Search Results (3,318)

Urban expressway projects are critical components of modern transportation infrastructure, yet their construction quality is often threatened by multi-source, latent, and dynamic risks. Traditional expert-driven risk identification methods frequently suffer from subjective bias and low efficiency, failing to meet the rigorous management requirements of complex engineering environments. To address these challenges, this study proposes a robust risk assessment framework integrating Large Language Models (LLMs) and the Delphi method within a Bayesian Network (BN) structure. First, LLM technology is leveraged to perform semantic mining on extensive engineering texts, including construction specifications and project reports, to pre-identify potential risk factors. Second, the Delphi method is applied through multiple rounds of expert consultation to refine a comprehensive inventory comprising 32 risk factors across five dimensions: personnel, machinery, materials, methods, and environment. Finally, a BN-based evaluation model is developed, utilizing forward inference, backward diagnosis, and sensitivity analysis to quantify risk levels and pinpoint critical risk drivers. The framework was empirically validated using the T Expressway Project in Hangzhou as a case study. Results demonstrate that the model effectively transforms empirical management into precise, data-driven diagnosis, providing project managers with a quantitative tool for optimizing construction quality control and decision making in complex urban bridge projects. Full article

Invasive candidiasis is a significant concern in healthcare environments, and awareness of these infections has increased in recent years. A growing number of risk factors, the ability of some Candida species to progress from colonization to tissue invasion, and their capacity to adhere to and survive on abiotic surfaces have all contributed to the spread of invasive candidiasis. The primary goal in cases of invasive candidiasis is to diagnose it as promptly as possible, as any delay can delay antifungal treatment. This review concentrates on clinical syndromes reunited under the definition of invasive candidiasis and the current diagnostic methods. Risk factor assessment is another major topic of this narrative review and recent updates are included. Research stage biomarkers are also explored and partial results are mentioned as there are continuous efforts to search for new tools for a more accurate prediction or an earlier identification of IC. Full article

Breast cancer is one of the leading causes of cancer deaths in women worldwide. Neoadjuvant chemotherapy (NACT) has increased rates of breast-conserving procedures and enabled the identification of patients with a particularly poor prognosis. Achieving a pathological complete response (pCR), an indicator of NACT efficacy, contrasts with residual disease (RD), which identifies patients at higher risk of recurrence. This review provides an overview of current evidence on the clinical and prognostic significance of pCR and RD in patients receiving NACT for breast cancer. The analysis is based on data from randomized clinical trials, meta-analyses, and current clinical guidelines for contemporary systemic treatment. Pathological complete response varies according to tumor subtype, with the highest rates observed in triple-negative and non-luminal HER2-positive breast cancer. In HER2-positive disease, the combination of chemotherapy with HER2-targeted therapies increases pCR rates, while the presence of RD supports escalation of postoperative treatment with antibody–drug conjugates. In triple-negative breast cancer (TNBC), the inclusion of platinum agents and immune checkpoint inhibitors improves treatment efficacy. In HER2-negative breast cancer and germline BRCA1/2 mutations, adjuvant PARP inhibitors improve survival independently of pCR, highlighting the complex relationship between pathological response and prognosis. Immunotherapy and targeted therapies are used alongside standard chemotherapy and hormone therapy in perioperative treatment. Further research is required to refine response assessment, integrate new biomarkers such as circulating tumor DNA (ctDNA), and optimize treatment selection, while clarifying the significance of reassessing hormone receptor and HER2 status in residual disease and its impact on subsequent treatment decisions. Full article

The Jimusar shale reservoir is characterized by saline lacustrine mixed sedimentation and strong reservoir heterogeneity, making continuous identification of formation elemental composition challenging. Although elemental capture spectroscopy (ECS) logging provides direct elemental measurements, its high cost and limited deployment restrict its large-scale application. This study investigates the feasibility of predicting ECS-derived elemental compositions from conventional logging data to support continuous reservoir characterization. A dataset comprising 115,668 depth-matched samples from three wells in the Jimusar Sag, Junggar Basin, was used. Conventional logging curves served as input features, while ECS-derived elemental concentrations were used as prediction targets. After data preprocessing and feature enhancement, correlation analysis identified seven relevant logging curves as key input variables. Four regression models—Random Forest, XGBoost, CatBoost, and LightGBM—were evaluated and compared with a stacked ensemble learning model. Model performance was assessed using five-fold cross-validation and multiple metrics, including R², RMSE, MAE, and relative error. The results show that all four individual models achieved satisfactory predictive performance, with R² values generally around 0.8, whereas the stacked ensemble model provided the highest prediction accuracy and stability. Compared with the individual models, the ensemble model improved R² by 2–10%, reduced RMSE by 5–15%, and decreased relative error by 8–15% across different elemental predictions. Among the predicted elements, Fe achieved the highest accuracy, with an R² value of 0.87. As an exploratory engineering application, the predicted elemental compositions were further compared with hydraulic-fracturing response parameters, achieving a conformity rate of 74.8% with fracturing-operation status. These results suggest that predicted elemental data may provide useful auxiliary constraints for fracture-response interpretation and abnormal-risk identification. Nevertheless, further validation using independent well data is required, and the generalizability of the proposed workflow to other wells and lacustrine shale oil systems remains to be further assessed. Full article

