Search Results (392)

Background/Objectives: Hemodynamic disturbances in sepsis and septic shock arise from the vasoactive effects of inflammatory mediators involved in the immune response. Relaxin-1 is a pleiotropic hormone associated with inflammation, angiogenesis, tissue repair, and vasodilation. This study aimed to investigate the changes in relaxin-1 levels in septic shock and to evaluate their association with mortality. Methods: This prospective observational study was conducted in a Level II intensive care unit. Demographic characteristics, vital signs, APACHE II and SOFA scores, comorbidities, and routine laboratory parameters were recorded at admission and at 48 h. Serum relaxin-1 levels were measured at both time points and analyzed in relation to survival status. Binary logistic regression was additionally performed to evaluate variables associated with mortality in a multivariable framework. Results: A total of 48 patients with sepsis and septic shock were included (54.2% female; mean age 73.4 ± 14.7 years). Overall mortality was 33.3%. Relaxin-1 levels significantly increased from baseline (11.25 ± 4.85 pg/mL) to 48 h (12.64 ± 4.81 pg/mL) (p = 0.047). Baseline relaxin-1 levels were significantly higher in non-survivors compared to survivors (14.62 ± 4.47 pg/mL vs. 11.65 ± 4.73 pg/mL, p = 0.043). Conclusions: Elevated Relaxin-1 levels were associated with mortality in patients with sepsis and septic shock. The observed increase in Relaxin-1 during early follow-up suggests a potential link with the underlying pathophysiological processes. Although Relaxin-1 was associated with mortality, its independent prognostic value could not be established in multivariable analysis due to the limited sample size. Larger, adequately powered multicenter studies are required to confirm these findings. Full article

Urban ethnic enclaves are historically layered cultural landscapes whose public signage encodes community vitality, power relations, and cultural identity in ways that conventional land-use inventories cannot capture. Addressing the absence of scalable, longitudinal computational methods for monitoring such at-risk landscapes, this study develops a reproducible digital-mapping pipeline that operationalises linguistic-landscape analysis as a cultural-heritage monitoring tool for heritage-sensitive land-use planning. Taking Daerim-dong—Seoul’s primary Joseonjok (Korean Chinese) enclave—as a case, we process 38,640 Kakao Map Road View images across 17 annual cross-sections (2008–2024). The pipeline integrates four methodological components: a bounded Spatial Weighting Correction that adjusts for uneven historical coverage; zero-shot semantic sign-function classification using the Qwen2-7B-Instruct model; an exploratory Difference-in-Differences design probing the 2016–2017 THAAD geopolitical disruption; and a Boundary Permeability Ratio (BPR) for tracking enclave edge dynamics. The results document a three-phase trajectory—rapid bilingual expansion (2008–2016), stabilisation (2016–2019), and a COVID-period contraction (2019–2024)—and show that raw sign-count metrics can systematically overstate minority-language decline during economic crises once crisis-period signage is isolated. The BPR is presented as a candidate leading indicator of enclave contraction whose operational thresholds remain to be calibrated through multi-enclave validation. As a methodological proof-of-concept, the study illustrates how computational street-view analysis can support cultural-landscape governance, offering urban planners and heritage managers an actionable, transparent baseline for monitoring at-risk multicultural urban landscapes. Full article

