Search Results (15,462)

Background: Pre-travel health consultations require individualised risk assessment across itinerary, destination, traveller characteristics, vaccine and medication history, comorbidities, pregnancy and immune status, activities, and access to care. Artificial intelligence (AI), particularly large language models (LLMs), may support pre-consultation education, structured history collection, guideline retrieval, multilingual communication and post-consultation reinforcement, but unsafe use may introduce hallucinated, outdated or insufficiently personalised recommendations. Objectives: This scoping review maps the current evidence on AI tools relevant to pre-travel health consultations, characterises implementation gaps, identifies patient-safety risks and proposes a supervised implementation model for travel medicine clinics. Original contribution: Unlike previous reviews of clinical AI, patient-education LLMs or chatbots in chronic illness, this is the first scoping review focused specifically on AI in pre-travel consultations. It uniquely combines a five-tier evidence hierarchy that separates direct travel-medicine AI evidence from indirect clinical-AI safety and equity evidence, and provides a travel-medicine-specific clinical safety risk taxonomy and a supervised implementation framework anchored to authoritative travel-medicine guidance and current AI regulatory regimes. Methods: A scoping review was conducted following PRISMA-ScR reporting, using a Population–Concept–Context eligibility framework and a targeted retrieval in May 2026 covering January 2017 to May 2026. Sources were screened and charted by a single reviewer using a structured eligibility checklist. Quality and applicability were appraised conceptually using MMAT, AMSTAR 2 and JBI text-and-opinion criteria, with GRADE-informed certainty. Results: Of 70 records identified, 11 were included: four direct pre-travel AI sources, one adjacent travel-related decision-support study, four guideline and context sources and two clinical LLM safety sources. The only patient-level implementation involved 26 travellers using a GPT-4 Travel Clinic Assistant in Singapore, where physicians and travellers reported acceptability and workflow benefit but objective effectiveness outcomes were not measured. Broader clinical LLM evidence indicates heterogeneous evaluation methods, vulnerability to hallucinated guidelines, and accuracy that varies widely across model versions and specialties. Conclusions: Current evidence supports supervised AI augmentation of pre-travel consultations but does not support autonomous AI-led vaccine selection, malaria prophylaxis, contraindication screening or individualised travel-risk clearance. Near-term deployment should be restricted to clinician-supervised education, structured intake, source-grounded guideline retrieval, after-visit reinforcement and escalation-triggered workflow support. Priority research includes travel-medicine-specific hallucination audits; equity testing in visiting-friends-and-relatives, migrant, older-adult, First Nations Australian, and Pacific Islander travellers; and prospective trials reported under CONSORT-AI, SPIRIT-AI and TRIPOD + AI. Full article

Software defect prediction (SDP) aims to identify defect-prone modules before testing, but severe class imbalance and redundant software metrics often limit prediction performance. Many conventional feature selection methods estimate feature relevance with the original imbalanced empirical distribution and mainly emphasize marginal relevance or global classifier-oriented criteria, which may under-prioritize features that are informative for the minority defective class. To address this issue, this paper proposes SCAR-CMB, a simplified class-reweighted and interaction-aware feature selection method for imbalanced SDP. SCAR-CMB estimates feature-label dependency with a class-balanced empirical distribution, controls redundancy using weighted conditional dependency information, and incorporates an interaction-aware conditional-gain term as an auxiliary re-prioritization signal within a relevance-screened feature pool. Rather than performing full causal structure discovery or formal synergy estimation, SCAR-CMB adopts a Markov-blanket-inspired conditional dependency design as a practical guide for feature selection. The final configuration excludes both hardness-aware weighting and false discovery rate filtering. SCAR-CMB is evaluated on ten public NASA and PROMISE defect datasets under a leakage-free cross-validation protocol. Compared with seven representative baselines, SCAR-CMB achieves competitive overall performance and obtains the highest average defective-class recall, G-mean, and balanced accuracy. However, it is not uniformly superior across all metrics, and the recall advantage is not confirmed by the omnibus Friedman test. Additional mechanism-level, stability, and sensitivity analyses show that class reweighting changes feature prioritization, the selected feature subsets are relatively stable across folds, and the interaction-aware term provides limited and dataset-dependent auxiliary effects. Sensitivity analyses further indicate that the main conclusions are not solely determined by a specific feature budget, discretization-bin setting, or downstream classifier. Overall, SCAR-CMB should be interpreted as a practical minority-class-oriented feature selection method that provides a trade-off among defective-class detection, feature subset control, and computational cost. Full article

