Search Results (10,121)

Background/Objectives: Acute myocardial infarction with ST-segment elevation (STEMI) remains a clinical condition with high mortality. The Ion Shift Index (ISI) and Naples Prognostic Score (NPS) are two prognostic indicators that have recently come to the fore. The aim of this study is to compare the predictive value of ISI and NPS in predicting in-hospital mortality in STEMI patients. Methods: This retrospective study included 214 STEMI patients (1 January 2022–1 January 2024). Exclusion criteria included active cancer, infection, autoimmune disease, or chronic kidney disease. ISI and NPS were calculated from laboratory results obtained from the emergency department at the time of initial presentation. Patients were categorized according to in-hospital survival. Logistic regression and ROC curve analyses were performed for in-hospital mortality. Results: The mean age of participants was 64.8 ± 11.2 years, and 40.2% were female; a total of 36 patients (16.8%) died during hospitalization. Hypertension and female gender were more common in those who died, and LDL cholesterol and inflammatory markers were higher. The ISI value was significantly increased in the mortality group, whereas no significant difference was observed in NPS. ROC analysis revealed that at a threshold value of 3.0, ISI had a sensitivity of 68% and specificity of 71%, with an area under the curve (AUC) of 0.70, while NPS had an AUC of 0.55 and did not demonstrate significant discriminatory power. In the multivariate analysis, ISI and increased LDL cholesterol were independently associated with mortality; decreased lymphocyte/monocyte ratio and female gender were also additional independent predictors. NPS did not emerge as an independent factor in predicting in-hospital mortality. Conclusions: ISI was found to be a superior and independent early risk predictor of in-hospital mortality in STEMI patients compared to NPS. ISI may serve as a rapid and inexpensive risk classification tool in the acute phase, as it reflects sudden changes in intracellular–extracellular ion balance, whereas NPS may not be sufficiently sensitive in the hyperacute phase, as its components reflect chronic nutritional and inflammatory states. Due to limitations such as a single-center retrospective design and low mortality rates, validation through multicenter prospective studies is required for the integration of ISI into clinical practice. Full article

Background/Objectives: Reflecting the interaction between dissolution and absorption, the biphasic dissolution system is an appealing approach for estimating the intestinal absorption of drugs in humans. The study aims to characterize the suitability of the biphasic in vitro dissolution testing to set up an in vitro–in vivo correlation (IVIVC) for the original and generic immediate-release (IR) tablets of a Biopharmaceutics Classification System (BCS) Class II drug, bicalutamide (BIC). Methods: USP apparatus II paddle was used to conduct dissolution testing. A level A IVIVC was obtained between in vitro partitioning and in vivo absorption data of the original drug. The single-compartmental modeling was used for pharmacokinetic (PK) analysis. The generic product’s plasma concentrations were estimated. Results: There was a good correlation between in vitro and in vivo data (r² = 0.98). The area under the concentration–time curve (AUC) and maximum plasma concentration (C_max) ratios for generic/original were 1.04 ± 0.01 and 0.951 ± 0.026 (mean ± SD), respectively. Conclusions: The biphasic dissolution testing may present an in vivo predictive tool for developing generic products of poorly soluble and highly permeable drugs such as BIC, which are characterized by pH-independent poor solubility. Full article

