Search Results (1,354)

The transition to autonomous driving is creating mixed traffic environments in which human-driven vehicles, partially automated vehicles, and autonomous vehicles must continuously interact, adapt, and respond to one another. This paper presents a comprehensive review of driver behavior in mixed traffic with autonomous vehicles, with emphasis on the sociotechnical nature of human–machine coexistence. The review synthesizes recent evidence on behavioral adaptation in car-following and tactical decision-making, trust calibration, situational awareness, takeover performance, internal and external human–machine interface design, surrogate safety metrics, vehicle-to-vehicle communication, operational design domains, and data-driven scenario generation. The literature shows that drivers do not respond to autonomous vehicles uniformly. Instead, behavior varies by driving style, perceived predictability of the automated vehicle, interface transparency, and traffic context. The review also emphasizes that these interaction patterns are context-dependent and may differ substantially across regions, particularly in dense mixed traffic environments. While some adaptations can improve stability and safety, others can encourage opportunistic maneuvers, overtrust, confusion, or degraded takeover quality. The review also highlights that crash data alone are insufficient to assess safety in mixed traffic, and that near-miss analysis, surrogate conflict metrics, and scenario-based evaluation are essential for understanding safety-critical interactions. Across the literature, a central inference emerges: adaptation to autonomous vehicles is real, but it is not automatically stabilizing. Safe deployment therefore depends not only on technical vehicle performance but also on behavioral legibility, transparent communication, calibrated trust, and robust evaluation under diverse real-world conditions. The paper concludes by identifying major research gaps, including the lack of longitudinal studies, incomplete standardization of surrogate metrics, limited understanding of vehicle conspicuity effects, and the need for integrated frameworks that jointly assess driver behavior, system design, and scenario-based safety. Full article

Manipulation-detection models calibrated in developed markets are routinely applied to frontier economies without validation, yet the institutional conditions that make such tools function as pricing signals are rarely present in those settings. This study provides the first systematic test of the Beneish M-Score and Dechow F-Score as return predictors in Vietnam, a frontier market navigating staged International Financial Reporting Standards (IFRS) convergence. Apparent negative associations between manipulation scores and excess returns under System Generalized Method of Moments (System GMM) do not survive panel fixed effects, Fama–MacBeth, or between-firm estimation. Persistent second-order serial correlation confirms that the GMM signal reflects frontier-market return momentum rather than manipulation pricing. By contrast, Big Four audit quality generates a robust cross-sectional return premium, establishing audit credibility as the operative governance channel where regulatory enforcement is absent. Survival analysis further shows that high-risk firms face substantially elevated exit hazards, demonstrating that reporting risk shapes long-run viability even where short-run pricing is absent. These findings constitute an IFRS paradox: Vietnam has adopted the institutional form of international reporting standards while lacking the informational infrastructure to support detection models that function as reliable pricing signals. Governance infrastructure, not standards convergence, is the operative condition for market discipline in frontier settings. Full article

Additive manufacturing enables the development of low-cost, self-built robotic systems; however, their performance is typically not characterized by validated metrics. The paper presents a photogrammetric concept intended for system-independent application to characterize planar positioning accuracy and repeatability without access to internal controller data. The method is based on a Raspberry Pi 4 camera system, image processing in Python 3.12.0 and OpenCV 4.12.0, and a universal additively manufactured robot tool attachment. Two position estimation strategies are investigated: a marker-based approach using ArUco markers and a markerless blob-analysis method based on a ruby sphere. Camera calibration is evaluated using different patterns, with a compact CharUco board exhibiting the lowest RMS reprojection error (~1 px). Experimental validation follows selected elements of ISO 9283:1998 and comprises 30 repetitions at five target poses for linear and axial motion strategies. The results show lower positional deviations for marker-based methods compared to the markerless approach, with a two-marker configuration yielding the lowest mean deviation under the investigated conditions. Sub-millimeter positioning accuracy and repeatability are achieved, and linear motion exhibits lower repeatability deviations than axial motion. The proposed approach provides a cost-effective and flexible solution for external robot characterization, particularly suited for self-built and resource-constrained systems. Full article