Fermented foods and beverages based on plant-derived ingredients are of growing technological interest, especially when they are designed as alternatives to conventional dairy products. This study evaluated the effects of herbal extracts from Verbascum thapsus L., Cnicus benedictus L., and Fumaria officinalis L. on lactic acid bacteria growth, acidification kinetics, and viable cell counts during the fermentation of organic milk, coconut beverage, and soy beverage. The extracts were characterized for extraction yield, total phenolic content, and antioxidant activity before use in fermentation trials. Mixtures of organic solvents and water produced extracts with higher total phenolic content and antioxidant activity than water alone. The highest values were obtained for F. officinalis extracts prepared with water and methanol or water and acetone, while for C. benedictus, the most effective solvents were water and acetone or water and ethanol. The agar well-diffusion assay showed no relevant antibacterial activity against the tested LAB strains under the applied conditions. No biologically relevant inhibition zones were observed in any of the 84 extract-strain combinations under the tested conditions. The only borderline response was observed for Lactobacillus acidophilus La-14 exposed to the 70% ethanolic extract of C. benedictus. The clear halo did not exceed 1.50 mm outside the 5 mm well and was treated as a weak, strain-specific screening result. Fermentation kinetics depended mainly on the food matrix. The coconut beverage acidified most rapidly, reaching pH 4.38 to 4.79 after 6 h, whereas the soy beverage required 24 h to reach pH 4.31 to 4.56. Organic milk showed the slowest acidification, and selected C. benedictus extracts delayed pH reduction. All analyzed fermented samples contained more than 7 log CFU/mL of viable LAB. These results indicate that selected herbal extracts can be used in fermented milk and plant-based beverages without reducing LAB survival, but their suitability should be assessed separately for each strain and matrix. Full article

Foodborne diseases remain a major global public health concern. According to the World Health Organization (WHO), contaminated food causes nearly 500,000 deaths annually. Consequently, the development of novel strategies to control foodborne pathogens and prevent microorganism-induced plant diseases remains an important research priority, attracting considerable scientific attention worldwide. This study describes the design and synthesis of novel dihydropyrazolines containing an ethylpiperazine moiety as potential antimicrobial agents. The antibacterial activity of the newly synthesized heterocyclic compounds, along with their corresponding chalcone precursors, was evaluated against Gram-positive and Gram-negative foodborne pathogens and phytopathogenic bacteria. The assessment was performed using a two-step protocol comprising an initial qualitative screening of selected bacterial species, followed by a quantitative evaluation of the inhibitory effects of individual compounds and their combinations. Among the tested compounds, pyrazolines 4a and 4b and chalcone 3b exhibited notable strain-dependent antibacterial activity, particularly against phytopathogenic strains of Pseudomonas syringae, while compound 4a demonstrated the highest efficacy against the Gram-negative foodborne pathogens Escherichia coli, Salmonella enterica and Listeria monocytogenes. Thus, the potential of the N-ethylpiperazine moiety as a key structural feature contributing to the antimicrobial activity of the studied compounds is revealed. Full article

Autonomous driving sensor fusion housings (SFHs) integrate LiDAR, radar, camera, and computing modules within a shared mechanical and thermal enclosure. This review examines how coefficient of thermal expansion (CTE) mismatch among housing polymers, aluminum heat spreaders, substrates, and solder joints can contribute to interfacial delamination, solder joint fatigue, optical misalignment, and Thermomechanical Coupling Interference (TMCI). Using a structured narrative review of 99 publications and authoritative standards from primarily 2009 to 2026, the article organizes the evidence into a 4 × 4 taxonomy linking four failure mechanisms with experimental, computational, AI/ML, and qualification-oriented approaches. The review explicitly distinguishes direct literature evidence, transferred package-level evidence, model-based extrapolation, and author-derived conceptual estimates. Accordingly, TMCI temperature increments, sensor spacing values, optical drift estimates, and lifetime projections are discussed only as case-specific screening-level hypotheses unless directly validated in the cited literature. Five research gaps are identified: standardized multi-sensor TMCI validation, aging-corrected material and solder fatigue databases, long-term qualification of thermally conductive nanocomposites, SFH-specific validation of physics-informed digital twins, and integrated multi-failure testing. The contribution of this article is therefore primarily structural and agenda setting: it clarifies what is supported by direct evidence, what is transferred from adjacent domains, and what remains to be validated before robust SFH-level reliability guidance can be established. Full article

