Intracranial steno-occlusive lesions are characteristic of moyamoya disease (MMD), but increasing reports of extracranial vascular involvement suggest a possible systemic arteriopathy. We conducted a PRISMA-guided systematic review (1974-May 2025) to identify extracranial vascular involvement in classical MMD. Seventy-four studies were included. Because the search was limited to English-language publications, language bias may be present. A subset of 85 patients from 46 studies was used for exploratory multilabel machine-learning and deep-learning modeling based on clinical variables. Seventy-four studies comprising 143 patients (59.4% female) were included; mean age was 24.9years. Bilateral intracranial disease predominated. Among patients with isolated extracranial involvement, coronary lesions were most frequent (33.1%), followed by renal (31.5%), pulmonary (14.6%), external carotid (10.8%), vertebral (4.6%), and celiac/mesenteric (3.1%) involvement; other rare single-vessel lesions accounted for 2.3%. Multisite involvement occurred in 9.1%. Of 25 genotyped patients, 23 (92.0%) carried RNF213 variants. Mortality was highest in the pulmonary subgroup (31.6%). Logistic regression and random forest showed the best overall balance of accuracy, F1 score, and ROC area, whereas one-dimensional convolutional neural networks were the strongest deep-learning models, although deep learning overall underperformed machine learning. Feature selection identified vascular risk factors and bilateral MMD as the strongest predictors. Extracranial vascular involvement in MMD appears age- and vascular bed-specific. These findings are consistent with, but do not establish, a systemic arteriopathy framework and support further prospective validation.

Insomnia and nocturia are prevalent, highly comorbid conditions and share a mutual connection. We aim to further elucidate the link between both in order to improve sleep and eventually overall health. The objective of this analysis was to construct a predictive model for nocturia, using retrospective clinical and polysomnographic data from patients with severe nocturia in line with the TUCSON project: Tackling Underlying Causes Of Sleep Related Nocturia (NCT05404828). Retrospective data was collected from all adult patients consulting a third line center for insomnia between 2019 and 2021. Data on demographics, medical history, nocturia, insomnia severity index (ISI), Pittsburgh Sleep Quality Index (PSQI), polysomnographic sleep parameters and First Uninterrupted Sleep Period (FUSP) were collected. Potentially significant variables related to nocturia were incorporated into a Classification and Regression Trees (CART) model to predict nocturia using the R "rpart" package. A total of 170 patients presenting with insomnia were analyzed, including 106 without and 64 with nocturia. The median age was 45 years (IQR 31-55), and 58.2% were women. In exploratory CART analyses, FUSP, apnea-hypopnea index (AHI), and REM sleep percentage were identified as potential predictors of nocturia. The predictive model's accuracy was 73.08%, with a sensitivity of 79.41%, specificity of 61.11%, and an ROC area of 0.82. However, logistic regression outperformed CART in internal validation (cross-validated AUC 0.76 vs. 0.52; bootstrap AUC 0.76 vs. 0.64), and both models showed poor generalization on a held-out test set. This study is the first to investigate clinical and polysomnographic predictors of nocturia within a select insomnia population. While exploratory CART modeling allowed the identification of potential non-linear relationships and clinically relevant subgroups, its predictive performance was limited and inferior to logistic regression after internal validation. Although FUSP was not consistently retained across resampled models, its role as a potential contributor to nocturia remains of interest and warrants further investigation in larger, well-defined cohorts. ClinicalTrials.gov identifier: NCT05404828.

