Search Results (19,848)

Cardiovascular disease remains the leading cause of morbidity and mortality worldwide, especially in regions like Eastern Europe, South Asia, and Latin America. A significant portion of these cases (80%) is linked to atherosclerosis, which can lead to severe conditions like ischemic heart disease and stroke, with atherosclerosis (ATS) responsible for the majority of cases. This review explores the multifaceted relationship between insulin resistance (IR) and ATS, highlighting their roles as both independent and interrelated contributors to cardiovascular risk. ATS is characterized by lipid accumulation and chronic inflammation within arterial walls, driven by factors such as hypertension, dyslipidemia, and genetic predisposition, with endothelial dysfunction as a key early event. The early detection of subclinical ATS is critical and can be achieved through a combination of non-invasive imaging techniques—such as coronary artery calcium scoring and carotid ultrasound—and comprehensive risk profiling. IR, marked by impaired glucose uptake in liver, muscle, and adipose tissue, often precedes early diabetes and is associated with metabolic disturbances, including dyslipidemia and chronic inflammation. The diagnosis of IR relies on surrogate indices such as HOMA-IR, the QUICKI, and the TyG index, which facilitate screening in clinical practice. Compelling evidence indicates that IR independently predicts the progression of atherosclerotic plaques, even in non-diabetic individuals, and operates through both traditional risk factors and direct vascular effects. Understanding and targeting the IR–ATS axis is essential for the effective prevention and management of cardiovascular disease. Full article

Purpose: To investigate associations between 18F-FDG-PET/CT semiquantitative and radiomic features with pathologic axillary lymph node (ALN) status in stages I–III breast cancer patients. Methods: A search was conducted across MEDLINE, EMBASE, and CENTRAL databases. Quality assessment was performed with QUADAS-2 and the radiomics quality score (RQS). Descriptive statistical analysis was performed. Results: Most studies were retrospective cohort studies (27/28) and reported only on semiquantitative features (26/28). Most studies were at high risk of bias in patient selection (22/28) and feature extraction (26/28). Semiquantitative features included maximum standardized uptake value (SUVmax), metabolic tumour volume (MTV), and total lesion glycolysis (TLG). Although associations between tumour semiquantitative features and ALN status were reported, the mean/median reported values of tumour SUVmax (3.2–8.6 vs. 2.4–9.4), MTV (2.7–19.2 vs. 1.9–10.5), and TLG (10.6–59.3 vs. 5.6–29.6) in ALN+ vs. ALN− patients were inconsistent between studies. Fourteen studies reported a significantly higher ALN SUVmax in ALN+ patients. Two studies developed models using tumour radiomic features with high accuracy for predicting ALN metastases (81.2% and 80%) but scored low on the RQS. Conclusions: Feature-based analysis of PET/CT demonstrates potential for predicting pathologic ALN status in breast cancer patients. However, establishing a clinically meaningful relationship requires higher quality evidence. Full article

In modern digital transactions involving government agencies, financial institutions, and commercial enterprises, reliable identity verification is essential to ensure security and trust. Traditional methods, such as submitting photocopies of ID cards, are increasingly susceptible to identity theft and fraud. To address these challenges, this study proposes a novel and robust identity verification framework that integrates super-resolution preprocessing, a convolutional neural network (CNN), and Monte Carlo dropout-based Bayesian uncertainty estimation for enhanced facial recognition in electronic know your customer (e-KYC) processes. The key contribution of this research lies in its ability to handle low-resolution and degraded facial images simulating real-world conditions where image quality is inconsistent while providing confidence-aware predictions to support transparent and risk-aware decision making. The proposed model is trained on facial images resized to 24 × 24 pixels, with a super-resolution module enhancing feature clarity prior to classification. By incorporating Monte Carlo dropout, the system estimates predictive uncertainty, addressing critical limitations of conventional black-box deep learning models. Experimental evaluations confirmed the effectiveness of the framework, achieving a classification accuracy of 99.7%, precision of 99.2%, recall of 99.3%, and an AUC score of 99.5% under standard testing conditions. The model also demonstrated strong robustness against noise and image blur, maintaining reliable performance even under challenging input conditions. In addition, the proposed system is designed to comply with international digital identity standards, including the Identity Assurance Level (IAL) and Authenticator Assurance Level (AAL), ensuring practical applicability in regulated environments. Overall, this research contributes a scalable, secure, and interpretable solution that advances the application of deep learning and uncertainty modeling in real-world e-KYC systems. Full article