Traditional methods for screening high-melanin-yielding Auricularia heimuer strains are time-consuming and environmentally unfriendly. To address this issue, fifteen A. heimuer strains were used to determine the mycelial biomass, absorbance at 500 nm, CIE L*a*b* colorimetric values, and melanin yield of the fermentation broth. Pearson correlation analysis was performed to clarify the correlations among these indicators, and a regression equation was fitted to establish a rapid screening method. A total of 84 A. heimuer strains were used to verify this method, of which one high-melanin-yielding strain was obtained. The structural characterization and in vitro antioxidant activities of A. heimuer melanin (AHM) were determined. The results showed that the melanin yields of fifteen A. heimuer strains were extremely significantly positively correlated with absorbance at 500 nm (r = 0.880, p < 0.01). The fitted linear regression equation was Y = 0.0246X + 0.00094 (R² = 0.8756, p < 0.01). When 84 tested strains were investigated with this method, 8 strains (53.33%) exhibited relative differences below 10%, which is consistent with the satisfactory accuracy of the absorbance-based screening method. Finally, a high-melanin-yielding strain HMCC50028 was obtained, with a melanin yield of 0.0540 g/100 mL. The results of UV-Vis spectroscopy, Fourier-transform infrared (FTIR) spectroscopy, and scanning electron microscopy (SEM) of AHM indicated that the melanin exhibited structural characteristics consistent with fungal melanins, belonging to the natural melanin family. In vitro assays demonstrated that AHM possessed excellent superoxide anion radical scavenging activity and ferric reducing power. Full article

Background/Objectives: Acute type A aortic dissection (ATAAD) has high preoperative mortality, and an interpretable multivariable model based on clinically accessible data is crucial for clinical risk stratification. Methods: The data for this study were obtained from West China Hospital, Sichuan University. Centroid regression was used to construct the predictive model, with logistic regression, classification and regression tree, explainable boosting machine and extreme gradient boosting as the reference. Variables were screened by iterative selection, the literature review and clinical experience. Model performance was evaluated by accuracy, sensitivity, precision, Youden’s index, AUROC and AUPRC. Results: The vital signs and tests of 361 ATAAD patients during the first 24 h of their first admission were included in the final analysis. Centroid regression outperformed logistic regression, achieving accuracy (90.7% vs. 81.5%), sensitivity (0.813 vs. 0.741), specificity (0.983 vs. 0.900), Youden’s index (0.796 vs. 0.641), AUROC (area under the receiver operating characteristic curve, 0.953 vs. 0.843) and AUPRC (area under the precision–recall curve, 0.978 vs. 0.863) in the test set. It revealed that the use of α-blocker (the weights w = −1.20) and hydrochlorothiazide (w = −1.20), clinical features like dyspnea (w = −0.94), chest pain (w = 0.91) and lactate dehydrogenase (w = −0.95) were variables that had the greatest impact on model prediction. Conclusions: The centroid regression model not only has relatively high predictive performance and interpretability but also can be easily implemented in hospital systems to provide a practical and cost-effective tool for ATAAD preoperative risk stratification. Full article

Essential oils are multicomponent natural products with potential neurobehavioral activity, but integrated comparative studies remain limited. This study compared the essential oils of Satureja brevicalyx, Peperomia dolabriformis, and Rosmarinus officinalis in relation to their chemical profiles, predicted target interactions, preliminary acute oral safety, anxiolytic-like and antidepressant-like effects, antagonist-sensitive behavioral patterns, and exploratory serum biomarkers. Oils were characterized by GC-MS, and their constituents were screened by molecular docking against anxiety-, depression-, sleep-, and stress-related targets. Independent cohorts of male BALB/c mice received oral essential oils (25–100 mg/kg) and were assessed in anxiety-related, depression-related, and locomotor behavioral paradigms, including the elevated plus maze, light–dark box, marble burying, tail suspension, forced swim, and open field tests. Flumazenil and WAY-100635 were used to examine whether the behavioral responses were sensitive to γ-aminobutyric acid type A (GABA-A)/benzodiazepine- and serotonin 1A (5-HT1A)-related pharmacological modulation, respectively. In a preliminary 24-h acute oral toxicity screen, no mortality was observed up to 5000 mg/kg. The three oils produced anxiolytic-like and antidepressant-like effects without reducing spontaneous locomotor activity. Within its experimental block, S. brevicalyx showed the most consistent flumazenil-sensitive anxiolytic-like pattern and FDR-significant reductions in corticosterone and TNF-α, together with increased IL-4. P. dolabriformis showed a broader predicted multitarget docking profile and antagonist-sensitive behavioral attenuation compatible with mixed pathway participation. R. officinalis produced significant but more moderate behavioral effects. WAY-100635 partially attenuated the antidepressant-like effects of all three oils. These findings support differentiated but convergent functional neurobehavioral profiles among the oils. The docking, antagonist, and biomarker results should be interpreted as hypothesis-generating evidence of possible pathway involvement, supporting further validation in chronic stress models, receptor-specific assays, pharmacokinetic studies, and expanded safety evaluations. Full article