Background: Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous syndrome characterized by frequent underdiagnosis, diverse etiologies, and limited therapeutic options. Given its complexity, artificial intelligence (AI) and machine learning (ML) offer promising avenues to decode high-dimensional, multi-modal healthcare data. This review aims to synthesize the current landscape of AI/ML applications in HFpEF, evaluating their potential to address critical unmet clinical needs. Methods: We conducted a comprehensive review of the literature focusing on AI/ML paradigms in HFpEF. Key methodological frameworks were examined, including supervised, unsupervised, semi-supervised, and reinforcement learning, alongside advanced techniques such as deep learning and natural language processing (NLP). The analysis focused on the application of these techniques across four domains: diagnosis, sub-phenotyping, risk prediction, and optimization of diagnostic modalities, with specific emphasis on studies incorporating external validation. Results: Current evidence demonstrates that AI approaches effectively enhance diagnostic accuracy and facilitate the identification of distinct HFpEF phenotypes beyond traditional classifications. These technologies show significant utility in refining prognostic assessments and optimizing diagnostic testing strategies. Furthermore, ML-driven analytics provide a robust framework for improving patient selection and streamlining clinical trial design, potentially overcoming historical barriers to drug development in this population. Conclusions: AI represents a transformative tool capable of dissecting the heterogeneity of HFpEF to enable precision medicine. While the potential to improve clinical outcomes is substantial, challenges regarding model interpretability, bias, and clinical integration persist. Future efforts must focus on rigorous external validation and prospective trials to ensure the responsible translation of these technologies into routine clinical practice. Full article

Background/Objectives: The frontal sinus exhibits individual morphological variability that may support human identification. Its development progresses through childhood and adolescence and stabilizes in early adulthood, with age-related changes potentially affecting radiological comparisons. This study presents a forensic case report and discusses it in light of the literature on frontal sinus development and forensic identification. Methods: A comparative radiological analysis was conducted using images obtained at two distinct stages of biological maturation (14 and 21 years of age). Manual delineation combined with semi-automated computational analysis was applied to assess morphological features of the frontal sinus, including contour configuration, lobulation, and dimensional parameters. Results: The intra vitam record was obtained at 14 years of age, during an active developmental phase, and the post mortem examination was obtained at 21 years, corresponding to early adulthood. Comparative analysis revealed significant morphological differences, including increased lobulation, contour complexity, and sinus expansion. These changes limited the reliability of frontal sinus morphology for identification in this case. Friction ridge examination independently established positive identification. Conclusions: This study highlights the limitations of frontal sinus analysis when applied across periods of active development and underscores the risk of misinterpretation if age-related changes are not adequately considered. Full article

Linear heritage corridors are increasingly exposed to spatially heterogeneous pressures from climate change and human activities, yet integrated geospatial frameworks for corridor-scale risk identification remain limited. Taking the Beijing–Hangzhou Grand Canal as a representative linear World Heritage corridor, this study developed a GIS-based spatiotemporal assessment framework to quantify natural risk, anthropogenic pressure, and their coupled patterns during 1995–2024. Approximately 350 canal segments were constructed as comparable assessment units and linked with 49 heritage sites and 18 World Heritage canal sections through a multi-scale spatial framework integrating canal sections, buffer zones, and heritage sites. Natural risk was characterized using extreme temperature, precipitation, and drought indices, while anthropogenic pressure was represented by nighttime lights, population density, impervious surface, and road density. The results reveal a clear north–south gradient in integrated natural risk, with higher values concentrated in the southern canal sections. Among the three natural-risk modules, temperature, precipitation, and drought contributed weights of 0.594, 0.242, and 0.164, respectively, indicating the dominant role of heat-related processes. The first two principal components of anthropogenic pressure explained 80.8% of the total variance. Four dominant coupling types were identified, among which the dual high-pressure type was concentrated mainly in the southern canal and marked the most critical areas of compound risk. This study provides a geospatial approach for hotspot detection and spatial decision support for the conservation of large linear heritage systems. Full article