Accurately modelling and simulating the stiffness modulus of asphalt mixtures is essential for reliable pavement design and performance prediction under varying environmental and loading conditions. The preceding is commonly achieved through master curves, which relate stiffness to loading frequency at a reference temperature. However, conventional master curves face two primary limitations. Firstly, temperature is not treated as a state variable; instead, its effect is indirectly considered through shift factors, which can introduce inaccuracies due to their lack of thermodynamic consistency across the entire range of possible temperatures. Secondly, conventional master curves often encounter convergence difficulties when calibrated with experimental data constrained to a narrow frequency spectrum. In order to address these shortcomings, this investigation proposes a novel formulation known as the Thermo-Stiffness Integration (TSI) model, which explicitly incorporates both temperature and frequency as state variables to predict the stiffness modulus directly, without relying on supplementary expressions such as shift factors. The TSI model is built on thermodynamics-based principles (such as Eyring’s rate theory and activation free energy) and leverages the time–temperature superposition principle to create a physically consistent representation of the mechanical behaviour of asphalt mixtures. This manuscript presents the development of the TSI model along with its application in a case study involving eight asphalt mixtures, including four hot-mix asphalts and four warm-mix asphalts. Each type of mixture contains recycled concrete aggregates at replacement levels of 0%, 15%, 30%, and 45% as partial substitutes for coarse natural aggregates. This diverse set of materials enables a robust evaluation of the model’s performance, even under non-traditional mixture designs. For this case study, the TSI model enhances computational stability by approximately 4 to 45 times compared to conventional master curves. Thus, the main contribution of this research lies in establishing a valuable mathematical tool for both scientists and practitioners aiming to improve the design and performance assessment of asphalt mixtures in a more physically realistic and computationally stable approach. Full article

Objective: This study aimed to investigate the prognostic value of the preoperative monocyte-to-high-density lipoprotein cholesterol ratio (MHR) and clinicopathological parameters for predicting survival outcomes in patients undergoing curative-intent gastrectomy for gastric adenocarcinoma. Methods: This retrospective cohort study analyzed data from 304 patients with histopathologically confirmed gastric adenocarcinoma who underwent curative-intent gastrectomy with standardized D1+ or D2 lymphadenectomy. The MHR was calculated using preoperative monocyte counts and HDL cholesterol levels. Patients were dichotomized based on the optimal MHR cutoff determined via receiver operating characteristic curve analysis with the Youden index. Survival outcomes, including overall survival (OS) and progression-free survival (PFS), were assessed using Kaplan–Meier analysis and compared with log-rank tests. Results: ROC analysis determined an optimal MHR cutoff of ≥11.02 (AUC: 0.654; 95% CI: 0.59–0.718), yielding sensitivities and specificities of 62.6% and 62.4%, respectively. Patients with an elevated MHR (≥11.02) had worse 5-year OS (51.4 vs. 72.2%; p < 0.001) and PFS (65.2 vs. 80.5%; p = 0.003). In the multivariate Cox regression model, elevated MHR emerged as an independent predictor of disease progression (HR: 1.93; 95% CI: 1.17–3.18; p = 0.010), while parameters such as signet ring cell histology, lymphovascular invasion, and perineural invasion were significant in univariate analyses but not in the adjusted multivariate model. Conclusions: MHR should not be regarded as a definitive predictor in isolation but rather as a cost-effective, readily obtainable adjunct within a broader preoperative risk assessment framework. Integration with other inflammation-based and clinicopathological factors may enhance predictive performance and clinical applicability. Full article

Ensuring the safety, reliability, and longevity of Lithium-ion (Li-ion) batteries is crucial for sustainable integration of Electric Vehicles (EVs) within Intelligent Transportation Systems (ITSs). However, thermal stress and degradation-induced anomalies can cause sudden performance failures, posing critical operational and safety risks. Capturing complex, non-linear, and high-dimensional patterns remains challenging for traditional Machine Learning (ML) models. We propose a hybrid anomaly detection method that incorporates a Variational Quantum Neural Network (VQNN), which uses the principles of quantum mechanics, such as superposition, entanglement, and parallelism, to learn complex non-linear patterns. The VQNN is integrated with Isolation Forest (IF) and a Median Absolute Deviation (MAD)-based spike characterisation method to form a Quantum Anomaly Detector (QAD). This method distinguishes between normal and anomalous spikes in battery behaviour. Using an Arrhenius-based model, we simulate how the State of Health (SoH) and voltage of a Li-ion battery reduce as temperatures increase. We perform experiments on NASA battery datasets and detect abnormal spikes in 14 out of 168 cycles, corresponding to $8.3$% of the cycles. The QAD achieves the highest Receiver Operating Characteristic Area Under the Curve (ROC-AUC) of $0.9820$, outperforming the baseline IF model by $7.78\%$. We use ML to predict the SoH and voltage changes when the temperature varies. Gradient Boosting (GB) achieves a voltage Mean Squared Error (MSE) of $0.001425$, while Support Vector Regression (SVR) achieves the highest $R^2$ score of $0.9343$. These results demonstrate that Quantum Machine Learning (QML) can be applied for anomaly detection in Battery Management Systems (BMSs) within intelligent transportation ecosystems and could enable EVs to autonomously adapt their routing and schedule preventative maintenance. With these capabilities, safety will be improved, downtime minimised, and public confidence in sustainable transport technologies increased. Full article