Background: This study compared the effects of mindfulness training (MT) and cognitive-behavioral training (CBT) on ADHD symptoms, test anxiety, and self-concept in children with ADHD and explored potential pathways of change associated with the two interventions. Methods: After a two-stage screening and eligibility confirmation procedure, 63 children with ADHD were included in the final analyses, including 19 in the MT group, 25 in the CBT group, and 19 in the control group. The MT and CBT groups received an eight-week intervention, whereas the control group continued regular classroom activities. ADHD symptoms, test anxiety, and self-concept were assessed before and after the intervention. Data were analyzed using 3 × 2 repeated-measures ANOVA and exploratory cross-lagged analyses. Results: Significant Group × Time interactions were found for ADHD symptoms, F(2, 60) = 29.74, p < 0.001, η_p² = 0.5; test anxiety, F(2, 60) = 7.77, p = 0.001, η_p² = 0.21; and self-concept, F(2, 60) = 18.9, p < 0.001, η_p² = 0.39. Simple effects analyses showed that both the MT and CBT groups showed significant reductions in ADHD symptoms and test anxiety and significant improvements in self-concept, whereas the control group showed no significant pretest-to-posttest changes. Exploratory cross-lagged analyses showed different patterns of association between ADHD symptoms and test anxiety in the two intervention groups. Conclusions: Both MT and CBT were associated with improvements in ADHD symptoms, test anxiety, and self-concept in children with ADHD. The exploratory pathway findings suggest that the two interventions may be linked to partly different patterns of change. However, further studies are needed to verify these preliminary findings. Full article

Background: Vulvar cancer is a rare gynecologic malignancy comprising 1–3% of all cases. No established standard exists for advanced disease, and treatment is often extrapolated from cervical cancer. Although HER2 overexpression is well defined in breast cancer and recognized across multiple solid tumors, its prevalence and significance in vulvar cancer remain unclear. Recent activity of HER2-directed antibody–drug conjugate Trastuzumab deruxtecan in solid tumors with an objective response rate (ORR) of around 37% highlights the need to better characterize HER2 expression in vulvar cancer. Methods: We performed a systematic search of Medline, Embase, and the Cochrane Library up to May 2025. Eligible studies included ≥10 vulvar cancer cases, predominantly vulvar squamous cell carcinoma (VSCC), excluding vulvar Paget’s disease, with available HER2 assessment by immunohistochemistry and/or in situ hybridization. Two reviewers independently screened the studies. A random-effects model was used to estimate pooled HER2 positivity. Heterogeneity was assessed using Cochrane’s Q and Higgins’s I². Results: Of 506 records, nine retrospective studies including 769 patients with predominantly squamous cell carcinoma histology (98%, n = 752) met inclusion criteria. A total of 50 HER2-positive cases were observed. Median age at diagnosis of vulvar cancer was between 55 and 78, reported in three studies. Molecular profiling was limited. Among three studies with known TP53 status (n = 206), 59% of the tumors expressed TP53 (n = 122), and among two studies with known human papilloma virus (HPV) status (n = 128), 21% (n = 27) were HPV-positive. Six studies used American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) HER2 testing guidelines in breast cancer. Pooled HER2-positive expression across ASCO/CAP-based studies was 2% (95% CI: 1%, 3%) and for non-ASCO/CAP-based studies was 21% (95% CI: 2%, 52%). Exploratory pooled estimated proportion of HER2-positive expression was 5% (95% CI: 0.4%, 14%). There was substantial heterogeneity across studies, I² value of 91.1% [95% CI: 85.4%; 94.6%], and no significant publication bias was observed (Egger’s test p = 0.364). This study could not assess prognostic value of HER2 overexpression in VSCC. Conclusions: HER2 positivity in VSCC appears uncommon but it remains to be fully explored. Standardized assessment using contemporary ASCO/CAP breast, endometrial-specific and/or gastric criteria are needed to clarify the prevalence of HER2-positive versus HER2-low/ultralow disease to inform potential use of HER2-targeted therapy. Full article