Background: Household food insecurity (HFI), defined as the lack of reliable access to adequate food because of limited money or resources, may influence children’s nutritional status. This study aimed to examine the association between HFI risk, based on a single screening item, and underweight and obesity among kindergarten children in Nevada. Methods: Cross-sectional data from the Kindergarten Health Survey (KHS) collected across three school years (2022–2023, 2023–2024, and 2024–2025) were analyzed using a pooled sample of 7267 children. HFI risk was assessed using one item from the Hunger Vital Sign. Weight status was determined using Body Mass Index (BMI) guidelines from the Centers for Disease Control and Prevention (CDC). Descriptive statistics and multinomial logistic regression examined associations between HFI risk and underweight and obesity, adjusting for confounders. Results: Across the pooled sample, 16.3% were at risk for HFI, 16.0% were underweight, and 21.9% had obesity. In pooled analysis, HFI risk was associated with higher odds of obesity (Adjusted Odds Ratio [AOR] 1.29; 95% Confidence Interval [CI]: 1.05–1.59), but not underweight, compared with food-secure children. In year-specific analyses, higher odds of underweight were observed in 2023–2024 (AOR 1.74; 95% CI: 1.14–2.66) and 2024–2025 (AOR 1.58; 95% CI: 1.04–2.38). Conclusions: HFI risk was associated with obesity among kindergarten children in Nevada, while associations with underweight were observed only in certain school years and should be interpreted cautiously. These findings suggest HFI risk as an important early childhood health concern and support the need for nutrition support, family assistance, and longitudinal research. Full article

Non-contact physiological monitoring based on remote PPG (rPPG) offers a viable alternative for the care of pediatric populations, particularly for children with cerebral palsy (CP) who present unique communication and mobility challenges. This paper presents a review of the literature on the use of rPPG for the estimation of vital signs and its application in emotional monitoring. Following the PRISMA 2020 guidelines as a methodological framework for searching and filtering, an exhaustive search was conducted in the IEEE Xplore and Scopus databases covering the period from 2017 to 2024. A total of 35 studies were selected for analysis. The review examines the evolution of rPPG algorithms—from classical mathematical approaches to recent deep-learning-based architectures—identifying critical technical challenges such as motion artifacts caused by spasticity and variations in lighting conditions. The results reveal that while rPPG has reached technical maturity for monitoring core physiological parameters such as heart rate, its application to robust emotion detection in children with CP remains limited. The main limitation identified across the surveyed literature is the critical scarcity of public or clinical datasets featuring pediatric CP cohorts. Finally, the potential of multimodal integration—combining rPPG with eye-tracking and wearable sensors—is discussed as a promising pathway toward objective emotional monitoring. Such an approach could enhance communication, support rehabilitation processes, and ultimately improve the quality of life of children with cerebral palsy and their caregivers. Full article

Undertriage—the misclassification of acutely ill patients into low-acuity triage categories—is a persistent patient safety concern, and prior machine learning approaches restricted to vital signs have yielded modest predictive performance. We hypothesized that this ceiling reflects feature restriction rather than an inherent predictive barrier. In this retrospective cohort study of 10,792 adult patients (age ≥ 18) initially triaged as Korean Triage and Acuity Scale (KTAS) level 4 or 5 across two tertiary academic centers during 2025, the primary outcome was triage reclassification—change from initial KTAS 4/5 to final KTAS 1–3 (n = 941; 8.7%). Five nested feature sets of increasing breadth were compared using logistic regression (LR) and gradient-boosting classifiers (GBC). Calibration (slope, intercept, Brier score), sensitivity/specificity/positive and negative predictive values at operating thresholds of 3%, 5%, and 10%, and decision-curve net benefit were evaluated on a held-out test partition. NEWS alone yielded an AUROC of 0.58, whereas the full triage-time panel (Set E; 43 features) achieved a GBC AUROC of 0.72 (95% CI 0.68–0.76; 5-fold CV 0.73 ± 0.02) and an AUPRC of 0.23, approximately doubling the NEWS baseline (0.12). The model was well calibrated, with a Brier score of 0.075, a calibration slope of 0.85 (95% CI 0.70–1.01), and an intercept of −0.30 (95% CI −0.65 to 0.07); both intervals included the ideal values of 1 and 0, indicating that predicted probabilities can be interpreted as approximate absolute event likelihoods. At a 5% operating threshold, sensitivity was 0.79, capturing 79% of reclassifications while flagging 53% of the cohort. Decision curve analysis demonstrated positive net clinical benefit across thresholds of 3–20%, exceeding both a vital-signs-only model and the treat-all/treat-none baselines. Feature importance analysis identified pain score, onset-to-arrival time, heart rate, systolic blood pressure, and age as the dominant predictors. Contextual variables routinely documented at triage—particularly pain score and onset-to-arrival time—together with heart rate and systolic blood pressure form a discriminative composite that exceeds the performance of vital-signs-only models in the KTAS 4/5 subpopulation. The resulting model is well calibrated and provides positive net clinical benefit across the 3–20% threshold range, supporting its potential role as a secondary screening flag for low-acuity patients warranting clinician re-review. External validation in independent cohorts is needed before clinical deployment. Full article