T-bar penetrometers have been widely used to measure strength parameters of marine clay in laboratory and in situ tests. However, using the deep resistance factor derived from full-flow conditions to evaluate the undrained shear strength of shallow clay layers may lead to significant underestimation. Furthermore, the deep resistance factor is inherently influenced by the strain-softening behavior of clay rather than maintaining the constant value (typically 10.5) as conventionally assumed in practice. To address this issue, large-deformation finite element (LDFE) simulations incorporating an advanced exponential strain-softening constitutive model were performed to replicate the full T-bar penetration process—from shallow embedment to deeper depths below the mudline. A series of parametric studies were conducted to examine the influence of key parameters on the resistance factor and the associated failure mechanisms during penetration. Based on numerical results, empirical formulas were derived to predict critical penetration depths for both trapped cavity formation and full-flow mechanism initiation. For penetration depths shallower than the full-flow depth, an expression for the softening correction factor was developed to calibrate the shallow resistance factor. Finally, combined with global optimization algorithms, a computer-aided back-analysis procedure was established to estimate strain-softening parameters using resistance-penetration curves from initial penetration tests in marine clay. The reliability of the back-analysis procedure was validated through extensive comparisons with a series of numerical simulation results. This procedure can be applied to the interpretation of T-bar in situ test results in soft marine clay, enabling the evaluation of its strain-softening behavior. Full article

Background/Objectives: Gallstone disease, a prevalent and costly digestive system disorder, is influenced by multifactorial risk factors, some of which may predispose to malignancy. This study aims to evaluate the association between gallstone disease and malignancy using advanced machine learning (ML) algorithms. Methods: A dataset comprising approximately 1000 patients was analyzed, employing six ML methods: random forests (RFs), support vector machines (SVMs), multi-layer perceptron (MLP), MLP with PyTorch 2.3.1 (MLP_PT), naive Bayes (NB), and Tabular Prior-data Fitted Network (TabPFN). Comparative performance was assessed using Pearson correlation, sensitivity, specificity, Kappa, receiver operating characteristic (ROC), area under curve (AUC), and accuracy metrics. Results: Our results revealed that age, body mass index (BMI), and history of HRT were the most significant predictors of malignancy. Among the ML models, TabPFN emerged as the most effective, achieving superior performance across multiple evaluation criteria. Conclusions: This study highlights the potential of leveraging cutting-edge ML methodologies to uncover complex relationships in clinical datasets, offering a novel perspective on gallstone-related malignancy. By identifying critical risk factors and demonstrating the efficacy of TabPFN, this research provides actionable insights for predictive modeling and personalized patient management in clinical practice. Full article

Pre-stacked uncured metal–fiber-reinforced polymer (FRP) laminates, which are critical for aerospace components like double-curved fuselage panels, wing ribs, and engine nacelles, exhibit better deformation behavior than their fully cured counterparts. However, accurate process simulation requires precise material characterization and process optimization to achieve a defect-free structural design. This study focuses on two core material behaviors of uncured laminates—inter-ply friction at metal–prepreg interfaces and out-of-plane bending—and optimizes process parameters for their non-coherent deformation. Experimental tests included double-lap sliding tests (to quantify inter-ply friction) and clamped-beam bending tests (to characterize out-of-plane bending); a double-curved dome part was designed to assess the effects of the material constituent, fiber orientation, inter-ply friction, and clamping force, with validation via finite element modeling (FEM) in Abaqus software. The results indicate that the static–kinetic friction model effectively predicts inter-ply friction behavior, with numerical friction coefficient–displacement trends closely matching experimental data. Additionally, the flexural bending model showed that greater plastic deformation in metal layers increased bending force while reducing post-unloading spring-back depth. Furthermore, for non-coherent deformation, higher clamping forces improve FRP prepreg deformation and mitigate buckling, but excessive plastic deformation raises metal cracking risk. This work helps establish a combined experimental–numerical framework for the defect prediction and process optimization of complex lightweight components, which address the core needs of modern aerospace manufacturing. Full article