Background: Acute knee trauma is a common reason for emergency department visits, yet early identification of patients requiring surgical intervention remains challenging. Most existing prediction studies focus on discrimination metrics and provide limited guidance for clinical decision-making. Methods: We conducted a retrospective study of 905 patients presenting to the emergency department with acute knee trauma. Prediction models were developed using logistic regression, random forest, and extreme gradient boosting (XGBoost) based on routinely available clinical variables. Model performance was evaluated in terms of discrimination (AUROC, AUPRC), calibration, and clinical utility. Threshold-specific performance metrics and decision curve analysis were used to assess clinical applicability, and patients were stratified into risk groups based on predicted probabilities. Results: Among 905 patients, 163 (18.0%) underwent surgical intervention. Logistic regression and random forest demonstrated comparable performance (AUROC 0.748 and 0.744, respectively), whereas XGBoost showed lower discrimination (AUROC 0.632). Calibration was acceptable overall but less stable at higher predicted probabilities. Threshold-specific analysis demonstrated meaningful trade-offs between sensitivity and specificity across probability thresholds. Decision curve analysis showed that the model provided greater net benefit than default strategies within a threshold range of approximately 0.05–0.25. Risk stratification showed increasing surgical rates across risk groups, although the degree of separation was modest. Conclusions: Prediction models based on routinely available clinical variables can support early risk assessment in acute knee trauma. Their clinical usefulness depends on threshold-specific evaluation and decision-analytic approaches rather than overall performance metrics alone. These findings highlight the importance of interpreting prediction models within a clinical decision-making framework to facilitate real-world application. Full article

Background/Objectives: Respiratory dysfunction in Parkinson’s disease (PD) is a clinically relevant but frequently underrecognized manifestation associated with functional impairment and increased risk of respiratory complications. This study compared spirometry and impulse oscillometry (IOS) in the assessment of respiratory function in PD, with particular focus on the detection of subtle or peripheral airway abnormalities. Methods: A prospective, single-center, cross-sectional study was conducted, including 108 participants (55 patients with PD and 53 control subjects). Pulmonary function was evaluated using standardized spirometry and IOS protocols. Group comparisons were performed using non-parametric tests, while multivariable regression analyses adjusted for potential confounding factors, including age, body mass index, smoking status, pollutant exposure, and cardiovascular comorbidities. Results: IOS identified a higher frequency of abnormal categorical findings compared with spirometry, including among subjects with normal spirometric values. Although dyspnea was more frequent in patients with PD in unadjusted analyses, multivariable regression demonstrated that PD was not an independent predictor of respiratory dysfunction. Pollutant exposure was significantly associated with abnormal IOS findings (p = 0.011). No significant differences were observed between PD and control groups regarding continuous spirometric or oscillometric parameters. Only a weak association between disease severity and FEV1 (%) was identified, whereas no significant correlations were observed for oscillometric parameters. Conclusions: IOS may provide complementary information regarding subtle or peripheral respiratory abnormalities in patients with PD. The findings suggest that respiratory alterations in this population are likely multifactorial and not independently determined by PD itself. Incorporating oscillometric assessment into respiratory evaluation may contribute to the identification of subtle respiratory mechanical alterations in patients with PD. Full article

Sexual violence against women frequently originates during adolescence, when structural inequalities and gendered power dynamics heighten vulnerability, making early identification of risk factors essential to prevent trajectories leading to sexual exploitation. This study presents the psychometric validation of the EDR-ESIA, a screening instrument designed to detect vulnerability to Child Sexual Exploitation (CSE) in healthcare, education, and social care settings, with particular relevance for prevention strategies targeting adolescent girls. The sample comprised 199 adolescents aged 11–17 years (M = 15.23; SD = 1.59) residing in Spain (58.8% female, 40.2% male, 1.0% unspecified), assessed by trained professionals using case records and reports. The 88-item instrument underwent expert review and pilot testing prior to validation, and its internal structure was examined using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicated that all subdimensions and higher-order constructs showed an adequate fit to the theoretical model, supporting the instrument’s validity. Female adolescents scored significantly higher than males on CSE target indicators, reflecting a medium-to-large gender difference in vulnerability levels. Overall, the EDR-ESIA constitutes an evidence-based instrument for the timely recognition of CSE vulnerability, supporting prevention, education, and intervention efforts aimed at reducing sexual violence against women from early developmental stages. Full article