Background/Objectives: In-hospital cardiac arrest (IHCA) remains a devastating event associated with high morbidity and mortality among general ward patients. While Rapid Response Systems (RRS) can help identify deteriorating patients, maintaining these systems in secondary hospitals is frequently hindered by severe fiscal and personnel constraints. Consequently, evidence regarding the real-world clinical effectiveness of artificial intelligence software as a medical device (AI-SaMD) for predicting deterioration in such resource-constrained settings remains limited. Methods: We conducted a retrospective analysis on a multicenter, staggered-implementation study evaluating 164,761 eligible adult general ward admissions across three secondary hospitals in South Korea. The intervention involved deploying an AI-SaMD (DeepCARS), which utilizes four routine vital signs to predict ward IHCA within 24 h. The primary outcome was ward IHCA. Secondary outcomes included in-hospital mortality and length of stay (LOS). Exploratory analyses investigated the mechanisms of clinical outcomes by evaluating lead-times to interventions, outcomes in sepsis subgroups, changes in care directives, and post-arrest neurological outcomes. Results: AI-SaMD implementation was associated with a 21% reduction in ward IHCA incidence (adjusted rate ratio 0.79; 95% CI, 0.65–0.96; p = 0.016) and a 15% reduction in in-hospital mortality (aRR 0.85; 95% CI, 0.79–0.90; p < 0.001), alongside significantly shorter hospital and intensive care unit LOS. These associations were also observed in patients with sepsis (IHCA aRR 0.71; 95% CI, 0.54–0.93; p = 0.013). Lead-times to critical care intervention and to antibiotic escalation were numerically shorter in the AI-SaMD group by 16.3 h (p = 0.066) and 2.6 h (p = 0.523); poor neurological outcome at discharge among ward IHCA cases was 85/108 (78.7%) in the AI-SaMD group versus 63/102 (61.8%) in the standard-care group (aRR 1.13; 95% CI, 0.99–1.33; p = 0.058); and the full-code death rate did not differ between groups (aRR 0.94; 95% CI, 0.76–1.15)—none of these additional analyses reached statistical significance. Conclusions: In secondary hospitals unable to operate an RRS due to fiscal limitations, implementation of an AI-SaMD as an additional informational layer was associated with lower ward IHCA and in-hospital mortality. The AI-SaMD may serve as an actionable and scalable additional safety layer for general-ward patients in resource-constrained environments where RRS infrastructure is not feasible. Although this was a multicenter, large-scale study, the present analysis was retrospective and quasi-experimental in design; rigorous randomized studies are needed to confirm these associations. Full article