Background/Objectives: Perioperative organ injury (POI) is frequently observed following hepatectomy as acute kidney injury (AKI), perioperative myocardial injury (PMI), or both. We aimed to determine the incidences of POI, PMI, and AKI, reveal the risk factors and predictive tools for POI occurrence, and evaluate the relationship between POI and patient outcomes. Methods: This was a single-center historical cohort study of consecutive patients. The primary endpoint was the occurrence of POI within 3 days following hepatectomy. Results: Out of 128 patients, POI, PMI, and AKI occurred in 48 (37.5%), 36 (28.1%), and 23 (18%) patients, respectively. Ten (7.8%) patients suffered from both PMI and AKI. The presence of chronic kidney disease or systolic/valvular heart disease, fluid balance more than 365 mL/h, and intraoperative bleeding more than 950 mL were the risk factors for POI. A tool created by using the intraoperative decline of central venous oxygen saturation and lactate value during skin closure performed well in predicting POI (area under the ROC curve: 0.79, p < 0.001). In patients with POI, the number of those who needed intensive care unit (ICU) follow-up for more than 1 day was significantly higher (21% vs. 6%, p: 0.01). The length of hospital stay for these patients was significantly longer as well (11 (8–18) vs. 9 (7–13) days, p: 0.02). Two patients (20% of 10 patients who suffered from both AKI and PMI) died in the 90-day follow-up. Conclusions: POI is a common complication following hepatectomy and is associated with longer hospital and ICU stays. Patients who suffer from both AKI and PMI have a higher risk of mortality. Full article

Objectives: This study aims to explore the role of radiomics features (RFs) from prostate subregions, including the tumor microenvironment (TME), in predicting persistent PSA. Methods: In retrospective analysis, we segregated 354 patients with pathologically confirmed localized prostate cancer (PCa) into training, internal validation, and external validation cohorts. The prostate on ¹⁸F-prostate-specific membrane antigen (PSMA)-1007 positron emission tomography/computed tomography (PET/CT) was partitioned into three zones based on the maximum standardized uptake value (SUVmax) (zone-intra: 45–100% SUVmax; zone-peri: 20–45% SUVmax; zone-norm: 0–20% SUVmax). RFs from these zones were harnessed to develop five radiomics models [model-intra; model-peri; model-norm; model-ip; model-ipn]. Three optimal radiomics models were further integrated with the PSA model to construct combined models. Model performance was evaluated using the receiver operating characteristic (ROC) curves and the area under the curve (AUC). Results: Utilizing least absolute shrinkage and selection operator (LASSO) and logistic regression, five radiomics models were constructed, with model-ip, model-ipn, and model-intra showing superior performance [training cohort AUCs: 0.76 (0.68–0.83), 0.75 (0.68–0.83), 0.76 (0.68–0.83); internal validation cohort AUCs: 0.76 (0.65–0.88), 0.72 (0.57–0.86), 0.70 (0.55–0.86); external validation cohort AUCs: 0.70 (0.50–0.86), 0.55 (0.36–0.73), 0.53 (0.34–0.72)]. Notably, the combined model incorporating model-ip and the PSA model exhibited optimal performance [training cohort AUC: 0.78 (0.71–0.85); internal validation cohort AUC: 0.78 (0.67–0.90); external validation cohort AUC: 0.89 (0.72–0.98)]. Conclusions: The RFs in different subregions on ¹⁸F-PSMA-1007 PET/CT have varying effectiveness in predicting persistent PSA. A radiomics model that encompasses the 20–45% SUVmax and 45–100% SUVmax zones, when combined with the PSA model, enhances predictive accuracy. Full article