This work introduces the use of Bayesian Neural Networks (BNNs) for inferring chlorophyll-a concentration ([CHL-a]) from remotely sensed data. BNNs are probabilistic models that associate a probability distribution to the neural network parameters and rely on Bayes’ rule for training. The performance of the proposed probabilistic model is compared to that of standard ocean color algorithms, namely ocean color 4 (OC4) and ocean color index (OCI). An extensive in situ bio-optical dataset was used to train and validate the ocean color models. In contrast to established methods, the BNN allows for enhanced modeling flexibility, where different variables that affect phytoplankton phenology or describe the state of the ocean can be used as additional input for enhanced performance. Our results suggest that BNNs perform at least as well as established methods, and they could achieve 20–40% lower mean squared errors when additional input variables are included, such as the sea surface temperature and its climatological mean alongside the coordinates of the prediction. The BNNs offer means for uncertainty quantification by estimating the probability distribution of [CHL-a], building confidence in the [CHL-a] predictions through the variance of the predictions. Furthermore, the output probability distribution can be used for risk assessment and decision making through analyzing the quantiles and shape of the predicted distribution. Full article

Background and Objectives: Maternal dyslipidaemia and low-grade inflammation are recognised drivers of in utero vascular remodelling, yet composite dynamic markers that integrate lipid–glycaemic, inflammatory and endothelial signals have not been evaluated. We investigated whether eight-week trajectories in the triglyceride–glucose index (TyG), interleukin-6 (IL-6) and flow-mediated dilation (FMD) outperform single-timepoint lipids for predicting fetal aortic remodelling. Materials and Methods: In a prospective repeated-measures study, 90 singleton pregnancies were examined at 24–26 weeks (Visit-1) and 32–34 weeks (Visit-2). At each visit, we obtained fasting lipids, TyG index, hsCRP, IL-6, oxidative-stress markers (MDA, NOx), brachial flow-mediated dilation (FMD), carotid IMT and uterine-artery Doppler, together with advanced fetal ultrasonography (abdominal-aorta IMT, ventricular strain, Tei-index, fetal pulse-wave velocity). Mothers were grouped by k-means clustering of the visit-to-visit change (Δ) in TG, TyG, hsCRP, IL-6 and FMD into three Metabolic-Inflammatory Response Phenotypes (MIRP-1/2/3). Linear mixed-effects models and extreme-gradient-boosting quantified associations and predictive performance. Results: Mean gestational TG rose from 138.6 ± 14.1 mg/dL to 166.9 ± 15.2 mg/dL, TyG by 0.21 ± 0.07 units and FMD fell by 1.86 ± 0.45%. MIRP-3 (“Metabolic + Inflammatory”; n = 31) showed the largest change (Δ) Δ-hsCRP (+0.69 mg/L) and Δ-FMD (–2.8%) and displayed a fetal IMT increase of +0.17 ± 0.05 mm versus +0.07 ± 0.03 mm in MIRP-1 (p < 0.001). Mixed-effects modelling identified Δ-TyG (β = +0.054 mm per unit), Δ-IL-6 (β = +0.009 mm) and Δ-FMD (β = –0.007 mm per %) as independent determinants of fetal IMT progression. An XGBoost model incorporating these Δ-variables predicted high fetal IMT (≥90th percentile) with AUROC 0.88, outperforming logistic regression (AUROC 0.74). Conclusions: A short-term surge in maternal TyG, IL-6 and endothelial dysfunction delineates a high-risk phenotype that doubles fetal aortic wall thickening and impairs myocardial performance. Composite dynamic indices demonstrated superior predictive value compared with individual lipid markers. Full article