Background and Perspectives: Obstructive sleep apnea (OSA) is a prevalent condition associated with increased perioperative risks. Endobronchial ultrasound (EBUS), a diagnostic and staging procedure requiring deep sedation, may pose additional risks for patients at high risk of OSA. The Integrated Pulmonary Index (IPI), derived from capnography and vital signs, offers a single numerical value reflecting respiratory status. This study aimed to assess the association between high OSA risk and adverse events using the IPI during EBUS under sedation. Materials and Methods: This prospective cohort study included 65 patients undergoing EBUS with sedation between December 2024 and April 2025 at a tertiary referral center. STOP-Bang questionnaire scores were used to stratify patients into high- (≥3) and low-risk (<3) OSA groups. During the procedure, IPI, oxygen saturation, end-tidal carbon dioxide, respiratory rate, and hemodynamic parameters were recorded at multiple time points. Hypoxemia, hypoventilation, and apnea were defined using standard thresholds. Logistic regression and Generalized Linear Mixed Models (GLMM) were applied to examine associations between OSA risk and respiratory outcomes. Results: Forty-three patients (66.2%) were classified as high risk for OSA. Patients with high STOP-Bang scores were older and had higher BMI, comorbidity rates, and ASA scores (all p < 0.05). IPI values were lowest between 5 and 10 min, accompanied by more frequent interventions. Logistic regression showed no significant association between STOP-Bang scores and low IPI or hypoxemia. GLMM analysis also indicated no significant association between high OSA risk and low IPI (OR = 1.02; 95% CI = 0.36–2.86; p = 0.974). Hypoxemia was nearly threefold higher in high-risk patients, though not statistically significant (p = 0.080). Conclusions: Although no statistically significant association was identified between high OSA risk and adverse respiratory events, GLMM analyses revealed that patients with high STOP-Bang scores demonstrated approximately three times higher odds of developing hypoxemia (OR = 2.76; 95% CI = 0.99–7.66; p = 0.052), a result that approached statistical significance. The present findings do not support the routine use of IPI-based monitoring in this setting, and further adequately powered studies are warranted. The early procedural period (5–10 min) is critical for hypoxemia and respiratory compromise. Full article

This study investigated the association between serum hypoxia-inducible factor 1-alpha (HIF-1α) levels and clinical severity in patients with coronavirus disease 2019 (COVID-19). This prospective case–control study included 91 patients with confirmed COVID-19, of whom 51 had severe-critical disease with pneumonia and 40 had mild disease without pneumonia, as well as 39 healthy controls. Vital signs, including body temperature, pulse rate, respiratory rate, oxygen saturation, and blood pressure, were recorded. Biochemical parameters such as complete blood count, D-dimer, ferritin, creatinine, urea, and high-sensitivity cardiac troponin T were analyzed. Serum HIF-1α levels were measured using ELISA. Median HIF-1α levels were 132.9 pg/mL (IQR: 131.7–138.0) in the severe-critical disease group, 137.35 pg/mL (IQR: 131.65–152.75) in the mild disease group, and 136.6 pg/mL (IQR: 132.2–162.2) in controls. Significant differences were observed between groups (p = 0.012). ROC analysis showed a discriminatory performance for HIF-1α, with a sensitivity of 89.01% and specificity of 35.90% at a cut-off value of ≤154 pg/mL for distinguishing mild disease from controls, and a sensitivity of 86.3% and specificity of 42.5% at a cut-off value of ≤141.1 pg/mL for distinguishing severe-critical disease from mild disease. HIF-1α levels decreased with increasing disease severity. HIF-1α levels were found to be associated with disease severity; however, the low AUC values indicate that this parameter has limited discriminative ability for clinical use when used alone. Full article

Background: Fiberoptic bronchoscopy (FOB) is a procedure routinely performed in clinical practice for both diagnostic and therapeutic purposes. FOB frequently impairs respiratory function, which may exacerbate respiratory failure. Currently, conventional oxygen therapy (COT) is the most commonly used form of respiratory support; however, non-invasive ventilation (NIV) and high-flow nasal cannula (HFNC) are being used increasingly. The optimal settings and indications for NIV and HFNC in patients with respiratory acidosis undergoing FOB have not yet been determined. Methods: This is a prospective, multicenter, randomized controlled trial including two parallel study populations defined by the indication for bronchoscopy and the type of respiratory acidosis. Therapeutic FOB (Study 1): Patients with decompensated type 2 respiratory failure (pH < 7.35 and PaCO₂ > 45 mmHg) will be randomized to receive one of four methods of respiratory support during bronchoscopy: COT, NIV, HFNC, or invasive mechanical ventilation (IMV) (n = 315). Diagnostic FOB (Study 2): Patients with chronic respiratory acidosis (pH ≥ 7.35, PaCO₂ > 45 mmHg, and/or HCO₃⁻ > 27 mmol/L) will be randomized to receive COT, NIV, or HFNC during bronchoscopy (n = 210). Before FOB, patients in both groups will undergo arterial blood gas (ABG) analysis. During FOB, vital signs will be continuously monitored, including SpO₂, FiO₂, TcCO₂, ECG, and heart rate. After FOB, ABG analysis will be repeated, and study endpoints and complications, if any, will be recorded. The planned study period is from April 2026 to April 2029. Results: Based on the study results, we aim to evaluate the effectiveness and safety of different respiratory support strategies during flexible bronchoscopy, with the primary objective of comparing the rate of treatment failure among COT, HFNC, NIV, and IMV. Treatment failure is defined as the need for endotracheal intubation, premature termination of the procedure, or escalation of respiratory support. Additionally, we aim to identify the optimal NIV and HFNC settings, as well as complication rates in both study groups. Conclusions: The results of this study will help define the role of optimal respiratory support in patients with respiratory acidosis undergoing FOB, potentially leading to a shorter time from admission to diagnosis, better tolerance of the procedure, and faster recovery afterward. Full article