Stroke is a major global cause of death and disability, and improving outcomes requires early prediction. Although class imbalance in datasets causes biased predictions and inferior classification accuracy, machine learning (ML) techniques have shown potential in stroke prediction. We used the Synthetic Minority Oversampling Technique (SMOTE) to balance datasets and lessen bias in order to address these problems. Furthermore, we suggested a method that combines a linear discriminant analysis (LDA) model for classification with an autoencoder for feature extraction. A grid search approach was used to optimize the hyperparameters of the LDA model. We used criteria like accuracy, sensitivity, specificity, AUC (area under the curve), and ROC (Receiver Operating Characteristic) to guarantee a strong evaluation. With 98.51% sensitivity, 97.56% specificity, 99.24% accuracy, and 98.00% balanced accuracy, our model demonstrated remarkable performance, indicating its potential to improve stroke prediction and aid in clinical decision-making. Full article

Background: Diabetes remains a major global health challenge, contributing significantly to premature mortality due to its potential progression to organ failure if not diagnosed early. Traditional diagnostic approaches are subject to human error, highlighting the need for modern computational techniques in clinical decision support systems. Although these systems have successfully integrated deep learning (DL) models, they still encounter several challenges, such as a lack of intricate pattern learning, imbalanced datasets, and poor interpretability of predictions. Methods: To address these issues, the temporal inception perceptron network (TIPNet), a novel DL model, is designed to accurately predict diabetes by capturing complex feature relationships and temporal dynamics. An adaptive synthetic oversampling strategy is utilized to reduce severe class imbalance in an extensive diabetes health indicators dataset consisting of 253,680 instances and 22 features, providing a diverse and representative sample for model evaluation. The model’s performance and generalizability are assessed using a 10-fold cross-validation technique. To enhance interpretability, explainable artificial intelligence techniques are integrated, including local interpretable model-agnostic explanations and Shapley additive explanations, providing insights into the model’s decision-making process. Results: Experimental results demonstrate that TIPNet achieves improvement scores of 3.53% in accuracy, 3.49% in F1-score, 1.14% in recall, and 5.95% in the area under the receiver operating characteristic curve. Conclusions: These findings indicate that TIPNet is a promising tool for early diabetes prediction, offering accurate and interpretable results. The integration of advanced DL modeling with oversampling strategies and explainable AI techniques positions TIPNet as a valuable resource for clinical decision support, paving the way for its future application in healthcare settings. Full article

In software development, software requirement engineering (SRE) is an essential stage that guarantees requirements are clear and unambiguous. However, incomplete inconsistency, and ambiguity in requirement documents often occur, which can cause project delay, cost escalation, or total failure. In response to these challenges, this paper introduces a machine learning method to automatically identify the risk levels of software requirements according to ensemble classification methods. The labeled textual requirement dataset was preprocessed utilizing conventional preprocessing techniques, label encoding, and oversampling with the synthetic minority oversampling technique (SMOTE) to handle class imbalance. Various ensemble and baseline models such as extra trees, random forest, bagging with decision trees, XGBoost, LightGBM, gradient boosting, decision trees, support vector machine, and multi-layer perceptron were trained and compared. Five-fold cross-validation was used to provide stable performance evaluation on accuracy, area under the ROC curve (AUC), F1-score, precision, recall, root mean square error (RMSE), and error rate. The bagging (DT) classifier achieved the best overall performance, with an accuracy of 99.55%, AUC of 0.9971 and an F1-score of 97.23%, while maintaining a low RMSE of 0.03 and error rate of 0.45%. These results demonstrate the effectiveness of ensemble-based classifiers, especially bagging (DT) classifiers, in accurately predicting high-risk software requirements. The proposed method enables early detection and mitigation of requirement risks, aiding project managers and software engineers in improving resource planning, reducing rework, and enhancing overall software quality. Full article