Background: Early identification of bacteremia in immunosuppressed cancer patients remains difficult, especially in neutropenia. This study evaluated the diagnostic accuracy of NLR, PLR, and NLPR for identifying bacteremia and sepsis in patients undergoing blood culture episode. Methods: We conducted a retrospective diagnostic accuracy study at a tertiary oncology center between January 2023 and December 2024. All bacteremia identified were included as cases. Culture-negative episodes were subsequently sampled as controls using a frequency-matching strategy. Hematological parameters were obtained within ±24 h of first blood culture episode. Diagnostic performance was assessed using ROC curve analysis and multivariable logistic regression. Results: Of 369 screened episodes, 337 from 323 unique patients were included after excluding 31 records. NLPR showed the highest accuracy for bacteremia (AUC 0.730; 95% CI 0.671–0.788). The optimal cut-off was 0.038 (sensitivity 69.2%, specificity 72.3%) and remained consistent after excluding episodes with antibiotic therapy (AUC 0.768), corticosteroids (AUC 0.708), or growth factor use (AUC 0.718). In severe neutropenia, NLPR showed the highest accuracy (AUC 0.887; 95% CI 0.797–0.978). In multivariable analysis (n = 304), NLPR remained independently associated with bacteremia (p < 0.001), with good model discrimination (AUC 0.815; 95% CI 0.763–0.866). Diagnostic performance for sepsis was lower and not statistically significant. Conclusions: These findings suggest that NLPR may represent a simple, inexpensive, widely accessible adjunctive biomarker to support early bacteremia risk stratification in immunosuppressed cancer patients, particularly in patients with severe neutropenia. Although its overall discrimination was comparable to isolated lymphocyte count, NLPR may provide clinically relevant contextual information by integrating multiple dimensions of immune dysregulation. Further prospective multicenter validation is warranted. Full article

Background: Human papillomavirus (HPV)-related diseases remain a major global health problem, with cervical intraepithelial neoplasia (CIN) representing a key precursor to cervical cancer. Identification of high-risk HPV genotypes is essential for early diagnosis and appropriate management. This study aimed to evaluate the usefulness of in situ hybridization (ISH) for detecting HPV DNA in formalin-fixed, paraffin-embedded (FFPE) cervical tissue and to compare automated signal detection with manual histopathological assessment. Methods: This prospective, non-randomized study included 83 women undergoing diagnostic procedures for abnormal cytology or confirmed CIN between 2022 and 2023. Tissue specimens obtained during a loop electrosurgical excision procedure (LEEP) were examined using two ISH probes: ISH II for low-risk HPV types 6 and 11, and ISH III for high-risk HPV genotypes. Staining patterns and distributions were evaluated and correlated with molecular HPV testing and histopathological outcomes. Results: ISH II distribution was significantly associated with the presence of HPV type 6 or 11 (p < 0.001), although stain structure itself was not. ISH III stain structure was significantly associated with high-risk HPV genotypes (p = 0.020). A positive ISH II result predicted low-risk HPV infection with a sensitivity of 62.5% and specificity of 64.0%, while ISH III predicted high-risk HPV infection with a sensitivity of 86.36% but lower specificity (23.53%). Overall diagnostic accuracy was 63.86% for ISH II and 73.49% for ISH III. Conclusions: ISH proved to be a reproducible method for detecting HPV in archived cervical tissue, enabling assessment even years after specimen collection. Although PCR-based methods remain more widely used due to higher sensitivity and less invasive sampling, ISH provides valuable morphological context and may serve as a complementary diagnostic tool, particularly when only archival tissue is available. Full article