The Triglyceride-Glucose (TYG)-Index has been increasingly used as a simple surrogate marker of insulin resistance and has been associated with adverse cardiometabolic outcomes in several populations. TYG has gained great attention as a predictive index for mortality, but comparisons with other predictive indexes are unexplored, except for TYG-derived indexes such as TYG-body mass index (TYG-BMI). We conducted a comparative prognostic study in two independent cohorts of adults with obesity: a general obesity cohort (n = 1,359) and a bariatric surgery cohort (n = 854), both with long-term follow-up approaching 14years and with different mortality rates (11.5 vs. 5.5%, respectively). We compared TYG index, TYG-BMI index, blood glucose, age, Charlson Index, metabolic syndrome, glucose tolerance, diabetes mellitus, through Cox proportional hazard models with Harrell'C index, and through ROC analysis. We also evaluated the possible incremental predictive value of the above prognostic indexes when combined with blood glucose, the TYG-index, and TYG-BMI index. Across both cohorts, several metabolic and clinical indices were significantly associated with all-cause mortality in univariable analyses. However, age and Charlson Comorbidity Index consistently showed the strongest discrimination and prognostic performance. The various indexes significantly predicted mortality at Cox proportional hazard models (p always < 0.001). Harrell'C index correlated with ROC area under the curves of each index (p < 0.001), and both Harrell and ROC correlated with quality indexes of Cox analysis (LR, p < 0.001) and with quality indexes of linear regression (F, p < 0.001). Findings were directionally consistent in the bariatric surgery cohort, although lower event rates attenuated overall discrimination. The combined use of more indices together was not uniformly useful to increase the predictive value of the above indices. In obesity, TyG-based indices are associated with long-term mortality risk but add limited prognostic value beyond age and multimorbidity burden. These markers may be considered complementary tools for metabolic characterization rather than primary instruments for mortality risk stratification. This study reinforces the concept that various mortality indexes are as valid as, or even more predictive than, TYG index.

This study aimed to evaluate the diagnostic accuracy of [99mTc]-Tc-HYNIC-PSMA-11 SPECT/CT in differentiating glioblastoma recurrence from post-treatment changes, comparing both visual and quantitative parameters with follow-up MRI as the reference standard. In this cross-sectional study, 31 lesions with suspected glioblastoma recurrence underwent [99mTc]-Tc-HYNIC-PSMA-11 SPECT/CT imaging. PSMA uptake was assessed visually (0: no uptake, 1: mild, 2: moderate, 3: severe) and quantitatively using the tumor-to-background ratio (TBR). Diagnostic performance was evaluated through ROC analysis, sensitivity, specificity, and area under the curve (AUC) comparisons. Among the 30 patients with 31 lesions (70% male, mean age 46 years), initial MRI identified recurrence in 38.71% (n = 12), no recurrence in 35.48% (n = 11), and was indeterminate in 25.81% (n = 8) of lesions. Follow-up MRI and other clinical data confirmed recurrence in 58.06% (n = 18) and no recurrence in 41.94% (n = 13). Quantitative SPECT/CT analysis revealed a mean TBR of 13.08 ± 12.09. Visual PSMA assessment demonstrated a sensitivity of 83% and specificity of 92% when considering moderate/high uptake as positive. TBR quantitative analysis showed excellent discrimination (AUC = 0.87), with an optimal threshold of 7.55 yielding 89% sensitivity and 91% specificity. Initial MRI had 100% sensitivity and 91.67% specificity when excluding indeterminate cases. For the 8 patients with indeterminate initial MRI, both visual and quantitative PSMA parameters showed 100% accuracy in predicting the final diagnosis. [99mTc]-Tc-HYNIC-PSMA-11 SPECT/CT, particularly using quantitative TBR analysis, offers a reliable alternative to MRI for detecting GBM recurrence, especially in cases where MRI findings are equivocal.

Cardiac arrhythmia is a disorder caused by disruptions in the regular heart rhythm. Arrhythmias are categorized into two classes: sinus and non-sinus rhythms. Whereas sinus rhythms are generally low-risk, non-sinus rhythms are associated with higher risks of morbidity and mortality, including stroke and death. This research proposes a novel method, Optimized Hierarchical Fused Fuzzy Deep Reinforcement Learning OHFFDRL, which incorporates three steps for arrhythmia prediction: data preprocessing, reinforcement learning, and fuzzy deep learning. We evaluated the proposed method using a 12-lead electrocardiogram dataset comprising 10,646 patients. Our approach leverages recent advances in machine learning and medical science to achieve an accuracy of 94% in predicting non-sinus rhythms. Furthermore, the area under the ROC curve for OHFFDRL was 0.91, and the empirical ROC area was 0.90. In addition, the interpretability of the model has been analyzed with SHAP, LIME, Calibration Curve, Adversarial vulnerability, and Integrated Gradients. Our experimental results, among other findings, indicate that the most important feature for distinguishing heart rhythms is TAxis (the movement range in ventricular repolarization). These results demonstrate the potential of machine learning for the early prevention of heart disease through non-sinus rhythm prediction. The source code is available at (https://github.com/arman-daliri/OHFFDRL).