Background/Objectives: Infective endocarditis (IE) carries substantial mortality, particularly in middle-income settings where patient profiles and microbial ecology differ from those of cohorts used to derive international prognostic scores. Syndrome-specific, locally grounded decision aids for empirical therapy are also scarce. We aimed to identify predictors of in-hospital mortality, externally evaluate the RiskE and ICE scores, and construct a Bayesian weighted-incidence syndromic combination antibiogram (WISCA) for IE. Methods: We conducted a retrospective cohort study of consecutive adults with definite or possible IE admitted between January 2019 and January 2026. Candidate predictors were screened in two phases, and a clinically specified model was estimated with maximum-likelihood and Firth penalization, with 1000-replicate bootstrap optimism correction. Calibration was assessed with bootstrap calibration plots and the Hosmer–Lemeshow test. Discrimination was compared against RiskE and ICE using DeLong’s test and reclassification metrics. For empirical coverage, we built a WISCA using identified pathogens, reporting both non-Bayesian bootstrap estimates and Bayesian hierarchical partial-pooling estimates with species- and antibiotic-level random intercepts; analyses were also stratified by IE type. Results: In-hospital mortality was 22.9% in a young cohort (median 37 years) characterized by high hemodialysis prevalence (47.4%), substantial right-sided IE (46.4%), and Staphylococcus aureus predominance (32%) with no methicillin-resistant isolates. Vasopressor-requiring shock (Firth OR 9.23, 95% CI 2.40–40.61) and acute heart failure (OR 10.01, 95% CI 2.78–41.07) were the strongest predictors; the final model achieved an AUC of 0.922 (optimism-corrected 0.908), significantly outperforming RiskE (0.598) and ICE (0.632). The Bayesian WISCA identified multiple carbapenem-sparing and anti-MRSA–sparing regimens with adequate coverage (≥80%), particularly for community-acquired IE, supporting stewardship-oriented empirical selection. Coverage was consistently lower in healthcare-associated IE. Conclusions: A parsimonious three-variable model provided strong, locally valid mortality prediction in this hemodialysis-predominant, MRSA-free cohort, substantially outperforming European-derived scores. External validation in independent cohorts is required before clinical adoption. The Bayesian WISCA demonstrated that adequate empirical coverage is achievable without routine broad-spectrum agents, offering institution-specific guidance for stewardship-compatible regimen selection; multicenter validation is warranted. Full article

Objective: To identify risk factors for progression to sepsis in patients with non-ventilator hospital-acquired pneumonia (NV-HAP) and to develop a practical nomogram for individualized risk assessment in this population. Methods: We retrospectively screened 408 hospitalized patients with hospital-acquired pneumonia at Peking University Third Hospital between January 2017 and December 2021. After excluding patients with an unclear diagnosis date or missing critical variables required for SOFA score calculation, 368 eligible patients with NV-HAP were included and randomly divided into a training cohort (n = 260) and an internal validation cohort (n = 108). An independent temporal validation cohort of 68 patients admitted between January 2022 and December 2022 at the same center was further used for temporal validation. Univariable and multivariable logistic regression analyses with backward stepwise selection were performed in the training cohort to identify predictors associated with progression to sepsis. A nomogram was then constructed based on the final model and evaluated by discrimination, calibration, and decision curve analysis. Results: A total of 368 patients were included in the model development dataset. The final multivariable model retained six predictors: male sex (OR = 2.393, 95% CI: 1.333–4.296), diabetes (OR = 2.205, 95% CI: 1.126–4.319), coagulation dysfunction (OR = 3.327, 95% CI: 1.726–6.413), PaO₂/FiO₂ (OR = 0.955 per 10-unit increase, 95% CI: 0.912–1.001), platelet count (OR = 0.900 per 10 × 10⁹/L increase, 95% CI: 0.853–0.949), and bilirubin (OR = 1.176 per 1 μmol/L increase, 95% CI: 1.100–1.258). The nomogram showed acceptable performance, with an apparent C-index of 0.809 and a bootstrap-corrected C-index of 0.792 in the training cohort. The C-index was 0.750 (95% CI: 0.658–0.841) in the internal validation cohort and 0.754 (95% CI: 0.639–0.870) in the temporal validation cohort. Calibration analysis showed acceptable agreement between predicted and observed probabilities, and decision curve analysis indicated a positive net clinical benefit across clinically relevant threshold probabilities. Conclusions: In patients with NV-HAP, male sex, diabetes, coagulation dysfunction, lower PaO₂/FiO₂, lower platelet count, and higher bilirubin were associated with progression to sepsis. The developed nomogram showed acceptable discrimination, calibration, and clinical utility, and may serve as a practical tool for early individualized risk stratification in patients with NV-HAP. Full article

