Search Results (12,020)

With the continuous increase in coal mining depth in China, concealed geological structures—such as collapsed columns and faults—pose a severe threat to mine safety by triggering water inrush accidents. Mine DC resistivity methods exhibit high sensitivity to water-bearing characteristics but suffer from limited resolution, while mine seismic exploration offers superior resolution but weak sensitivity to water-rich bodies. Single-method inversion is inevitably plagued by solution non-uniqueness. This study aims to enhance the detection accuracy of concealed structures by implementing a joint seismic–electric inversion that exploits the complementary strengths of both methods. For the DC resistivity component, a forward model was established using the finite element method with unstructured meshes, and inversion was performed via Occam regularization. For seismic exploration, forward modeling employed curved-ray tracing, and inversion was conducted via the LSQR algorithm. Cross-gradient constraints were incorporated into the joint inversion to establish a structurally coupled framework. The novelty of this study lies in the integration of unstructured mesh discretization, curved-ray seismic tomography, and cross-gradient-constrained joint inversion for mine water detection. Numerical simulation results demonstrate that joint inversion effectively constrains the spatial extent of anomalies, accurately characterizes the morphology of multiple anomalous bodies and water-conducting fault channels, and substantially reduces solution non-uniqueness compared to single-method inversions. This research provides a reliable methodology for the refined detection of concealed hazard-inducing structures, offering considerable practical value for safeguarding coal mine safety. Full article

Background: Children with food allergy (FA) may develop eating difficulties that extend beyond medically required allergen avoidance, including food neophobia, selective eating, and avoidant/restrictive food intake disorder (ARFID)-related symptoms. This mixed-methods study aimed to assess ARFID symptoms, food neophobia, diet quality, and food-related psychological experiences in children with FA. Methods: Seventy children with confirmed FA were enrolled in an observational cross-sectional mixed-methods study conducted at a tertiary pediatric allergy center. Quantitative assessment included the Eating Disorders in Youth-Questionnaire (EDY-Q), the Italian Child Food Neophobia Scale (ICFNS), and the Mediterranean Diet Quality Index for Children and Adolescents (KIDMED). Qualitative data were collected from a purposive subsample of 26 children through semi-structured child-friendly interviews and drawing-based elicitation activities and were analyzed using inductive thematic analysis. Results: A positive EDY-Q screening result for ARFID-related symptoms was observed in 3 children (4.3%). Selective Eating showed the highest median EDY-Q domain score (median 1.50, IQR 0.67–4.00), followed by Food Avoidance Emotional Disorder (median 1.00, IQR 0.00–2.00) and Functional Dysphagia (median 0.00, IQR 0.00–2.00). Moderate food neophobia was observed in 45 children (64.3%), while 24 children (34.3%) were classified within the high food-neophobia category. KIDMED score was negatively correlated with EDY-Q total score (Spearman’s ρ = −0.302; p = 0.011), Selective Eating (ρ = −0.356; p = 0.002), and Functional Dysphagia (ρ = −0.238; p = 0.048). In an exploratory hierarchical regression, Selective Eating remained inversely associated with KIDMED after adjustment for demographic and FA-related covariates (β = −0.386; p = 0.002); however, the overall model did not reach statistical significance (p = 0.081), and the additional clinical covariates did not significantly improve model fit. Accordingly, the fully adjusted findings should be interpreted as exploratory. Qualitative analysis identified five themes: anticipatory fear of food and tasting, emotional memory of allergic reactions and generalization of risk, selective avoidance and food rigidity, reassurance-seeking and decision-making dependence, and ambivalent parental dynamics. Conclusions: Within this tertiary-care sample, positive EDY-Q screening results were uncommon, whereas broader food-related difficulties were identified. Pediatric FA assessment may benefit from attention to eating behavior, diet quality, and avoidance extending beyond confirmed allergens. Full article