The lack of transparency in many AI systems continues to hinder their adoption in critical domains such as healthcare, finance, and autonomous systems. While recent explainable AI (XAI) methods—particularly those leveraging large language models—have enhanced output readability, they often lack traceable and verifiable reasoning that is aligned with domain-specific logic. This paper presents Nomological Deductive Reasoning (NDR), supported by Nomological Deductive Knowledge Representation (NDKR), as a framework aimed at improving the transparency and auditability of AI decisions through the integration of formal logic and structured domain knowledge. NDR enables the generation of causal, rule-based explanations by validating statistical predictions against symbolic domain constraints. The framework is evaluated on a credit-risk classification task using the Statlog (German Credit Data) dataset, demonstrating that NDR can produce coherent and interpretable explanations consistent with expert-defined logic. While primarily focused on technical integration and deductive validation, the approach lays a foundation for more transparent and norm-compliant AI systems. This work contributes to the growing formalization of XAI by aligning statistical inference with symbolic reasoning, offering a pathway toward more interpretable and verifiable AI decision-making processes. Full article

The prediction of agricultural commodity futures returns is crucial for understanding global economic trends, alleviating inflationary pressures, and optimizing investment portfolios. However, current research that uses full-sample decomposition to predict agricultural futures returns suffers from data leakage, and the resulting forecast bias leads to overly optimistic outcomes. Additionally, previous studies have lacked a comprehensive consideration of key economic variables that influence agricultural prices. To address these issues, this study proposes the “Rolling VMD-LASSO-Mixed Ensemble” forecasting framework and compares its performance with “Rolling VMD” against univariate models, “Rolling VMD-LASSO” against “Rolling VMD”, and “Rolling VMD-LASSO-Mixed Ensemble” against “Rolling VMD-LASSO”. Empirical results show that, on average, “Rolling VMD” improved MSE, MAE, Theil U, ARV, and DA by 3.05%, 1.09%, 1.52%, 2.96%, and 11.11%, respectively, compared to univariate models. “Rolling VMD-LASSO” improved these five indicators by 2.11%, 1.15%, 1.09%, 2.13%, and 1.00% over “Rolling VMD”. The decision tree-based “Rolling VMD-LASSO-Mixed Ensemble” outperformed “Rolling VMD-LASSO” by 1.98%, 0.96%, 1.28%, 2.55%, and 4.18% in the five metrics. Furthermore, the daily average return, maximum drawdown, Sharpe ratio, Sortino ratio, and Calmar ratio based on prediction results also show that “Rolling VMD” outperforms univariate forecasting, “Rolling VMD-LASSO” outperforms “Rolling VMD”, and “Rolling VMD-LASSO-Mixed Ensemble” outperforms “Rolling VMD-LASSO”. This study provides a more accurate and robust forecasting framework for the global agricultural futures market, offering significant practical value for investor risk management and policymakers in stabilizing prices. Full article

In this work, we introduce a framework to combine arbitrary image segmentation algorithms from different agents under data privacy constraints to produce an aggregated prediction set satisfying finite-sample risk control guarantees. We leverage distribution-free uncertainty quantification techniques in order to aggregate deep neural networks for image segmentation tasks. Our method can be applied in settings to merge the predictions of multiple agents with arbitrarily dependent prediction sets. Moreover, we perform experiments in medical imaging tasks to illustrate our proposed framework. Our results show that the framework reduced the empirical false positive rate by 50% without compromising the false negative rate, with respect to the false positive rate of any of the constituent models in the aggregated prediction algorithm. Full article

The article presents an analysis of the complex mechanism of abrasion of shorelines built of non-lithified sediments as a result of rising water levels in the reservoir, along with its quantitative assessment. It allows forecasting the actual risks of coastal areas intendent for urbanization with similar morphology and geological structure. The task of the article is also to point out that for proper assessment of abrasion it is necessary to take into account the greater complexity of the mechanism in which abrasion is the result of co-occurring processes of erosion and landslides. During the analysis, the classic Kachugin method of abrasion assessment was combined with an analysis of the stability of the abraded slope, taking into account the circular slip surface (Bishop and Morgenster–Price methods) and the breaking slip surface (Sarma method). This approach required the assessment of the geotechnical properties of the soil using, among other things, advanced in situ methods such as static sounding. The results indicate that the cliff edge is in limit equilibrium or even in danger of immediate landslide. At the same time, it was possible to determine the horizontal extent of a single landslide at 1.2 to 5.8 m. In the specific cases of reservoir filling, the consideration of the simultaneous action of both failure mechanisms definitely worsens the prediction of shoreline sustainability and indicates the need to restrict construction development in the coastal zone. Full article

