage subsets that predominate during progression from experimental preclinical to established HFpEF. To generate the model, wild-type male C57BL/6N mice were randomized to control chow or a combination of high-fat diet plus L-NAME in drinking water for 5 weeks (asymptomatic pre-HFpEF) or 15 weeks (established HFpEF). At the end of each period, we measured body weight, running distance, metabolic biomarkers, systolic and diastolic blood pressure, myocardial function and morphology, cardiac remodeling by hypertrophic markers, morphometric analyses, fibrosis, cytokines TNF-α and IL-10, cardiac macrophage phenotype profiles (CCR2⁺ and CCR2⁻), and AMP-Activated Protein Kinase (AMPK)activity.Significant changes in myocardial macrophage populations were observed at 5 weeks (pre-HFpEF), specifically a decrease in resident reparative CCR2⁻MHCII⁻ and increase in proinflammatory CCR2⁺MHCII⁺ macrophages. These early changes were associated with higher circulating TNF-α, decreased myocardial AMPK activation, and more severe myocardial fibrosis. At 15 weeks (established HFpEF), proinflammatory CCR2⁺MHCII⁺ macrophage levels remained elevated in the myocardium; whereas the initial number of resident reparative CCR2⁻MHCII^- levels was reduced, it subsequently returned to baseline. In this model of HFpEF induced by a high-fat diet and L-NAME, which produced obesity, glucose intolerance, and hypertension, myocardial resident reparative CCR2⁻MHCII⁻ macrophages decreased and proinflammatory CCR2⁺MHCII⁺ macrophages increased during preclinical stages. These early changes in cardiac macrophage profile were associated with low-grade inflammation and myocardial remodeling and preceded the onset of HFpEF. Full article

New-generation batteries are attracting increasing interest in response to today’s energy storage challenges, as evidenced by the steady rise in scientific publications on the topic. However, their industrial deployment remains limited due to the complexity of aging mechanisms, which are still poorly understood and difficult to control. While several promising developments have emerged in laboratory settings, they remain too immature to be scaled up. These aging processes, which directly affect the performance, safety, and lifespan of battery systems, also determine their technical and economic viability. This review offers a comparative analysis of aging phenomena—both specific to individual technologies and common across systems—drawing on findings from accelerated testing, post-mortem analyses, and modeling. It highlights critical failures such as interface instability, loss of active material, and mechanical stress, while also identifying shared patterns and the unique features of each technology. By combining experimental data with theoretical approaches, the article proposes an integrated framework for understanding and prioritizing aging mechanisms by technology type. It underscores the limitations of current characterization techniques, the urgent need for harmonized testing protocols, and the importance of standardized data sharing. Finally, it outlines possible avenues for improving the understanding and mitigation of aging phenomena. Full article

Understanding the diversity of star formation histories (SFHs) of galaxies is key to reconstructing their evolutionary paths. Traditional models often assume parametric forms such as delayed-$\tau$ or exponentially declining models, which may not reflect the actual variety of formation processes. We aim to assess what types of SFHs are consistent with the observed present-day star formation rates (${\mathrm{SFR}}_0$) and time-averaged star formation rates ($⟨\mathrm{SFR}⟩$) of galaxies in the Local Volume, without assuming any fixed functional form. We construct a non-parametric framework by generating large ensembles of randomized SFHs for each galaxy in the sample. For each SFH, we compute its predicted stellar mass and present-day SFR and retain only those consistent with the observed values within a 20% tolerance. We then infer the statistical distribution of power-law slopes $\eta$ (fitted as $\mathrm{SFR}\left(t\right)\propto {(t-t_{\mathrm{start}})}^{\eta }$) and 50% stellar mass formation times $t_{50}$. Across the full sample of 555 galaxies, we find that ≈70% have flat SFHs ($\left|\eta \right|\le 0.01$), ≈24% are mildly declining ($\eta <-0.01$), and ≈6% are rising ($\eta >0.01$). In the low-mass bin ($M_{★}<3\times {10}^9{M}_{\odot }$), rising SFHs slightly increase (≈7%) but remain a minority as the majority have flat SFHs. Both $\eta$ and $t_{50}$ correlate strongly with the SFR ratio (Spearman $\rho >0.75$, $p\ll {10}^{-16}$), indicating that the shape and timing of star formation are primarily governed by this ratio. The $t_{50}$ distribution shows sharp spikes near 7.74 and 7.86 Gyr, which we attribute to grid discretization combined with filtering, rather than a physical bimodality. Our results confirm that strongly declining SFH templates are disfavored in the Local Volume: most systems are consistent with flat long-term SFHs, with only mild decline or occasional rising. Importantly, this is demonstrated through a fully model-independent, data-driven approach, with per-galaxy uncertainties quantified using the standard error of $\eta$ and $t_{50}$ from the ensemble of accepted SFHs. Full article

The tire-road friction coefficient (TRFC) is a critical parameter that significantly influences vehicle safety, handling stability, and driving comfort. Existing estimation methods based on vehicle dynamics suffer from a substantial decline in accuracy under conditions with insufficient excitation, while vision-based approaches are often limited by the generalization ability of their datasets, making them less effective in complex and variable real-driving environments. To address these challenges, this paper proposes a novel, low-cost fusion method for TRFC estimation that integrates weather conditions and road image data. The proposed approach begins by employing semantic segmentation to partition the input images into distinct regions—sky and road. The segmented images will be fed into the road recognition network and the weather recognition network for road type and weather classification. Furthermore, a fusion decision tree incorporating an uncertainty modeling mechanism is introduced to dynamically integrate these multi-source features, thereby enhancing the robustness of the estimation. Experimental results demonstrate that the proposed method maintains stable and reliable estimation performance even on unseen road surfaces, outperforming single-modality methods significantly. This indicates its high practical value and promising potential for broad application. Full article

Food antioxidant supplementation has been widely proposed for cancer prevention and adjuvant therapy due to the pleiotropic role of antioxidants. Herein, particular attention is given to recent clinical trials based on the use of dietary supplements in cancer patients, both as monotherapy and in combination with standard treatments, exploring both their potential benefits and risks. This review focuses on the efficacy of the most important food antioxidants, highlighting how their action may change depending on different factors such as cancer type, dose, timing of administration and antioxidant status of the patient. The results of clinical trials are often contradictory, and the clinical benefit of dietary antioxidants appears more consistent in patients with a baseline antioxidant deficiency. Furthermore, by analyzing the mechanisms underlying the contradictory clinical evidence and critically addressing the issues related to the methodologies used in preclinical models, this review could be helpful in guiding the personalized use of antioxidant supplementation in cancer patients. Full article