In scrubbing tower (ST) systems, outlet vapor temperature regulation is subject to strict thermal constraints, requiring the bottom temperature to remain below a critical safety threshold. Transient overshoots can rapidly trigger pyrolytic coking, foul mass-transfer packed beds and spray nozzles, and ultimately force complete production shutdowns. Due to feedstock compositional drifts, high thermal inertia, and significant transport delays, high-fidelity predictive identification is essential for proactive early warning and for overcoming the limitations of reactive feedback control. To address these bottlenecks, this paper introduces an offset-free, hard-constrained, data-driven adaptive optimal control paradigm, designated as the improved GRU-coupled conjugate gradient linear quadratic regulator (IGRUCG-LQR). First, by constructing an augmented state space embedded with integral error, the proposed paradigm eliminates permanent tracking offsets induced by long-term nonstationary drifts. Second, automatic differentiation is used to extract the time-varying Jacobian matrix of a gated recurrent unit (GRU) online, thereby tracking the nonlinear evolution of the underlying thermodynamic baseline with high fidelity. To manage the critical trade-off between strict actuator saturation and short real-time sampling intervals, the conjugate gradient (CG) method is fused with a hard-boundary projection operator, enforcing physical constraints with high computational efficiency and without complex matrix inversions. Experimental validation on a real-world industrial dataset demonstrates that the proposed paradigm secures the safety baseline while achieving high-resolution transient tracking. Furthermore, it significantly suppresses high-frequency valve chattering to mitigate mechanical fatigue, establishing a solid theoretical and engineering foundation for the prolonged stable operation of safety-critical processes. The proposed framework achieves an Integral Absolute Error (IAE) of 86.66 and an Integral Time Absolute Error (ITAE) of 4064.49. Full article

Memory-bank methods such as PatchCore are widely used in industrial quality control for visual anomaly detection since they require no training, are fast to deploy and achieve strong accuracy. However, they are memory-intensive. Furthermore, a single production line typically involves different products or cameras, so using a single anomaly detection method with a shared nearest-neighbor memory bank is attractive since it simplifies deployment and makes new products easy to add. Nevertheless, embeddings from different products/cameras can interfere during retrieval, causing what we call “memory pollution”. In this work, we study this effect through a new diagnostic framework, which involves: (1) a new metric, the wrong-neighbor rate (WNR), which measures how often a query’s nearest neighbor belongs to a different product; (2) an empirically validated phenomenon, “oracle inversion”, where querying only the product’s own data can underperform the shared bank under a fixed memory budget; (3) a first-order analytical model of the WNR, which predicts how pollution grows with product count and memory budget; and (4) a minimal training-free router that removes the effect of memory pollution. Our results show that our system performs robustly across different datasets and backbones, with up to 25× memory reduction, which makes our framework attractive for industrial applications. Full article

Background: Influenza poses a serious threat to the health of older adults, and vaccination is a key strategy for prevention. As an economically developed city, Suzhou has a rapidly aging population, yet the influenza vaccination rate among older adults remains substantially lower than that in developed countries. There is an urgent need to understand the current vaccination status and its associated factors to inform future strategies for improving vaccination coverage. Objectives: To compare influenza- and vaccine-related knowledge, attitudes, practices, and information access channels between vaccinated and unvaccinated older adults (≥60 years) in Suzhou, and to identify factors associated with vaccination behavior in economically developed areas. Methods: A comparative cross-sectional survey was conducted from April to August 2024 across six districts or county-level cities randomly selected from Suzhou’s ten county-level administrative divisions. Participants were divided into vaccinated and unvaccinated groups based on their influenza vaccination status during the 2023–2024 influenza season. A two-stage sampling method was employed: the six districts were selected as primary sampling units; within each district, eligible older adults (aged ≥60 years) were randomly selected from two independent sampling frames—the vaccination information system for the vaccinated group and the basic public health service system for the unvaccinated group. Face-to-face structured interviews were conducted using a questionnaire consisting of four sections; data were entered using EpiData 13.1 and analyzed using SPSS 27.0. The χ² test, t-test, and multivariable logistic regression were used for statistical analysis. Results: Among 4622 valid older adults (vaccinated: n = 2007; unvaccinated: n = 2615), the vaccinated group scored significantly higher on influenza-related knowledge (4.06 ± 2.21 vs. 3.16 ± 2.33, p < 0.001), vaccine-related knowledge (2.10 ± 1.18 vs. 0.67 ± 1.07, p < 0.001), and perceived influenza risk scores (2.72 ± 1.42 vs. 2.26 ± 1.64, p < 0.001). Vaccine safety and efficacy endorsement were also markedly higher among the vaccinated group (70.3% vs. 22.0%; 49.7% vs. 9.3%). Regarding information sources, the vaccinated group relied more on healthcare institutions (54.56% vs. 46.92%), whereas the unvaccinated group relied more on television (60.34% vs. 51.17%). Multivariable logistic regression revealed that awareness of the influenza vaccine (aOR = 9.151, 95%CI: 6.32–13.25), having a planned vaccination site (aOR = 2.66, 95%CI: 2.01–3.54), and perceiving the vaccine as safe (aOR = 1.81, 95%CI: 1.31–2.49) were positively associated with vaccination, while rural household registration (aOR = 0.76, 95%CI: 0.63–0.93), full-time employment (aOR = 0.41, 95%CI: 0.28–0.60), and hypertension (aOR = 0.79, 95%CI: 0.65–0.94) were inversely associated. Conclusions: Compared with the unvaccinated group, the vaccinated group demonstrated significantly better influenza- and vaccine-related knowledge, risk perception, and recognition of vaccine safety and efficacy, and relied more on healthcare institutions for information. Multivariable regression analysis identified awareness of the influenza vaccine as the strongest associated factor. These findings suggest that vaccination coverage could be improved by strengthening vaccine knowledge promotion through healthcare professionals and increasing awareness of vaccination policies that integrate medical insurance reimbursement. Full article