Background/Objectives: Senior living communities have become a popular living arrangement for older adults seeking supportive environments for aging in place. However, older adults may enter these communities with existing health vulnerabilities. This study described the prevalence of frailty, sarcopenia risk, and cognitive risk among residents and examined their associations with relocation reasons and fitness amenity utilization. Methods: A cross-sectional survey was conducted among residents aged ≥65 years who had lived in the community for at least 3 months. Survey items included demographics, health status, living history, relocation reasons, physical activity, fitness amenity use, and screening tools for frailty, sarcopenia risk, and cognitive impairment. Descriptive statistics, Mann–Whitney U tests, and Spearman rank correlations were conducted. Results: A total of 147 residents (Mean age = 80.2 years, SD = 7.1) responded to the survey. Overall, 28.5% met criteria for frailty, 19.7% screened positive for sarcopenia risk, and 21.8% for cognitive risk. Residents who reported a health-related reason for relocation showed greater frailty (U = 410, p < 0.001), sarcopenia risk (U = 393, p < 0.001), and cognitive impairment (U = 1062, p = 0.01). More frequent fitness amenity use was associated with lower frailty and sarcopenia risk scores (Spearman’s Rho = −0.30 and −0.40, respectively, both p < 0.01), but not cognitive impairment. Conclusions: A meaningful subset of senior living residents were at risk for frailty, sarcopenia, and cognitive impairment. Routine screening and interventions promoting fitness amenity use may support healthy aging in senior living communities. Full article

Within the domain of Earth observation, tracking terrestrial transitions via high-resolution optical data plays a fundamental role. Nevertheless, current methods face critical challenges, including the difficulty in collaborative modeling of local details and global features and the singularity of bi-temporal difference representation, along with insufficient cross-scale feature communication, thereby constraining both the precision and resilience of models when applied to complicated environments. To solve these problems, we propose LADENet (Locally Adaptive Mamba and Multi-scale Feature Enhancement Network), an innovative framework that synergizes CNN, Transformer, and Mamba paradigms. By leveraging customized local contextual refinement alongside sophisticated hierarchical fusion, this integration delivers highly precise and resilient detection performance. LADENet adopts a weight-sharing multi-level Transformer encoder combined with a sequence reduction mechanism to generate multi-scale global features, achieving precise alignment of bi-temporal features and global context modeling while reducing computational complexity. To realize accurate localization and local enhancement of changed regions, we design a dual spatiotemporal adaptive local feature marking module based on State-Space Scanning (SSS). This module screens high-saliency changed regions through an adaptive scanning strategy, realizes pixel-aligned spatiotemporal feature fusion via cross-temporal state-space scanning, and introduces a sliding window boundary calibration mechanism to alleviate boundary information loss caused by window segmentation. To strengthen the feature representation of changed regions, a dual-branch difference enhancement module is constructed, which collaboratively captures global change trends and fine-grained local features through an attention-enhanced difference branch and a multi-scale convolution concatenation branch, effectively suppressing background interference. To address the semantic gap between cross-scale features, a global cross-scale spatial feature fusion decoder is proposed, which balances local detail preservation and global context perception through the synergy of spatial attention and two-dimensional selective scanning, completing refined multi-scale feature fusion and spatial resolution recovery. To rigorously validate the proposed LADENet, comprehensive experiments were conducted across four widely adopted bi-temporal benchmarks: LEVIR-CD, WHU-CD, CLCD-CD, and GVLM-CD. The presented architecture establishes substantial superiority over existing cutting-edge methodologies across primary evaluation criteria. Specifically, it yields an F1-measure of 91.06% alongside an IoU of 85.28% in the LEVIR-CD tests, while registering 90.51% (F1) and 82.45% (IoU) for WHU-CD. Similarly, robust outcomes are delivered on CLCD-CD (82.15% F1, 72.83% IoU) as well as GVLM-CD (89.12% F1, 77.78% IoU). These results demonstrate that LADENet possesses excellent detection accuracy, boundary delineation capability and generalization performance in diverse and intricate bi-temporal observation environments. Full article