Background: The aortic elastic properties are relevant at several sites of cardiovascular. Increased arterial stiffness is an independent risk factor and predictor of cardiovascular mortality and an early predictor of coronary risk useful in screening. Therefore the evaluation of arterial stiffness may be important for clinical diagnosis and intervention in cardiovascular disease. Aim: The purpose of this study was to show the effect of age and total serum cholesterol on the percent wall thickness change, after adjusting for stroke volume. Methodology: A cross-sectional study using two-dimensional and Doppler of blood flow of aortic root echocardiography was preformed for a random sample of 60 apparently healthy males with an age ranging between 17 and 75 year. Aortic wall elasticity was assessed by percent wall thickness change. Results: Using multivariate modelling it was shown that age, serum cholesterol and stroke volume were a statistically significant independent predictors of Aortic wall elasticity (assessed by percent wall thickness change), after adjusting for blood pressure and BMI. Age and serum total cholesterol had an average ROC area of round 0.7 when used to predict high rigidity of Aortic artery. An age of 68 years and above was 95% specific in detecting rigid artic artery with low sensitivity 30%. A serum total cholesterol of 199 mg/di and above was associated with sensitivity of 50% and specificity of 85% in detecting rigid aortic artery. Conclusion: Being in the sixth decade of life and having a serum cholesterol of &gt;200 mg/dl will predict a rigid aorta with a reasonably high specificity. Measurements of aortic artery elasticity using Echocardiography is simple and may contribute to cardiovascular risk assessment.

Background: Human cysticercosis, caused by the larval stage (cysticerci) of the pork tapeworm Taenia solium, is an important zoonotic disease. The disease is prevalent in developing countries where porcine cysticercosis is common and undercooked pork is habitually consumed. Objective: This study aimed to develop an immunochromatography-based test kit for the rapid diagnosis of human cysticercosis using low-molecular-weight antigens purified from cyst fluid of the T. solium Asian genotype to detect specific IgG antibodies in whole blood. The kit was designated as "the cysticercosis whole-blood test kit (iCys WB kit)." Methods: It was evaluated under laboratory conditions using 164 whole-blood samples, of which 21 were from confirmed cysticercosis cases. The results of the iCys WB kit, which detects anti-T. solium (cysticercus) antibodies in simulated whole blood samples, were compared with results from corresponding human serum samples. Results: When using both sample types, iCys WB kit demonstrated an accuracy of 98.8%, a sensitivity of 91.7%, a specificity of 100%, a positive likelihood ratio of 0, a negative likelihood ratio of 0.083, and an ROC area of 0.96. The agreement between results obtained from simulated whole-blood and serum samples showed perfect concordance. Conclusions: The iCys WB kit is a valuable easy-to-handle diagnostic tool and may be applicable for supporting clinical diagnosis at the point of care.