Objective: To develop and validate a composite model that combines lesion–symptom mapping (LSM), radiomic information, and clinical factors for predicting cerebellar mutism syndrome in pediatric patients suffering from posterior fossa tumors. Methods: A retrospective analysis was conducted on a cohort of 247 (training set, n = 174; validation set, n = 73) pediatric patients diagnosed with posterior fossa tumors who underwent surgery at Beijing Children’s Hospital. Presurgical MRIs were used to extract the radiomics features and voxel distribution features. Clinical factors were derived from the medical records. Group comparison was used to identify the clinical risk factors of CMS. Combining location weight, radiomic features from tumor area and the significant intersection area, and clinical variables, hybrid models were developed and validated using multiple machine learning models. Results: The mean age of the cohort was 4.88 [2.89, 7.78] years, with 143 males and 104 females. Among them, 73 (29.6%) patients developed CMS. Gender, location, weight, and five radiomic features (three in the tumor mask area and two in the intersection area) were selected to build the model. The four models, KNN model, GBM model, RF model, and LR model, achieved high predictive performance, with AUCs of 0.84, 0.83, 0.81, and 0.87, respectively. Conclusions: CMS can be predicted using MRI features and clinical factors. The combination of radiomics and tumoral location weight could improve the prediction of CMS. Full article

Educational data mining (EDM) is instrumental in the early detection of students at risk of academic underperformance, enabling timely and targeted interventions. Given that many undergraduate students face challenges leading to high failure and dropout rates, utilizing EDM to analyze student data becomes crucial. By predicting academic success and identifying at-risk individuals, EDM provides a data-driven approach to enhance student performance. However, accurately predicting student performance is challenging, as it depends on multiple factors, including academic history, behavioral patterns, and health-related metrics. This study aims to bridge this gap by proposing a deep learning model to predict student academic performance with greater accuracy. The approach combines a convolutional neural network (CNN) and a bidirectional gated recurrent unit (BiGRU) network to enhance predictive capabilities. To improve the model’s performance, we address key data preprocessing challenges, including handling missing data, addressing class imbalance, and selecting relevant features. Additionally, we incorporate optimization techniques to fine-tune hyperparameters to determine the best model architecture. Using key performance metrics such as accuracy, precision, recall, and F-score, our experimental results show that our proposed model achieves improved prediction accuracy of 97.48%, 90.90%, and 95.97% across the three datasets. Full article

Introduction: Spinal and bulbar muscular atrophy (SBMA) is an adult-onset, X-linked, progressive neuromuscular disease caused by abnormal CAG trinucleotide expansion in the androgen receptor gene. Patients with SBMA report difficulty with falls on self-reported activities of daily living scales. To our knowledge, no study has examined the relationship between falls and common clinical measures of strength, balance, mobility, and disease biomarkers. We performed a cross-sectional analysis of an SBMA cohort. Objectives: The objectives of this study are as follows: (1) compare demographics, clinical measures, and biomarkers between patients who did and did not fall; (2) determine which measures best discriminate fallers from non-fallers; and (3) identify cutoff scores to detect patients with a higher fall risk. Design: Cross-sectional analysis was used. Outcome Measures: Disease biomarkers included blood serum creatinine, and clinical measures included the Timed Up and Go (TUG), the Adult Myopathy Assessment Tool (AMAT), and posturography, including the Modified Clinical Test of Sensory Interaction on Balance and the Motor Control Test. The Maximal Voluntary Isometric Contractions (MVICs) of four lower extremity muscles were captured via fixed-frame dynamometry. Results: We identified three clinical measures that help detect fall risk in people with SBMA. A post hoc receiver operating characteristic curve analysis helped identify cut scores for each test. Impairments of mobility (TUG > 8 s), muscle endurance (AMAT endurance subscale < 14), and muscle strength (ankle plantar flexion MVIC < 45% of predicted) were different between fallers and non-fallers, via independent t-tests. Conclusions: These three clinical tests can help detect fall riskthat may help clinicians implement gait aid use or other fall prevention strategies before catastrophic falls occur. Full article