Background/Objectives: In critically ill patients, endotracheal intubation (EI) is often performed to secure the airway or mechanical ventilation. Accurately predicting the timing of intubation significantly affects patient outcomes. We developed an artificial intelligence (AI) model designed for real-time risk stratification of patients requiring EI. Methods: We utilized the Medical Information Mart for Intensive Care-IV (MIMIC-IV) 2.2 dataset and performed model development using 15 clinical variables, including vital signs, Glasgow Coma Scale (GCS) score, and arterial blood gas analysis results. Patients intubated before or within 1 h of intensive care unit (ICU) admission were excluded. Clinical data from the ICU inherently consists of continuous time-series measurements. Traditional machine learning models often treat this information as static tabular data, neglecting vital temporal dynamics and patient history. Conversely, deep learning time-series approaches can capture these complex patterns over time. Thus, we applied the Gated Recurrent Unit with Decay++ (GRU-D++) model to predict the need for EI. GRU-D++ is an extension of the GRU and GRU-D. It builds upon the GRU-D to provide improved performance when handling datasets with exceptionally high rates of missing values. GRU-D++ is a time series deep learning model with an automatic mechanism for imputing missing values. This built-in capability eliminates the need for additional data preprocessing and has previously demonstrated high predictive performance. Using the 15 variables, we evaluated the optimal timing for EI in ICU-admitted patients by applying various AI models. Results: Among these, the GRU-D++ model demonstrated AUROC of 0.888, AUPR of 0.481, sensitivity of 0.474, specificity of 0.995, precision of 0.511, and F1 score of 0.491 on MIMIC-IV dataset. For KNUH dataset, the model demonstrated AUROC of 0.913, AUPR of 0.063, sensitivity of 0.162, specificity of 0.997, precision of 0.137, and F1 score of 0.147 within the 2 h in advance scenario. Furthermore, when compared with conventional scoring systems such as the Heart rate, Acidosis, Consciousness, Oxygenation, Respiratory rate (HACOR) score and Respiratory rate-Oxygenation (ROX) index, the GRU-D++ model also showed better performance predictive accuracy. Conclusions: The AI-based intubation prediction model developed in this study holds potential as a real-time risk stratification tool, providing timely risk assessments regarding the need EI. While operational threshold recalibration is essential prior to clinical deployment, further prospective multicenter studies are required to validate the clinical utility of this model in real-time practice. Full article