The helicopter tail drive shaft is vulnerable to failure from projectile impacts during low-altitude flight. Stress wave-based inversion of impact parameters and quantitative damage assessment remain insufficiently explored. To address small-sample and nonlinear challenges, a framework based on stress wave characteristics and physics-guided hierarchical Gaussian process regression is proposed. Four key features, namely first-arrival wave trough amplitude, frequency standard deviation, ratio of low-frequency to high-frequency root mean square, and wavelet energy entropy, are extracted from transient signals to construct a hierarchical progressive architecture for damage mode discrimination, parameter inversion, and quantitative assessment. Perforation is identified using a wavelet energy entropy-based adaptive threshold. Incidence angle inversion is achieved by an adaptive composite kernel and Bayesian physical prior correction. Damage degree is assessed through residual learning guided by a physical prior surface mean function. Results show an incidence angle inversion root mean square error (RMSE) of 3.02°, with entry and exit hole equivalent failure area RMSEs of 13.32 mm² and 12.98 mm², respectively. The 95% prediction interval maintained reliable coverage across the validation samples. This framework provides a new method with both physical interpretability and uncertainty quantification for the assessment of impact damage in thin-walled tube structures. Full article

In the era of big data, the demand for vast quantities of diverse and representative datasets has surged across various domains, from healthcare and finance to artificial intelligence and machine learning. Synthetic data generation offers a promising solution by enabling the creation of data with specific properties that closely mimic real-world data while avoiding privacy concerns and regulatory limitations. However, generating high-quality synthetic data that accurately preserves complex dependencies remains a significant challenge. This paper addresses this gap by exploring a novel approach: Inverse Copula Sampling for Multi-Dimensional Data Synthesis. Utilizing copulas, which are powerful tools for modeling dependencies between variables, our method generates synthetic data that maintains intricate interdependencies. We demonstrate the effectiveness of this approach through various experiments and case studies, showing high fidelity in preserving dependencies and minor discrepancies in marginal distributions. The method’s robustness was validated through comparative analysis and statistical checks, including the Kolmogorov–Smirnov test. Our research contributes to the field by introducing a flexible and efficient method for synthetic data generation that is applicable to a wide range of data distributions and practical applications. Future work will explore the application of other copula types and the further refinement of the method to enhance its versatility. Full article

This paper develops a parameterized generalized transform framework for nonlinear differential models. The method combines a generalized Laplace-type transform, Adomian decomposition, Chebyshev–Padé rational reconstruction, and a $\mu$-scaled generalized transform to construct admissible semi-analytical approximations. The framework treats the transform geometry as part of the approximation process, allowing the transformed domain to be adjusted while preserving an explicit analytical structure. The theoretical analysis establishes admissibility conditions, existence of admissible minimizers, characterization of the admissible region for the parametric kernel, interior optimality conditions, inverse and residue inversion formulas, and fixed-point consistency with a first-order truncation estimate. The method is illustrated on a logistic–Allee tumor-growth model using experimental data. The resulting compact representations remain real-valued and admissible on the full data interval and produce errors comparable to standard numerical reference solutions. Full article