Background/Objectives: Cranial gunshot injuries represent severe traumatic brain injuries associated with high mortality rates. This study investigated whether integrating clinical findings at admission, including GCS score and pupillary response, with a CT-derived anatomical burden score and midline shift improves the prediction of in-hospital mortality. Methods: Adult patients aged 18 years and older with penetrating cranial gunshot injuries (n = 143) treated at a tertiary referral centre between 1 January 2005 and 31 December 2025 were retrospectively analysed using a single-centre cohort design. All included patients completed in-hospital follow-up, defined as hospital discharge or in-hospital death. Clinical variables, the anatomical burden score, and midline shift were evaluated using a multivariable logistic regression model where the primary outcome was in-hospital mortality. Model performance was assessed using ROC analysis, calibration measures, and bootstrap internal validation. Results: The in-hospital mortality rate was 56.6%, with early mortality occurring in 33.6% of patients. In the multivariable analysis, a low admission GCS score (≤8), bilateral non-reactive pupils, an increased anatomical burden score, and midline shift were independently associated with a higher risk of mortality. The model demonstrated good discrimination (AUC = 0.87; 95% CI 0.81–0.93), and similar performance was maintained following internal validation (optimism-corrected AUC = 0.86). The addition of radiological parameters to clinical variables improved model discrimination (ΔAUC = 0.07; 95% CI 0.02–0.11). Conclusions: The combined evaluation of admission clinical findings and CT-based anatomical parameters may support a more structured early estimation of in-hospital mortality risk in adult patients with penetrating cranial gunshot injuries. Full article

While Environmental, Social, and Governance (ESG) rating divergence poses a barrier to accurate sustainability measurement and sustainable investment, how internal managerial cognition addresses this external market misalignment remains underexplored. To address the research question of how executive focus shapes market consensus on corporate sustainability, this study integrates the Attention-Based View and Signaling Theory to examine the potential mitigating role of Top Management Team (TMT) environmental attention on ESG rating divergence. Utilizing high-dimensional fixed-effects regressions and textual analysis, we analyze a sample of Chinese A-share non-financial listed firms from 2015 to 2023. Empirical results indicate that a transparent and forthcoming managerial environmental focus helps reduce rating divergence, thereby partially aligning informational baselines. This cognitive alignment can act as an information calibrator, particularly when environmental issues match the firm’s core industry materiality, and this association appears more pronounced in regions with stringent environmental regulations. Robustness checks support the notion that substantive, quantitative sustainability disclosures driven by executive attention assist in alleviating informational misalignment among external rating agencies. These findings offer socio-economic and policy insights for advancing sustainable development, suggesting that regulators could consider encouraging structured sustainability reporting to support the role of executive cognition in standardizing ESG measurements. Full article

Since 2007, the porosity–to–cement relationship has been widely used as a unified parameter to predict mechanical strength, durability, expansion, and stiffness of stabilized soils. In this formulation, the volumetric binder content is adjusted by an internal exponent x, typically ranging between 0 and 1, to balance the relative contributions of porosity and cementation. Traditionally, the parameters of this relationship have been obtained using manual regression procedures. This study proposes an automated calibration methodology for the porosity–binder index, where the parameters A, B, and x are determined through an iterative optimization framework based on minimization of the sum of absolute errors (SAE) combined with a Monte Carlo search algorithm. The methodology is applied to a cement-stabilized clay blended with ground glass (GG), recycled gypsum (GY), and limestone residues (CLW). The predictive capability of the calibrated model is evaluated using unconfined compressive strength (q_u) and initial shear stiffness (Go) datasets. Two calibration strategies are considered: Calibration Process No. 1, based on CLW mixtures and q_u values only, and Calibration Process No. 2, incorporating all mixtures (CLW, GG, and GY) and both q_u and Go responses. The results indicate that Calibration Process No. 2 provides a more robust and physically consistent parameter set, yielding coefficients of determination of 0.9318 and 0.9412 for q_u and Go, respectively. The proposed algorithm-driven calibration framework improves predictive capability and provides a systematic approach for determining the parameters of the porosity–binder relationship. Full article