Pre-made dishes have drawn growing attention because of their convenience and rapid market expansion. Their food safety risks, however, are shaped not only by products themselves, but also by the gap between public perception, reported incidents, and inspection records. This study develops a three-stage analytical approach by combining Weibo public opinion data, news media reports, and food inspection records from Gansu Province. First, ERNIE and BERTopic are used to identify public sentiment and discussion topics. The results show that negative sentiment slightly exceeds positive sentiment, with school meals, additives, and food safety as the main concerns. Second, 11,110 pre-made dish-related food safety reports from Food Partner Network are clustered and assessed for incident severity. The results point to drug residues in aquatic products, microbial contamination in egg products, authenticity disputes over meat ingredients, and quality issues in frozen composite foods. Third, based on the 2024 official definition, 12,121 inspection records are screened, and 2783 definition-constrained pre-made dish-associated products are retained. Six imbalanced classification models are then constructed. The Weight + RF model performs relatively well for starch and starch products, with a Precision of 0.7857, an AUC-ROC of 0.7778, and an MCC of 0.4429. The study provides a reference for risk identification and inspection resource optimization under limited pre-made dish inspection data. Full article

Early literacy screening is essential for reducing long-term educational inequality, yet traditional paper-based assessments remain difficult to scale due to logistical constraints and delayed feedback. This study presents K-KOBUKI, a cloud-based prototype screening workflow that organizes early literacy assessment as a human-validated, data-driven process. The system integrates structured assessment responses with automated speech recognition-based analysis of oral reading performance across five literacy domains and incorporates a human-in-the-loop verification stage to ensure the reliability of speech-derived features. The system was evaluated using data from 195 first-grade students. Across repeated stratified cross-validation, multiple classification models achieved stable recall (≈0.85) under class imbalance conditions, supporting consistent identification of at-risk learners. Psychometric-informed feature refinement improved precision without reducing recall, indicating enhanced signal clarity through measurement-level stabilization. Explainable AI analysis further revealed that word reading and reading fluency contributed strongly to model-level decision boundaries, while vocabulary knowledge provided complementary influence at the individual level. These findings provide prototype-level evidence that a human-validated, multimodal screening workflow can support stable early-risk detection. From a sustainability perspective, the results suggest potential design-level contributions to improving accessibility and reducing delays in early identification processes. Full article

Background/Objectives: Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the leading cause of chronic liver disease globally, mirroring the increasing prevalence of obesity, insulin resistance, and type 2 diabetes. Early detection of hepatic steatosis is vital for cardiometabolic risk assessment; however, conventional imaging is costly and impractical for population screening. This study aimed to develop interpretable machine-learning models to predict ultrasound-detected hepatic steatosis within the MASLD spectrum using routinely available clinical and biochemical data. Methods: We analyzed data from 644 adults, 50% of whom had ultrasound-detected hepatic steatosis. Preprocessing, imputation, and feature selection were implemented within a single scikit-learn pipeline to avoid information leakage. An Elastic Net-regularized logistic regression identified the top 20 predictors, which were subsequently used across nine supervised machine learning (ML) classifiers. Model performance was evaluated via repeated stratified 5-fold cross-validation (25 resamples) using accuracy, F1 score, sensitivity, specificity, Youden’s J, balanced accuracy, and Area Under the Receiver Operating Characteristic Curve (AUROC). Interpretability was assessed using SHapley Additive exPlanations (SHAP). Results: Participants with ultrasound-detected hepatic steatosis exhibited greater adiposity, insulin resistance, and dyslipidemia compared with controls [p < 0.05 for body mass index (BMI), waist circumference, glucose, glycated hemoglobin (HbA1c), triglycerides]. Elastic Net selection highlighted Weight, Ponderal Index, Fibrosis-4 Index (FIB-4), blood urea nitrogen (BUN)/Creatinine ratio, Aspartate Aminotransferase to Platelet Ratio Index (APRI), and Visceral Adiposity Index as the strongest predictors. Logistic Regression and Gradient Boosting achieved the best performance (accuracy = 0.65 ± 0.03; AUROC = 0.71 ± 0.04; balanced accuracy = 0.66 ± 0.06), outperforming rule-based indices such as Fatty Liver Index (FLI) and Hepatic Steatosis Index (HSI) reported in the literature. SHAP analysis confirmed clinically coherent feature effects, with higher anthropometric and hepatic injury indices increasing the predicted probability of ultrasound-detected hepatic steatosis. Conclusions: Routinely available clinical and biochemical parameters can predict hepatic steatosis with moderate accuracy using transparent, interpretable ML models. Logistic Regression and Gradient Boosting provided best discrimination and robust internal performance, offering a pragmatic, low-cost approach for early identification of ultrasound-detected hepatic steatosis within the MASLD spectrum in primary and metabolic care settings. Full article