Background: Effective communication with patients is vital for improving health outcomes in chronic disease management. In this study, we investigated WoundScribeAI’s Scribe AI, also known as Ambient Technology, and its patient education and engagement app, Pingoo.AI. It employed a multi-agent AI model that leveraged Large Language Models (LLMs) and NotebookLM to enhance patient communication in clinical settings. Methods: The system comprised specialized agents that transcribed healthcare provider–patient conversations through ambient dictation. This transcription generated medical notes that followed the Subjective, Objective, Assessment, and Plan (SOAP) format—a structured document used by healthcare providers to record and communicate information about patient encounters. Simultaneously, comprehensive visit summaries were also created. In the next step, these visit summaries were used to produce conversational and educational content by leveraging NotebookLM, an AI model introduced by Google that can generate podcast-style conversations from provided information. Integrating these agents allows clinicians to deliver engaging, empathetic, and actionable information to patients. Medical experts conducted a two-phase evaluation of the system’s performance based on multiple criteria, with a particular focus on diabetes education and diabetic foot care. The first phase used pre-recorded training videos, while the second phase involved simulated consultations by clinicians using the system. To validate the AI-generated educational content, we used several established frameworks in health communication that closely align with our enhancement goals. Results: The results showed that the AI model generated accurate clinical documentation and met the criteria for accurate SOAP Notes, visit summaries, and engaging educational content for patients. Given that hallucination is a significant concern related to large language models, especially in critical fields like healthcare, we meticulously analyzed the generated outputs to identify any signs of hallucinated information. Three outcomes successfully passed the validation criteria, including accuracy, completeness, comprehensiveness, absence of potential harm, and no hallucination. Additionally, the Conversational Education content was confirmed against established patient education frameworks and met criteria such as the use of metaphors, empathetic tone, and appropriate language, providing additional detail to help manage the condition. Conclusions: By providing specific instructions and prompts to NotebookLM to transform visit summaries into educational conversations, we significantly enhanced the comprehensiveness and engagement of the content for patients. In contrast to a traditional summary of the clinical visit, the podcast-style conversation enriched the content with background information, encouraging language, an empathetic tone, and helpful metaphors. Our analysis confirmed that the system did not exhibit any hallucinations, highlighting the effectiveness of our approach in mitigating this risk. These findings support the use of multi-agent AI models, combined with ambient dictation and tools like NotebookLM, to improve patient communication that surpasses traditional paper-based brochures, which are often impersonal, minimal, and do not always adhere to recommended factors for health literacy. Full article

Introduction: Stroke remains a leading cause of mortality and long-term disability worldwide. Accurate prognosis prediction is essential for timely intervention and personalized treatment planning. However, previous studies have often overlooked the role of patients’ medical history, age-specific risk factors, and time-dependent mortality patterns. This study aimed to develop and evaluate machine learning models for predicting mortality in stroke patients by incorporating vital signs, blood test results, demographic characteristics, and medical history, while also exploring subgroup-specific factors. Methods: We retrospectively analyzed data from 1780 stroke patients admitted to Hallym University Sacred Heart Hospital between 2018 and 2023. Input features included both original and binarized forms of vital signs and blood test values, along with age and medical history. Random Forest models were developed to predict mortality at 1, 2, and 3 years post-admission, as well as overall mortality. Model performance was assessed using AUC and 95% confidence intervals, and variable importance was evaluated using Mean Decrease Gini and SHAP values. Results: The highest predictive performance was observed in a model for patients under 60 using binarized input features, achieving an AUC of 0.995 (CI: 0.98–1). Across all models, pulse rate consistently emerged as the most important predictor. Additional key features included platelet count and diastolic blood pressure. SHAP analysis revealed that pulse rate was associated with higher mortality risk. Subgroup analyses based on age and medical history improved interpretability and predictive power. Conclusions: This study demonstrates that integrating clinical indicators with demographic and medical history variables can significantly enhance the accuracy and interpretability of mortality prediction models in stroke patients. The results underscore the importance of stratified modeling and continuous monitoring of vital signs, particularly pulse rate, to support precision stroke care. Full article