Abstract Background/Aims 14-3-3η is a joint-derived soluble biomarker linked to RA pathogenesis and joint damage. Prior studies have demonstrated its diagnostic utility. This study confirms and expands on earlier findings by evaluating the specificity of 14-3-3η in a clinically diverse RA and disease control cohort. Methods Serum 14-3-3η was measured in a total of 615 subjects (50 RA and 545 controls) using the Augurex ELISA. Group differences were assessed using Mann-Whitney and Kruskal-Wallis. ROC area under the curve (AUC) analysis and Fisher’s Exact test evaluated the diagnostic performance of 14-3-3η. 14-3-3η positivity was set at &amp;gt;0.19 ng/ml with statistical significance being p &amp;lt; 0.05. Results Median (IQR) serum 14-3-3η levels were significantly higher in established RA [0.94 ng/ml (0.20-14.15)] than all controls [0.02 (0.00-0.06)], p &amp;lt; 0.0001 (Table 1). 14-3-3η’s differential expression yielded a significant AUC of 0.93, p &amp;lt; 0.0001, with 76% sensitivity and 93% specificity. 14-3-3η positivity was significantly associated with RA diagnosis (OR = 42.3, 95% CI: 20.4-87.5, p &amp;lt; 0.0001). Conclusion This study reinforces the diagnostic value of serum 14-3-3η, demonstrating significantly elevated levels in RA patients compared to a broad spectrum of disease controls. With a high specificity of 93% and a strong association with RA diagnosis (OR = 42.3), 14-3-3η emerges as a robust biomarker that can enhance diagnostic confidence in clinical settings. Disclosure A. Marotta: Shareholder/stock ownership; Augurex Life Sciences Corp. W.P. Maksymowych: Consultancies; Augurex Life Sciences Corp, Abbvie, Eli-Lilly, Novartis, Pfizer, UCB, BMS, Celgene, Galapagos. R. Sengupta: Consultancies; Biogen, Abbvie, Pfizer, BMS, Novartis, UCB. Honoraria; Augurex Life Sciences Corp. S. Bleakley: Corporate appointments; Augurex Life Sciences Corp. S. Wichuk: None. N. Biln: Shareholder/stock ownership; Augurex Life Sciences Corp.

Background: Elevated levels of the C-3 epimer (3-epi-25(OH)D) of 25-hydroxyvitamin D (25(OH)D) have been identified in premature infants as compared to most adults, and an immature liver has been suggested as a possible cause. We hypothesised that patients with cirrhosis might present with elevated C-3 epimerisation due to impaired liver function. The aim was to assess whether 3-epi-25(OH)D levels differ in patients with chronic liver disease with cirrhosis vs. those without cirrhosis. Methods: A total of 309 patients were included (254 patients with cirrhosis vs. 55 without cirrhosis). Serum 25(OH)D and 3-epi-25(OH)D levels were determined using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Results: Patients with cirrhosis had significantly higher median relative 3-epi-25(OH)D concentrations, as compared to patients without cirrhosis (7.4% (5.5-10.4) vs. 4.8% (2.4-5.7), respectively; p < 0.001). They also had similar absolute 3-epi-25(OH)D levels (despite having lower 25(OH)D serum concentrations) than patients without cirrhosis. A progressive increase in relative 3-epi-25(OH)D levels was observed with more advanced cirrhosis (p < 0.001). An analysis of the ROC area under the curve determined 6% as the optimal cut-off for relative 3-epi-25(OH)D. All patients with Child-Pugh stage C and 88.6% with stage B were above the 6% cut-off and had significantly higher absolute serum 3-epi-25(OH)D concentrations (0.9 ng/mL vs. 0.6 ng/mL; p < 0.05) and lower serum 25(OH)D levels (9.3 vs. 14.1 ng/mL; p < 0.001) than patients <6% cut-off. Conclusions: These results reflect the marked increases in relative 3-epi-25(OH)D levels that occur with cirrhosis. The specific hepatic metabolic alterations still need to be unravelled, including whether cirrhosis might lead to reduced epimer clearance.