The G Block of Daqing Oilfield is a crucial area for sustainable development and stable production. In addressing the technical bottlenecks of high-resolution logging data interpretation for reservoir evaluation in the Block, this study proposes a resistivity curve prediction method based on machine learning algorithms. Traditional interpretation models relying on DLS logging data face two major challenges when applied to 0.2 m high-resolution logging: first, the interpreted effective thickness of the reservoir tends to be overestimated, and second, the accuracy of fluid property identification declines. Additionally, the lack of corresponding well-test data for new logging datasets further constrains the development of interpretation models. To tackle these challenges, this study employs the XGBoost algorithm to construct a high-precision resistivity prediction model. Through systematic analysis of various logging parameter combinations, the optimal feature set comprising HAC, MSFL, and GR curves was identified. Training and testing results demonstrate that the model achieves a mean absolute error (MAE) of 0.94 Ω·m and a root mean square error (RMSE) of 1.79 Ω·m in predicting resistivity. After optimization, the model’s performance improved significantly, with MAE and RMSE reduced to 0.75 Ω·m and 1.31 Ω·m, respectively. To evaluate the model’s reliability, an external validation test was conducted on Well GFX2, yielding MAE and RMSE values of 0.91 Ω·m and 1.43 Ω·m, confirming the model’s strong generalization capability. Furthermore, the RLLD-AC and RLLD-DEN crossplots constructed from the predicted results exhibit excellent fluid identification performance in practical applications, achieving an accuracy rate exceeding 89%, which aligns well with production test data. The findings of this study provide new technical support for fine reservoir characterization in the study area and offer significant practical guidance for development plan adjustments. Full article

Background: Chronic heart failure (CHF) continues to present significant prognostic challenges despite advances in diagnosis and therapy. While the N-terminal prohormone of brain natriuretic peptide (NT-proBNP) is widely recognized as a key marker of cardiac stress, and serum albumin reflects systemic inflammation and nutritional status, their integration into a single parameter—the NT-proBNP-to-albumin ratio (NTAR)—may improve risk stratification. This study aimed to evaluate the NTAR as a novel biomarker for predicting in-hospital mortality in patients with CHF. Methods: We performed an exploratory, retrospective, observational, single-center study involving 542 patients (306 males) admitted for CHF between January 2022 and August 2024. NTAR was calculated as log₁₀(NT-proBNP/albumin). Statistical analyses included ROC curves, univariate and multivariable Cox regression, and Kaplan–Meier survival analysis. Sex-specific performance of NTAR was compared against NT-proBNP and serum albumin alone. Results: Females had significantly lower serum albumin levels than males, while NT-proBNP levels were similar across sexes. NTAR increased with NYHA functional class and was highest in patients with heart failure with reduced ejection fraction (HFrEF). NTAR showed very good discriminatory performance for predicting in-hospital mortality (AUC = 0.840, 95% CI: 0.794–0.879, p < 0.001), marginally but statistically outperforming NT-proBNP in the male subgroup. In univariate Cox regression analyses, higher serum albumin was significantly associated with reduced in-hospital mortality risk in males (HR = 0.352; 95% CI: 0.154–0.803; p = 0.010) and females (HR = 0.169; 95% CI: 0.072–0.399; p < 0.001). Elevated NT-proBNP levels were associated with increased mortality risk in males (HR = 8.627; 95% CI: 1.956–38.042; p < 0.001) and females (HR = 6.060; 95% CI: 1.498–24.521; p = 0.002) with similar findings in NTAR (HR_males = 10.318, 95% CI: 2.452–43.417, p < 0.001 and HR_females = 7.542, 95% CI: 1.874–30.358, p < 0.001). Multivariable analysis identified NTAR as the strongest independent predictor for in-hospital mortality among males. Conclusions: These findings suggest that NTAR effectively integrates cardiac and systemic dysfunction to improve mortality risk stratification in CHF, particularly in male patients. Its ease of calculation from routinely available biomarkers supports its clinical applicability. Full article