Background/Objectives: The clinical profile of serum Maresin-1 (MaR1) in relation to diabetic retinopathy phenotypes and proteinuria in type 2 diabetes mellitus (T2DM) remains unclear. We evaluated serum MaR1 across healthy controls and patients with T2DM without diabetic retinopathy (DR), non-proliferative DR (NPDR), or proliferative DR (PDR), and examined the relationship between MaR1 and the urine protein-to-creatinine ratio (UPCR). Methods: This single-center cross-sectional study included 93 participants. Serum MaR1 was measured by ELISA. Group differences were assessed with the Kruskal–Wallis test and Holm-adjusted post hoc tests. DR phenotypes were analyzed among patients with T2DM. The MaR1-UPCR relationship was examined using correlation and adjusted regression models. Results: MaR1 differed across groups (H = 49.36, p < 0.001, epsilon² = 0.521). The median MaR1 was 89.8 (82.8–97.2) pg/mL in controls and 34.3 (33.0–36.0), 35.8 (34.4–36.7), and 34.0 (33.1–35.7) pg/mL in T2DM without DR, NPDR, and PDR, respectively. MaR1 was higher in controls than in all T2DM groups, whereas T2DM groups did not differ. Within T2DM, MaR1 was not associated with DR stage (H = 4.44, p = 0.109; rho = −0.001, p = 0.996). MaR1 was inversely related to the UPCR overall (rho = −0.272, p = 0.008), but not within T2DM (rho = −0.057, p = 0.634) or in adjusted models. Conclusions: MaR1 was markedly lower in T2DM than in controls. This reduction was not explained by DR stage or proteinuria. These findings indicate that MaR1 should be interpreted as a T2DM-associated systemic alteration rather than as a marker of retinopathy stage or proteinuria. Full article

Among the micropollutants of medium-depth waters in isolated inhabited areas, the inorganic ones deserve special attention: nitrate anion (NO₃⁻) and phosphate anions (HxPO₄^−(3−x)). The individual removal of these anions from water is widely studied, with different methods being found: chemical, ion exchange, or biological. This paper presents a membrane method for the simultaneous removal of nitrate anion and phosphate anions from dilute synthetic aqueous solutions. The developed method is nanofiltration using composite membranes made of cellulose acetate (CA) and silver nanoparticles (Agnp). The composite membranes were made by phase inversion of the dimethylformamide (DMF) solution containing the two components (CA–Agnp) on a polypropylene (PP) capillary fiber using deionized water as a coagulant. The DMF solution of CA containing Agnp was obtained by dissolving black-and-white cinematographic films (exposed and unexposed to light). CA–Agnp–PP composite membranes were tested for the simultaneous removal of nitrate anion and phosphate anions from aqueous solution by nanofiltration at pressures ranging from 5 to 25 bars. A removal of over 98% of phosphate anions and more than 95% of nitrate anions was achieved. Fluxes of 10 L·m⁻²·h⁻¹ were obtained for the working pressure of 15 atm, depending on the pH, flow rate, and concentration of the feed water. The variable parameters considered were also the concentrations of CA and Agnp. Full article

Background/Objectives: Caffeine citrate is widely used in very preterm infants, but whether initiation during the first postnatal hours is independently associated with respiratory outcomes remains uncertain. We evaluated the association between caffeine initiation within versus after the first 2 h of life and respiratory outcomes in infants born at <32 weeks’ gestation. Methods: This single-center observational cohort study included 84 infants born at <32 weeks’ gestation and with a birth weight of ≤1500 g who received caffeine citrate within the first 24 h of life. The 2 h threshold represented the unit’s protocol-defined target for caffeine loading rather than a biologically validated cutoff. The primary outcome was bronchopulmonary dysplasia (BPD) at 36 weeks’ postmenstrual age. Secondary outcomes included BPD or death, moderate/severe BPD or death, respiratory support requirements, and neonatal morbidities. Conventional multivariable logistic regression and expanded gestational age- and birth weight-based propensity score models with stabilized inverse probability of treatment weighting (IPTW) were used. The expanded propensity score models incorporated maternal, placental, perinatal, and early respiratory variables, including first-hour invasive mechanical ventilation and surfactant administration within the first hour. Results: Caffeine was initiated within the first 2 h in 41 infants and after the first 2 h in 43 infants. BPD occurred in 10/41 (24.4%) and 24/40 (60.0%) infants, respectively, while BPD or death occurred in 10/41 (24.4%) and 27/43 (62.8%), respectively. In expanded IPTW analyses, caffeine initiation after the first 2 h remained associated with BPD in both the gestational age-based model (OR 3.40, 95% CI 1.22–9.48) and the birth weight-based model (OR 3.23, 95% CI 1.16–9.00). Corresponding ORs for BPD or death were 3.89 (95% CI 1.41–10.71) and 3.68 (95% CI 1.34–10.12). Additional adjustment for residual imbalance in pretreatment respiratory support produced similar estimates. Conclusions: Caffeine initiation after the first 2 h of life was associated with higher odds of BPD and BPD or death. These findings support further investigation of very early caffeine timing but do not establish a causal effect because of the observational design, modest sample size, and potential residual confounding. Full article