Background/Objectives: This study aimed to investigate the impact of allergic diseases (AD) or obstructive sleep apnea (OSA) risk, as a host factor, on the development of severe Mycoplasma pneumoniae Pneumonia (SMPP) in children by analyzing the clinical data of pediatric patients with Mycoplasma pneumoniae Pneumonia (MPP). Methods: This retrospective study enrolled children hospitalized with Mycoplasma pneumoniae pneumonia (MPP) at Shanghai Ninth People’s Hospital from November 2024 to November 2025. Patients were classified into severe (SMPP) and mild (MMPP) groups. Demographic, clinical, laboratory, and questionnaire data were collected and compared between groups. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of SMPP and construct a nomogram. The model was validated for discrimination, calibration, and clinical utility using ROC curves, calibration plots, and decision curve analysis, with internal validation by bootstrap resampling. Results: Among the 150 enrolled children with MPP, 35 (23.3%) were classified as severe (SMPP) and 115 (76.7%) as mild (MMPP). Patients with SMPP exhibited significantly higher frequencies of allergic diseases, prolonged fever and steroid use, elevated inflammatory markers (CRP, LDH, D-dimer, ferritin, ALT), and higher PSQ and RQLQ scores (all p < 0.05). Disease severity was positively correlated with these clinical, laboratory, and questionnaire-based parameters. Multivariate logistic regression identified allergic diseases, PSQ score, LDH, and ferritin as independent predictors of SMPP. A nomogram incorporating these four factors demonstrated good predictive performance, with an internally validated C-index of 0.827, satisfactory calibration (Hosmer–Lemeshow p = 0.116), and clinical utility within a 0–25% threshold probability range on decision curve analysis. Conclusions: Children with MPP and comorbid AD or OSA risk are more likely to develop SMPP. Among children aged 6–12 years, RQLQ score is positively correlated with the severity of MPP. AD, PSQ score, LDH, and ferritin are independent risk factors for SMPP. Clinicians should be alert to the development of SMPP when children with MPP present with a history of AD, PSQ score >3.5, LDH >327.50 U/L, or ferritin >120.05 ng/mL. The visual nomogram model constructed by combining these risk factors demonstrates improved predictive performance for SMPP, with high predictive efficacy and accuracy. It has great clinical value and can be used for individualized risk assessment and early intervention. However, our proposed nomogram requires external validation prior to broader implementation. Full article

This study presents a comparative evaluation of solar cabinet drying and traditional open-air sun drying with respect to their influence on the retention of water-soluble vitamins and flavonoids in Plantago major L. leaves, aiming to identify an effective and sustainable drying strategy for functional food applications. Freshly harvested leaves were subjected to both drying methods under comparable environmental conditions. To account for possible structural heterogeneity, external and internal leaf tissues were analyzed separately. Qualitative and quantitative determination of bioactive compounds was performed using high-performance liquid chromatography with diode-array detection (HPLC-DAD). Flavonoids were analyzed at detection wavelengths of 254 and 276 nm, while water-soluble vitamins (C, B₂, B₃, B₆, and B₉) were determined at 250 nm. Quantification was carried out using external calibration, and results were expressed as concentrations (mg/g dry matter). The results demonstrate that solar cabinet drying provides superior preservation of oxidation- and light-sensitive bioactive compounds compared to open-air sun drying. In particular, vitamin C content in solar cabinet-dried samples reached 91.62 mg/g, which was more than three times higher than that observed after open-air drying (26.90 mg/g). Solar cabinet drying also enhanced the retention of key antioxidant flavonoids, notably dihydroquercetin (14.23 mg/g vs. 11.21 mg/g) and luteolin (0.38 mg/g vs. 0.26 mg/g). Although slightly higher concentrations of certain compounds, such as rutin and vitamins B₆ and B₉, were detected in open-air-dried samples, the overall nutraceutical profile favored solar cabinet drying. In conclusion, the controlled microclimate of the solar cabinet dryer significantly improves the stability and retention of critical water-soluble vitamins and antioxidant flavonoids in Plantago major L. leaves. These findings confirm that solar cabinet drying is a nutritionally advantageous, energy-efficient, and sustainable approach for producing high-quality plant-based ingredients suitable for functional food and nutraceutical applications. Full article