OBJECTIVE: To evaluate the diagnostic value of Revolution APEX energy spectrum CT imaging and quantitative parameters in T staging of pancreatic cancer (PCa). METHODS: Energy spectrum CT of 105 patients diagnosed with PCa confirmed by pathologic biopsy and 72 controls from June 2022 to January 2025 were retrospectively obtained. Patients were pathologically confirmed as PCa (15 patients with T1 stage, 23 patients with T2 stage, 46 patients with T3 stage, and 21 patients with T4 stage). Energy spectrum CT captured 40keV, 70keV, and 120 kVp-like images (Equivalent to conventional CT images at approximately 120 kVp), which were analyzed for contrast-to-noise ratio (CNR), noise, signal-to-noise ratio (SNR), and subjective scores. Normalized iodine concentration (NICtumor), energy spectrum slope (k), attenuation coefficient, and normalized CT value were quantitatively assessed, and the diagnostic efficacy was analyzed by ROC curve and area under the curve (AUC). RESULTS: The control group and each T-staging subgroup did not differ significantly in age, gender, and BMI, although tumor diameter significantly increased with advancing T-staging (P < 0.001). The 70keV images outperformed the control group in noise control, SNR, and subjective scores, despite having a significantly lower CNR in the tumor group (P < 0.001). Energy spectrum parameters changed significantly with T staging, including NICtumor (T1: 38.7% to T4: 64.0%), k value (T1: 1.3 to T4: 2.6 HU/keV), and attenuation coefficient (T1: 0.9 to T4: 1.6 HU/keV) (all P < 0.001). The AUC for NICtumor and k combined was 93.40% (sensitivity: 94.03%, specificity: 81.50%) in differentiating early from late stages, significantly surpassing the performance of a single parameter (P < 0.001). CONCLUSION: This preliminary study suggests that the 70keV images from Revolution APEX spectral CT, combined with a multiparametric model based on standardized iodine concentration and spectral slope, demonstrate potential value in distinguishing pancreatic cancer T staging. This approach may provide additional quantitative radiographic reference information for preoperative assessment.

<bold>Background</bold>Cancer heterogeneity results in patients with the same diagnosis responding differently to drugs, making treatments extremely challenging. Advances in computational power enable personalized treatments that suppress tumors and extend patient survival. Therefore, accurate prediction of cancer cell response to a particular medication is of utmost importance. Current deep learning-based models have achieved impressive accuracy, but they often function as a “black box” and cannot explain the reason for the prediction. To address this limitation, we develop a deep learning-based model, BKDRP, which incorporates prior biological information into the architecture, along with molecular fingerprints of drugs, while embedding biological priors into its architecture. Specifically, it incorporates the fact that genes encode proteins that combine to form protein complexes, which in turn regulate biological pathways, ultimately targeted by drugs.<bold>Results</bold>We evaluate BKDRP on the GDSC (Genomics of Drug Sensitivity and Cancer) cell line dataset using multi-omics gene expression, protein expression, mutation, and copy number variation. Four rigorous experiments have been conducted to test the model’s generalizability: prediction of unknown drug–cell line responses, responses to unseen drugs (LODO: Leave-One-Drug-Out), responses to unseen cell lines (LOCLO: Leave-On-Cell-Line-Out), and responses across unseen cancer types (LOCO: Leave-One-Cancer-Out). The performance of the proposed method and baseline algorithms is assessed using two metrics: Area Under the ROC Curve (AUC) and Area Under the Precision-Recall Curve (AUPR). The experimental results demonstrate that BKDRP performs well in different evaluation techniques. Notably, BKDRP has achieved an AUC of 0.8845, surpassing traditional machine learning and deep learning approaches and demonstrating robustness in handling biological variability across cancer types. A case study of lung adenocarcinoma (LUAD) highlights known biomarkers (KRAS, EGFR, STK11), key proteins (SOCS1, HSPA8, SMC3), and drugs (Erlotinib, Palbociclib) that are consistent with the literature.<bold>Conclusions</bold>In conclusion, BKDRP presents a novel biological knowledge-driven deep neural network model for cancer drug response prediction that shows strong predictive accuracy and interpretability. By integrating multi-omics data and incorporating domain knowledge, BKDRP has the strong potential for applications in biomarker discovery and the advancement of personalized oncology.