Cyber-physical power systems face multi-stage attacks that exploit both communication-network topology and power-grid interdependencies to reach critical substations from low-security entry points. Attack graphs systematically enumerate multi-step attack paths. However, existing approaches either ignore physical network topology or separate attack-graph construction from defense placement, limiting operational usefulness. This paper presents an enhanced topology-aware greedy (TAG) framework that couples source-to-critical attack-path search with dual-mode cyber defense and explicit cyber-physical interdependency modeling. A hierarchical attack graph is constructed directly on the physical network graph, encoding compromise probabilities conditioned on both cyber vulnerability profiles and power-grid criticality. TAG employs topology-aware candidate screening, deterministic probabilistic propagation, beam search, and one-swap local refinement, followed by a dual-mode defense package combining node hardening, micro-segmentation, and monitored-neighbor shielding. Monte Carlo experiments on the 179-bus medium-voltage feeder, IEEE 39-bus New England, IEEE 118-bus, and RTS-96 benchmarks demonstrate that TAG reduces critical-reach probability by $54.3$–$82.0\%$ versus no-defense baselines (mean $70.5\%$), and by $30.2$–$51.5\%$ versus vanilla greedy placement. A cyber-physical impact-weighted risk analysis further shows that TAG’s structural defense placement yields proportional reductions in power-flow-consequence-weighted risk. Parameter sensitivity across the segmentation/containment factor $\eta \in [0.30,0.80]$ and the monitored-neighbor shielding factor $\rho \in [0.05,0.30]$ confirms robust method superiority (68–$72\%$ risk reduction across all tested values). Full article

Prostate-specific antigen (PSA) is a well-established marker for prostate cancer screening, but current ≥4 ng/mL cutoff suffers from low specificity. This study aims to demonstrate the use of a long short-term memory (LSTM) network for accurate prediction of prostate cancer risk and next sequential PSA value. Hong Kong-wide PSA test data over a 25-year period, including PSA values, time difference between PSA tests, and PSA velocity change, were retrieved for model training with variable PSA cutoffs and sequence length. A total of 1,158,915 PSA tests from 499,342 patients (including 18,629 patients with prostate cancer) were included. Models predicting the next PSA level performed well (accuracy 0.724–0.910, AUROC 0.751–0.892). For a ≥4 ng/mL cutoff, the best model was at sequence length of 4 (AUROC 0.864). Temporal prediction of PSA performance was lower (accuracy 0.739–0.811, AUROC 0.740–0.849). Prostate cancer prediction performed excellent (AUROC 0.888–0.973), the sensitivity (0.734), and specificity (0.962) were high even at the shortest sequence length and a ≥4 ng/mL cutoff (4). Utilizing PSA levels only without additional markers or clinical data, LSTM-based models accurately predicted the next PSA level with modest temporal predictions while significantly improving the specificity of prostate cancer risk prediction, demonstrating their clinical utility. Full article

Clinical prediction rules often suffer from a translation gap, balancing high-dimensional statistical accuracy against practical bedside interpretability. This study presents the Surgical Complication Risk Index (S-CRI), an explainable, data-decoupled risk-stratification framework designed to predict post-operative complications using multi-center electronic health registry records (N = 19,965). To ensure strict validation integrity, data partitioning (70% development, n = 13,975; 30% independent holdout testing, n = 5990) was executed before any engineering or risk-tier group isolation. A parsimonious multivariate logistic regression model was fitted within the development cohort, utilizing five predictors: length of stay (LOS) accrued up to the morning of assessment, two institutional categorical groupings, and two historical entry-diagnosis empirical risk tiers. To bridge the translational gap, all fractional regression coefficients were scaled by the baseline anchor and rounded to the nearest whole integer, yielding a simple bedside scorecard where 1 point = 1 inpatient day. On the completely blinded independent holdout cohort, the whole-integer S-CRI demonstrated robust discriminative performance with an Area Under the Receiver Operating Characteristic curve (AUC) of 0.8741 (95% CI: 0.864–0.884) and a Precision–Recall AUC of 0.5785. Setting a baseline operational threshold ≥ 0 yielded an accuracy of 88.18%, a specificity of 96.43%, and a sensitivity of 35.43%, while an optimized integer screening cutoff score of ≥−4 maximized screening capacity (sensitivity: 63.95%; specificity: 91.68%). By enforcing strict temporal landmark constraints to eliminate reverse causality and removing all out-of-sample data leakage, the S-CRI provides an objective, transparent, and interpretable clinical decision support mechanism for early inpatient risk stratification, designed as a supplementary clinical decision-support aid, rather than as a definitive diagnostic replacement for independent clinical judgment. Full article