Background/Objectives: Uroflowmetry is done in the surveillance period after End-to-end Anastomotic Urethroplasty for pelvic fracture urethral injury. But is maximum flow rate a reliable surrogate for urethral calibre in these cases? The above question laid the foundation of the study. The aim of the study was: “Is uroflowmetry alone sufficient to predict a successful outcome following urethroplasty after pelvic fracture urethral injury (PFUI)?” Methods: We conducted a prospective masked study of all patients undergoing end-to-end anastomosis (EEA) urethroplasty for PFUI from January 2017 to September 2022. The first follow-up was 4 weeks after surgery, micturating cystourethrogram (MCU) was done after urethral catheter removal and at the same time, uroflowmetry was also done. The second follow-up was 6 months after surgery, when uroflowmetry was repeated, and urethroscopy was performed. The urologist performing urethroscopy was blinded to the uroflowmetry results. Results: In total, 26 patients were included in the study. After 6 months, 1 patient had poor flow (maximum flow rate [Q max] < 10 mL/s), 7 patients had flow with Q max 10–15 mL/s, and 18 patients had normal flow (Q max > 15 mL/s). On urethroscopy, all patients had a normal and easily passable urethra. The International Prostate Symptom Score (IPSS) and quality of life (QoL) scores showed a positive correlation. The urologist performing urethroscopy and the investigator recording uroflowmetry reached different conclusions. Conclusions: A reduced peak on uroflowmetry after EEA urethroplasty in PFUI does not always indicate surgical failure. Urethroscopy enables direct visualisation of the anastomotic site and provides more detailed information than uroflowmetry. The IPSS score and quality of life are more important than Q max alone. Full article

The automated production of large-scale labeled datasets from concrete X-ray computed tomography (CT) images is a fundamental prerequisite for training and validating deep learning-based segmentation models. However, existing methods either require extensive manual annotation or rely on domain-specific deep learning models that themselves demand labeled data—a circular dependency. This paper presents a parameter-driven three-class segmentation framework that automatically classifies each pixel in a concrete CT slice into one of three material phases: void (air pores and cracks), coarse aggregate, and mortar matrix, generating annotation masks suitable for large-scale dataset production without manual labeling. The proposed method combines: (1) fixed-threshold void detection calibrated to concrete CT grayscale characteristics; (2) adaptive percentile-based initial segmentation responsive to image-specific statistics; (3) multi-criteria connected component scoring based on area, shape descriptors (circularity, solidity, compactness, extent, aspect ratio), intensity distribution, and boundary gradient; (4) material science-informed size constraints aligned with concrete phase volume fractions; and (5) a material continuity enforcement module that applies topological hole-filling and conditional convex-hull consolidation to eliminate internal contamination within accepted aggregate regions, reducing boundary roughness by 7.6% and recovering misclassified boundary pixels. All parameters are centralized in a configuration file, enabling reproducible batch processing of 224 × 224 pixel CT slices at 0.07–1.12 s per image. Evaluated on 1007 224 × 224 concrete CT patches cropped from 200 representative scan frames, the framework produces three-class segmentation masks with physically consistent void fractions (mean 3.2%), aggregate fractions (mean 32.4%), and mortar fractions (mean 64.4%), all within ranges reported in the concrete CT literature (used as a dataset-scale QC screen, not a validation metric). Primary outputs and the archived image–mask pairs for this work are provided as an 8-bit patch archive. For pixel-wise validation, we report IoU, Dice, and pixel accuracy on an independently labeled subset that can be unambiguously paired with the released predictions: averaged over 57 matched patches, mean pixel accuracy is 88.6%, macro-mean IoU is 74.7%, and macro-mean Dice is 84.9%. The framework provides a fully automated annotation pipeline for dataset production, eliminating manual labeling costs for concrete CT image collections. The generated datasets are suitable for training semantic segmentation networks such as U-Net and its variants. Full article

Background and Objectives: To identify predictors of pulmonary infection in critically ill patients after abdominal surgery and to develop an early postoperative risk stratification model. Materials and Methods: Medical records of ICU patients after abdominal surgery (January 2016–June 2024) with Acute Physiology and Chronic Health Evaluation II (APACHE II) scores ≥10 were retrospectively analyzed. Patients were categorized according to the presence or absence of pulmonary infection. Candidate variables were screened using LASSO regression, followed by multivariate logistic regression to identify independent predictors. A nomogram-based prediction model was constructed and internally validated. Results: Among 4852 patients, 390 (8.0%) developed pulmonary infections. Overall, 8 independent predictors were identified: Male sex (vs. female) (OR 1.509, 95% CI: 1.091–2.087, p = 0.013), chronic obstructive pulmonary disease (OR 4.139, 95% CI: 2.872–5.966, p < 0.001), atrial fibrillation (OR 2.320, 95% CI: 1.366–3.939, p = 0.002), hypertension (OR 1.869, 95% CI: 1.372–2.539, p < 0.001), chronic renal insufficiency (OR 2.412, 95% CI: 1.143–5.091, p = 0.021), preoperative total bilirubin (OR 1.003, 95% CI: 1.001–1.004, p = 0.002), rectal surgery (OR 0.354, 95% CI: 0.151–0.830, p = 0.017), and invasive mechanical ventilation duration > 6 h (OR 2.206, 95% CI: 1.628–2.990, p < 0.001). The nomogram demonstrated good discrimination (AUC: 0.734 95% CI: 0.698–0.770) and calibration. Conclusions: This study identified 8 independent predictors of pulmonary infection and developed an internally validated early postoperative risk stratification model with satisfactory performance. The model may assist clinicians in identifying high-risk patients and guiding timely preventive strategies in ICU practice. Full article