Severe fear of childbirth (FOC) during pregnancy holds significant implications for maternal mental health, decisions about mode of birth, and potentially infant development. Until recently, there were no tools available to measure FOC within a UK population that assessed the full construct using acceptable phraseology. A new tool, the Fear of Childbirth Questionnaire, has been developed but requires full psychometric evaluation. This study aimed to assess the validity and reliability of the Fear of Childbirth Questionnaire and examine possible threshold scores indicating clinical severity. Pregnant women (N = 540) completed the Fear of Childbirth Questionnaire online alongside additional measures for current/previous obstetric history, anxiety, and depression to test dimensions of validity (ISRCTN62032021). Most women (N = 360) then completed the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders 5th Edition (Research Version) Module F (phobia) with trained interviewers over the telephone. Sensitivity and specificity were calculated. A subsample (N = 61) repeated Fear of Childbirth Questionnaire completion 2 weeks later to inform test-retest reliability. Internal consistency, validity (convergent, discriminant, and criterion-related), unidimensionality, and factor structure were examined. Internal consistency was excellent (α = 0.89; ω = 0.89) and test-retest reliability was good (r = 0.87). Strong association with the Fear of Birth Scale (r = 0.69) indicated convergent validity. Discriminant validity was indicated via moderate correlations with general measures for anxiety (r = 0.40-0.53), and less so with those of depression (r = 0.35). Scores were higher for those with current/previous mental health difficulties, previous birth trauma, those preferring an epidural, and those preferring a planned cesarean section, indicative of criterion-related validity. An optimal cut-off value of ≥33 total score and ≥4 current impact score is recommended to initiate further exploration of support/intervention needs (ROC Area under the curve = 0.80, 95% CI 0.75:0.85; sensitivity 71.9, specificity 85.4, positive predictive value 39.0, negative predictive value 95.9). The Fear of Childbirth Questionnaire is psychometrically robust and can identify FOC at levels commensurate with a clinical phobia. It is the only tool purposefully developed and clinically validated to identify FOC in a UK population.

The emergence of generative AI images has profoundly upended the art world. Distinguishing AI-generated images from human-created artwork is increasingly challenging. If this issue remains unresolved, dishonest individuals may exploit those who are willing to pay more for original artwork and businesses whose stated policies exclude AI graphics. There are a number of methods for differentiating AI photos from human art: Diffusion model-focused research tools, classifiers developed through supervised learning, and identification by experienced artists utilizing their understanding of creative methods. The objective of this study is to evaluate the performance of various machine learning algorithms when applied to datasets enhanced through two image enhancement techniques: Gabor Filter (GF) and Colour Layout Filter (GF). The focus is on comparing the effectiveness of these filters in improving the accuracy, precision, recall, ROC, and PRC of selected algorithms, thereby determining which technique yields superior results for different machine learning models. The problem statement addresses the challenge of optimizing machine learning model performance on image datasets. Specifically, it investigates whether the Gabor Filter, known for its effectiveness in feature extraction from images, can outperform the CLF Filter in enhancing the predictive capabilities of algorithms such as Bayes Net, Sequential Minimal Optimization (SMO), Instance-Based K-nearest neighbor (IBK), Bagging, Jrip, and Random Forest. The filters assist in retrieving more pertinent features in the dataset that subsequently increases both the model robustness and the accuracy of classification. Several performance measures were relied upon to evaluate the models and determine their comparative performance: The accuracy, the precision, the recall, the ROC area (Receiver Operating Characteristic), and the PRC region (Precision-Recall Curve). The best model was Random Forest with Gabor Filter (RF-GF) that got the highest accuracy (92.43), precision score (0.94), recall score (0.93), ROC (0.96), as well as PRC (0.99). The tests that provide statistically significant responses were Pairwise Wilcoxon signed-rank tests indicating that RF-GF performed significantly better when compared to other models, including SMO with Gabor Filter (SMO-GF) and CLF-based models. The proposed model was also compared to state-of-the-art methods and external benchmarks by the study proving the competitiveness of the model in regard to performance and efficiency of calculations. Moreover, a combination of the application in the AI detecting system like image classification, fraud detection, and medical diagnosis was considered. The findings show that the RF-GF model is resilient, effective and performant that can be used in real-life applications whereby there are constraints to the available computational resources.