Low-light image enhancement (LLIE) is a fundamental task in computer vision, required for restoring luminance, contrast, and structural fidelity in images captured under suboptimal lighting environments. This paper introduces an optimization-driven, scene-adaptive LLIE framework, designated as $O{M}_{\mathrm{I}\mathrm{D}\mathrm{C}\mathrm{P}}^{\mathrm{G}\mathrm{C}\mathrm{G}\mathrm{L}}$, which exploits the optical duality between low-light inversion and atmospheric scattering. The proposed methodology transforms low-light inputs into quasi-haze representations through an optical inversion process, followed by structural restoration using an Improved Dark Channel Prior ($M_{\mathrm{I}\mathrm{D}\mathrm{C}\mathrm{P}}$) baseline. To refine the restored output, a Gamma Correction with Global Limits (GCGL) module is integrated as a boundary constraint to mitigate localized over-exposure and preserve chromatic consistency. A core novelty of this framework lies in the deployment of metaheuristic optimization algorithms (MOAs)—specifically the Gray Wolf Optimizer (GWO), Harris Hawks Optimization (HHO), and Marine Predators Algorithm (MPA)—to autonomously resolve optimal, image-specific parameter configurations. This search paradigm is guided by perception-driven fitness functions, namely the Patch-based Contrast Quality Index (PCQI) or the Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE). Quantitative and qualitative evaluations across a comprehensive pool of 1092 benchmark images demonstrate that the proposed framework exhibits robust statistical resilience and cross-dataset generalization compared to four state-of-the-art deep learning methods. While data-driven deep learning architectures retain localized superiority under the extreme degradation boundaries of the DARK FACE dataset, the proposed physics-inspired optimization framework achieves the leading overall cross-dataset aggregate ranking ($\overline{R}=2.467$) across diverse evaluation environments due to its per-image dynamic solution space mapping. Full article

Objective: We aimed to quantify the prognostic value of glucose monitoring-derived time in range (TIR), including continuous glucose monitoring (CGM), flash glucose monitoring (FGM), and fingertip capillary glucose monitoring (FCGM), for predicting adverse clinical outcomes in patients with type 2 diabetes mellitus (T2DM). Research Design and Methods: PubMed, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL, via OVID) were systematically searched from 2017 to November 2025 for studies evaluating the risk of all clinically relevant outcomes associated with different TIRs in T2DM. Extracted data were standardized to evaluate the effect of a 10% increment in TIR. Pooled estimates were calculated using inverse-variance random-effects models incorporating dose–response analysis. The certainty of evidence was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) framework. Results: Twenty-four observational studies involving 20 distinct associations and 35,916 participants were included. Dose–response meta-analyses were conducted for nine associations. The results showed that each 10% increment in TIR was significantly associated with a reduced risk of multiple adverse complications, including all-cause mortality (odd ratio [OR] = 0.88, 95% confidence interval [CI]: 0.82–0.93), vision-threatening diabetic retinopathy (OR = 0.93, 95% CI: 0.87–0.98), diabetic retinopathy (OR = 0.92, 95% CI: 0.89–0.95), lower extremity atherosclerotic disease (OR = 0.86, 95% CI: 0.82–0.91), diabetic peripheral neuropathy (OR = 0.77, 95% CI: 0.71–0.84), and cardiovascular autonomic neuropathy (OR = 0.81, 95% CI: 0.68–0.97). In contrast, the associations for albuminuria (KDIGO [Kidney Disease: Improving Global Outcomes] A2 and A3) and amputation did not reach statistical significance in the primary meta-analysis. Conclusions: In conclusion, each 10% increment in TIR is consistently associated with a reduced risk of mortality and various micro- and macrovascular complications in T2DM. These findings suggest TIR as a robust prognostic indicator and actionable therapeutic target in diabetes management. Full article