Chimeric antigen receptor (CAR) T-cell therapy is an effective treatment for hematologic malignancies. However, it is limited by high costs, risk of severe toxicities such as cytokine release syndrome and neurotoxicity, and heterogeneous patient responses. The current therapy monitoring depends largely on subjective symptom assessment, routine laboratory tests, and basic vital signs, without real-time, quantitative evaluation of CAR T-cell expansion or activation in clinical practice. This lack of timely immune monitoring hampers individualized care and contributes to increased treatment costs. To address this need, we present a proof-of-concept, label-free rapid optical imaging (ROI) biosensor with automated machine learning analysis for direct quantification of CAR T-cells from whole blood. This microfluidic platform integrates red blood cell (RBC) removal, CAR T-cell capture, and imaging-based quantification on a single chip, eliminating the need for centrifugation, staining, and operator-dependent interpretation. For validation, 50 μL whole blood samples spiked with Jurkat cells expressing CD19 CARs underwent RBC depletion by agglutination and microfiltration. The remaining blood components were then incubated on a sensor chip functionalized with recombinant CD19 protein. Captured CAR T-cells were imaged by brightfield microscopy and automatically enumerated using a machine learning algorithm trained on fluorescence-validated cells. The CD-19 cells’ capture performance was validated by flow cytometry and fluorescence imaging. The trained machine learning model validated at 88% sensitivity and 96% specificity. Buffer and whole blood calibration curves were established across clinically relevant concentrations (1–1000 cells/µL) with triple replicates. The results showed high correlation (0.975 and 0.990 R²) between the spiked concentration and the detected CAR T-cells, with a 95% certainty limit of detection (LOD) and quantification (LOQ) of 0.6 and 1.1 cells/µL for spiked buffer, and 14 and 67 cells/µL for spiked whole-blood, respectively. Full article

Background: Postoperative anemia associated with intertrochanteric hip fracture is a detrimental complication that detrimentally impairs patients’ outcomes. This study is designed to develop an online predictive tool to assist physicians in developing surgical blood preparation strategies to prevent the occurrence of postoperative anemia. Methods: This study included data collected from June 2017 to June 2025 on intertrochanteric hip fracture patients at Tangdu Hospital, including demographic information, comorbidities, vital signs, and laboratory results. LASSO regression was used to select predictive variables, and seven machine learning techniques: Logistic Regression, Support Vector Machine, Decision Tree, LightGBM, XGBoost, Neural Networks, and Random Forest, were compared to identify the best tool for predicting postoperative anemia risk. We created a patient-specific risk prediction tool with SHAP-driven interpretability for clinical decision support. Results: A total of 815 patients were included in the analysis, of whom 208 (25.5%) presented with postoperative anemia. Eight variables were selected to build seven machine learning models. Among these, the SVM model exhibited the best predictive performance in terms of discrimination, calibration, and clinical applicability, with an AUC range of 0.827–0.831. In test sets encompassing diverse population characteristics, SVM achieved the highest sensitivity (72.73%), accuracy (77.78%), and F1 score (57.14%). Conclusions: We established an online prediction platform for clinical practice, enabling clinicians to assess anemia risk in intertrochanteric hip fracture patients and support early prevention of postoperative anemia. Full article

Background/Objectives: Following prior in vitro and in vivo investigations on the bone health benefits of green onions and oats, we aimed to assess the effects of WCO31, Allium fistulosum L. (green onion) root and Avena sativa L. (oat) mixtures, on height growth and safety. Methods: This multi-center, randomized, double-blind, placebo-controlled study included 150 children aged 6–8 years (75 males and 75 females) who fell between the 3rd and 50th percentiles of the Korean National Growth Charts but had not yet developed secondary sexual characteristics. They were randomly assigned to receive daily oral administration of WCO31 (1.2 g/day) or a placebo for 24 weeks. For efficacy analysis, height, growth rate, growth rate standard deviation score (SDS), height SDS, and growth-related parameters were measured. To evaluate the safety of the intervention, several safety parameters (including the incidence of adverse events, laboratory tests, and vital signs) were monitored. Results: The WCO31 group demonstrated significantly superior outcomes, including height, growth rate, growth rate SDS, height SDS, and height-for-age Z-score, than the placebo group (all p < 0.001). Moreover, no safety-related concerns were identified. Conclusions: WCO31 positively influences height growth and demonstrates a favorable safety profile, with no observable adverse effects. This study provides the first clinical evidence supporting growth enhancement using natural extracts, suggesting that WCO31 could serve as a cost-effective, safe, and accessible complementary strategy for promoting child growth. Full article