To analyze risk factors of enteral nutrition (EN) diarrhea in intensive care unit (ICU) patients and to construct nomogram model. A retrospective analysis was conducted on 402 patients who received EN treatment in the ICU of Shanghai Blue Cross Brain Hospital from January 2022 to January 2025. They were divided into diarrhea group and non-diarrhea group based on the occurrence of EN diarrhea. We used univariate and multivariate logistic regression analysis to identify risk factors for EN diarrhea. Construct a nomogram model for the occurrence of EN diarrhea in ICU patients based on independent risk factors and conduct goodness of fit tests. The incidence of EN diarrhea in ICU patients was 29.85% (120/402). Hypoproteinemia, abdominal surgery in the last five days, fasting time > 5 days, antibiotic use > 2 weeks, no gradual increase in EN preparations, oral potassium preparations, EN liquid infusion rate ≥ 100ml/h, albumin level ≥ 35g/L were identified as risk factors for EN diarrhea (p<0.05). Based on the above influencing factors, a nomogram model was constructed and internally validated using the Bootstrap method. The nomogram model basically fitted with the ideal model. ROC area under curve (AUC) was 0.860 (95% CI: 0.820-0.901), indicating a certain predictive value for EN diarrhea. The nomogram model has high predictive power in predicting the risk of EN diarrhea in ICU patients. It is beneficial for screening high-risk patients and further adjusting nutritional therapy.

Machine learning has been used for decades to analyze vast datasets, classify and cluster data, and make predictions using algorithms. One of its top use areas is cybersecurity, where it can help detect and prevent destructive threats such as malware. The use of machine learning in cybersecurity has proven to be a powerful tool in detecting and predicting malware attacks. In recent years, the number of Internet users has greatly increased and with it the number of malware attacks. This has made predicting malware a challenge. Consequently, to date, there is still a need to examine the numerous existing MLs’ performance. This study is presented to identify the best classification model for predicting malware using two datasets and 18 different classifiers belonging to six learning strategies. The results showed that the RandomForest classifier had the highest accuracy, precision, recall, F1-measure, and ROC Area metrics, Moreover, Trees and Bayes learning strategies showed the best predictive performance on the two datasets compared with the other five learning strategies.

Diagnostic accuracy of non-invasive fatty liver indexes and their association with ECORE-BF scale in a cohort of 386,924 Spanish workers.

Validation of the revised Baux score to predict mortality and formulation of a new model for burns mortality in Southeast Asian patients

Cardiometabolic multimorbidity (CMM) has become an increasing global public health challenge. In China, the prevalence of CMM is rising rapidly among middle-aged and older adults, with estimates ranging from 11.6% to 16.9%, posing a substantial burden on both individuals and healthcare systems. However, effective tools for predicting individual risk of CMM remain limited, hindering timely prevention and intervention. This study used data from the China Health and Retirement Longitudinal Study (CHARLS) between 2011 and 2015, including 7,913 participants aged ≥ 45 years without CMM at baseline. Incident CMM events were identified during the 2015 follow-up based on self-reported diagnoses of cardiometabolic diseases. Ten lipid metabolism biomarkers and derived composite indices (TC, TG, LDL-C, HDL-C, TyG, TyG-BMI, LAP, CTI, non-HDL-C, and RC) were evaluated. Predictive models were estimated using logistic regression, random forest, gradient boosting machine, eXtreme Gradient Boosting (XGBoost), support vector machine, naïve Bayes, deep learning (DL), and an ensemble model. The dataset was randomly split into training (75%) and validation (25%) subsets. Model discrimination was assessed using ROC curves and Area Under the Curve (AUC); calibration was evaluated with calibration plots and Brier scores; classification performance was examined using confusion matrices. Decision curve analysis (DCA) and clinical impact curves (CIC) were applied to assess clinical utility across risk thresholds. Feature importance ranking and SHapley Additive exPlanations (SHAP) were used to quantify variable contributions, marginal effects, and feature interactions. In addition, regional variations in CMM incidence were illustrated using choropleth maps, and correlations between lipid markers and CMM prevalence were analyzed with Pearson coefficients and heatmaps. Over the four-year follow-up, 1,355 participants (17.1%) developed CMM. Compared with controls, incident cases were older, had a higher proportion of women and urban residents, and showed higher BMI. They also had significantly elevated triglycerides (126.6 vs. 101.8 mg/dL), reduced HDL-C (45.2 vs. 50.3 mg/dL, P < 0.001), and increased TyG-BMI and LAP (P < 0.001). Geographical analysis revealed markedly higher CMM incidence in northern cold regions (> 40%) than in southern regions (< 20%). The ensemble model achieved robust predictive performance (AUC = 0.715), followed closely by the DL model (AUC = 0.716) and GBM (AUC = 0.714). These non-linear models consistently outperformed GLM (AUC = 0.696), SVM (AUC = 0.696), and XGBoost (AUC = 0.683). Ensemble, DL, and RF models also demonstrated the best calibration (lowest Brier score, 0.125) and provided the greatest net benefit across risk thresholds. SHAP analysis indicated that composite indices, particularly TyG-BMI, LAP, and TyG, contributed most to risk prediction, whereas HDL-C exerted a protective effect. In contrast, traditional single lipid markers such as LDL-C and TC ranked lower in predictive importance. This study demonstrates that machine learning models incorporating lipid metabolism biomarkers and derived indices can predict the risk of CMM. Composite indicators such as TyG and LAP, which capture insulin resistance and visceral adiposity, showed superior predictive value. DL and ensemble models provided higher discrimination and clinical utility compared with traditional approaches. These models may enable early identification of high-risk individuals, underscoring the importance of lipid and metabolic management in CMM prevention, with potential implications for clinical decision-making and public health strategies.