Optical inverse design faces a dilemma: neural surrogates enable fast, differentiable search but can yield physically unreliable pseudo-optima, whereas solver-in-the-loop optimization is reliable yet costly. Most surrogate methods also trust the surrogate throughout the search, train separate models for performance prediction and structure optimization, and remain largely black-box. We propose Physics-Verified Spectral Dreaming (PVSD), a unified framework for forward prediction, inverse design, and physical interpretability: a frozen differentiable spectral surrogate “dreams” structural mutations by input-gradient ascent to explore the design space, while a physical solver adjudicates every accepted update—the surrogate proposes, physics decides. We instantiate it as PVSD-TMM for one-dimensional multilayer radiative coolers. The forward predictor attains $R^2=0.9936/0.9964/0.9828$ for net cooling power, solar reflectance, and primary-window emissivity; neural dreaming lifts the population-mean net cooling power of 1000 random seeds from $-466.7$ to $65.8$ W m⁻² ($91.4\%$ reaching net cooling), and continuous-thickness refinement with 5 nm rounding yields a 14-layer manufacturable final design. Independent COMSOL finite-element and analytic TMM cross-validation converge to $P_{\mathrm{cool}}\approx 172$ W m⁻², $R_{\mathrm{solar}}\approx 0.970$, and ${\epsilon }_{\mathrm{win}}=0.9252$. This is a full-spectrum radiative-balance result for an idealized radiative-only case ($h_{\mathrm{conv}}=0$), not a window-emittance-only metric; PVSD thus achieves high simulated broadband radiative-cooling performance under the stated assumptions, without claiming global optimality. Full article

Background: Surgical sepsis complicated by acute kidney injury (AKI) is associated with high mortality, and early risk stratification remains challenging. The prognostic nutritional index (PNI), derived from serum albumin and lymphocyte count, reflects nutritional and immune status, but its prognostic value in surgical sepsis with AKI has not been well defined. Methods: In this retrospective cohort study based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) (version 3.1) database, we included adult patients admitted to a surgical or surgery-related intensive care unit (ICU) between 2008 and 2022 who met the Sepsis-3 criteria and developed AKI according to KDIGO. The primary endpoint was 90-day all-cause mortality. We used multivariable Cox proportional hazards models, restricted cubic splines, Kaplan–Meier analysis, and predefined subgroup analyses to examine the association between PNI at ICU admission and 90-day mortality. Results: A total of 1483 patients were included, with a 90-day mortality rate of 30.2%. Non-survivors had a lower median PNI than survivors (35 vs. 36, p < 0.001). After sequential adjustment for demographics, comorbidities, illness severity, and major interventions, each 1-point increase in PNI was associated with a 1.5% reduction in 90-day mortality (hazard ratio 0.985, 95% confidence interval 0.974–0.997, p = 0.012). Restricted cubic spline analysis showed an approximately linear inverse relationship between PNI and mortality risk (p for overall association = 0.046; p for nonlinearity = 0.534). Using an optimal cut-off of 29.51, patients with low PNI had significantly lower 90-day survival than those with high PNI (log-rank p < 0.0001), and subgroup analyses demonstrated generally consistent protective associations across age, sex, illness severity, and key treatments. In propensity score-matched analysis, the association was attenuated and no longer significant (HR = 1.174, 95% CI: 0.901–1.529, p = 0.234), although the direction of effect remained consistent. Conclusions: Lower PNI at ICU admission was associated with higher 90-day mortality in multivariable-adjusted models and may serve as a simple, routinely available adjunctive marker for early risk stratification in surgical sepsis patients with AKI. However, given the non-significant finding in propensity score-matched analysis, its independent prognostic value remains uncertain, and further prospective studies are needed to validate its clinical utility. Full article