Background/Objectives: Hospital readmissions are a key quality metric impacting both patient outcomes and healthcare costs. Traditional logistic regression models, including the LACE index (Length of stay, Admission type, Comorbidity index, and recent Emergency department visits), are commonly used for readmission risk stratification, though their accuracy may be limited by non-linear interactions with other clinical variables. This study compared the predictive performance of non-linear machine learning (ML) models with stepwise logistic regression (LR) and the LACE index for predicting 30-day general medicine readmissions. Methods: We retrospectively analysed adult general medical admissions at a tertiary hospital in Australia from 1 July 2022 to 30 June 2023. Thirty-two variables were extracted from electronic medical records, including demographics, comorbidities, prior healthcare use, socioeconomic status (SES), laboratory data, and frailty (measured by the Hospital Frailty Risk Score). Predictive models included stepwise LR and four ML algorithms: Least Absolute Shrinkage and Selection Operator (LASSO), random forest, Extreme Gradient Boosting (XGBoost), and artificial neural networks (ANNs). Performance was assessed using the area under the curve (AUC), with comparisons made using DeLong’s test. Results: Of 5371 admissions, 1024 (19.1%) resulted in 30-day readmissions. Readmitted patients were older and frailer and had more comorbidities and lower SES. Logistic regression (LR) identified the key predictors of outcomes, including heart failure, alcoholism, nursing home residency, and prior admissions, achieving an AUC of 0.62. LR’s performance was comparable to that of the LACE index (AUC = 0.61) and machine learning models: LASSO (AUC = 0.63), random forest (AUC = 0.60), and artificial neural networks (ANNs) (AUC = 0.60) (p > 0.05). However, LR significantly outperformed XGBoost (AUC = 0.55) (p < 0.05). Conclusions: About one in five general medicine patients are readmitted within 30 days. Traditional LR performed as well as or better than ML models for readmission risk prediction. Full article

Background: Although complete excision of suspicious melanocytic lesions is mandatory, it carries the risk of unnecessary scarring on one hand and inadequate treatment of misdiagnosed lesions on the other. Objectives: We evaluated the sensitivity, specificity, and predictive value of reflectance confocal microscopy (RCM) in diagnosing pigmented lesions with clinically ambiguous features―the so-called “gray zone” ―and compared its performance with the more established technique of epiluminescence microscopy (ELM). Results: Between 2019 and 2020, a total of 2282 melanocytic lesions were assessed using both ELM and RCM. Histopathological diagnosis aligned with the ELM risk classification in 91.6% of melanocytic lesions, specifically in 92.0% of very-high-risk lesions, 88.5% of high-risk lesions, 66.3% of medium-risk lesions, 96.3% of low-risk lesions, and 98.0% of very low-risk lesions. Similarly, histopathological diagnosis of these lesions corresponded with the RCM risk assessment in 91.2% of cases, including 90.9% of very-high-risk lesions, 84.4% of high-risk lesions, 93.1% of medium-risk lesions, 90.5% of low-risk lesions, and 96.2% of very low-risk lesions. Conclusions: Although ELM is a valuable tool for increasing the efficacy of clinical diagnosis, its reliability decreases for a group of lesions that appear suspicious during clinical skin examination. RCM, as a newer technique, appears to improve malignancy detection in suspicious melanocytic lesions without requiring excision; its sensitivity and specificity remain high even in lesions classified by ELM as posing a medium risk of malignancy. Full article

Unrealistic cash flow forecasts negatively affect project stakeholders and are a common issue for construction practitioners. This study proposes a new method for predicting the probabilistic cash flow of a project that can automate the calculation process while considering the impact of risks and their inter-related structure. This research integrates a Bayesian Belief Network (BBN) and 5D-BIM to provide a new probabilistic cash flow analysis approach. Here, 5D-BIM is used to facilitate cash flow calculations and automate the process. The BBN has also been implemented to assess the impact of risk factors on project cash flow, considering their complex inter-related structure. In addition, a hybrid approach combining fuzzy set theory, decision-making trial and evaluation laboratory (DEMATEL), and interpretive structural modeling (ISM) is used to form the BBN. The proposed method provides a robust tool for calculating the probabilistic cash flow of the project. The results showed that the project’s cash flow in the last month was IRR 14.4 billion without considering the impact of risks. The probabilistic cash flow of the project indicates that due to the impact of the risks, the project cash flow will be in the range of IRR −142.2 billion and IRR 1.11 billion at the end of the project. This shows the possibility of experiencing between 11 and 130% deviation in the project cash flow due to existing risks. In conclusion, project cash flow is unreliable without considering the impact of risks. This framework supports better financial decisions and allows for the evaluation of cash flow risk management scenarios. Full article