Internal thermal load fluctuations and variations in occupant density affect the performance of Hybrid Geothermal Heat Pump (HGHP) systems. Traditional control strategies cannot provide the rapid adjustments needed to operate efficiently in real time and can be inefficient, leading to increased energy consumption and reduced thermal comfort. A data-driven fuzzy logic control framework is developed in this paper to dynamically adjust the performance of an HGHP system in real time as a function of occupancy and environmental conditions (e.g., temperature and humidity differences). The controller analyzes input data related to real-time outdoor ambient conditions like temperature, humidity and occupied spaces; a real-time flow sensor attached to the occupants of the building (a count of the number of occupants currently in each occupied space); and the coefficient of performance (COP) of the HGHP system, and uses the analysis to generate a “smart” control decision for the following device types: variable speed drive (VSD), fan number, operating modes, system control and valve positions. The controller also controls the overall system. The model was developed and simulated in MATLAB Simulink^®, with realistic system parameters, and validated and calibrated using operational data from an HGHP system at a university, based on operating conditions. The simulation results indicate that our fuzzy controller achieves higher energy efficiency for thermal comfort than traditional thermostat-based controls, with COP improvements ranging from 7.36% to 11.76% and power consumption reductions between 4.13% and 8.55% across various occupancy scenarios. The improved COP also demonstrates the device’s responsiveness and effectiveness, even under frequent changes in occupancy patterns (dynamic occupancy), making it suitable for use in automated climate control systems in modern buildings. Full article

Cardiovascular disease (CVD) risk assessment in older adults requires models that are accurate, clinically interpretable, and able to retain performance in independent populations. This study developed an interpretable machine-learning framework for CVD risk stratification in individuals aged 65 years and older using routinely available clinical factors and a selected biochemical extension and then evaluated its performance in a substantially larger independent external cohort. Model development used a development cohort of 100 patients (Almaty, age ≥ 65) with leakage-free nested cross-validation and out-of-fold (OOF) probabilities. Three internally evaluated configurations were compared: a clinical logistic regression baseline (LR clinical), a biomarker-augmented logistic regression (LR selected), and a nonlinear random forest on the selected feature set (RF selected). Discrimination was assessed using ROC-AUC and PR-AUC; probabilistic accuracy using Brier score and log loss. Calibration was examined using OOF calibration curves with sigmoid calibration for selected models. Decision-analytic utility and exploratory operational thresholds were assessed using Decision Curve Analysis (DCA), yielding a three-tier scale with thresholds t__low = 0.23 and t__high = 0.40. In nested cross-validation, LR clinical achieved ROC-AUC 0.9425 ± 0.0188 and PR-AUC 0.9574 ± 0.0092 with Brier 0.1004 ± 0.0215 and log loss 0.3634 ± 0.0652; LR selected performed worse, while RF selected showed competitive discrimination. External validation on an independent cohort (n = 695) showed retained discrimination (ROC-AUC 0.8355; PR-AUC 0.9376) with acceptable probabilistic accuracy (Brier 0.1131; log loss 0.3760), and recalibration (intercept + slope) slightly improved probability metrics. Explainability analyses (odds ratios, permutation importance, SHAP) consistently identified heredity, BMI, physical activity, and diabetes as influential model-associated factors, with clinically plausible directionality. The results suggest that an interpretable model trained on a small geriatric cohort can retain meaningful predictive performance on a substantially larger external cohort, supporting the potential value of transparent risk stratification in older adults, while broader prospective and multi-center validation remains necessary before routine clinical implementation. Full article