ObjectiveThis study aimed to characterize the epidemiology of pertussis in children, evaluate the diagnostic performance of clinical and laboratory features, assess the effectiveness of different suspected-case criteria, and identify independent risk factors for ICU admission.MethodsWe retrospectively analyzed demographic, clinical, and laboratory data from patients aged ≤14 years who underwent pertussis testing. Participants were stratified by test results, and the sensitivity and specificity of individual symptoms and laboratory parameters were calculated. Diagnostic performance was further assessed using univariate and multivariate logistic regression and nomogram analyses.ResultsAmong 2,015 children tested, 724 (35.9%) were positive for pertussis, with 187 (30.2%) hospitalized and 55 (8.9%) admitted to the ICU. Cases increased from August 2023, peaking in May 2024, with the majority aged 3–11 years (51.8%). Compared with B. pertussis–negative children, positive cases exhibited higher WBC counts and procalcitonin (PCT) levels. Incorporating PCT into the WHO suspected-case definition improved the ROC area from 0.710 to 0.870, with an optimal cutoff of 0.1665 ng/mL. Cyanosis, post-tussive vomiting, neutrophil count, and assisted ventilation emerged as independent predictors of ICU admission, with a neutrophil cutoff of 6.62 × 109/L.ConclusionThe age distribution of pediatric pertussis shifted from young infants to preschool- and school-aged children. Clinical features alone are insufficient for reliable diagnosis due to overlap with non-pertussis cases. Elevated neutrophil count serves as an independent predictor of ICU admission, highlighting its potential utility for early risk stratification in pediatric pertussis.

Background:The onset of chronic kidney disease (CKD) results in a high level of cardiovascular morbidity and mortality. Conventional risk assessment systems tend to fail in considering the cardiovascular risk of this group of people. A biomarker of myocardial damage, high-sensitivity cardiac troponin I (hs-cTnI), has become a candidate biomarker of subclinical cardiac damage.Materials and Methods:A prospective observational study was carried out of 120 patients with CKD stages 3 to 5 that excluded dialysis patients. A high-sensitivity immunoassay was used to measure the baseline value of hs-cTnI. The follow-up was thorough, with patients being kept under observation after 12 months of occurrence in their patients major adverse cardiovascular events (MACE), i.e., myocardial infarction, heart failure hospitalization, and cardiovascular death.Results:Out of 120 CKD patients (mean age 58.2+/- 9.6 years; 62 males), 48 patients (40%) had elevated hs-cTnI (>19.0 ng/L). The number of patients who developed MACE at the time of the follow-up was 29 (24.2%). There was a significantly greater proportion of MACE in an elevated hs-cTnI group (52.1%) than in a normal hs-cTnI group (10.4%) (P < 0.001). The Cox regression analysis indicated that an independent outcome predictor of MACE was hs-cTnI (HR 3.89, 95% CI 1.91793, P = 0.002). The ROC area under the curve of hs-cTnI at predicting MACE was 0.81 (95% CI 0.7290).Conclusion:High-sensitivity T-I is a good prognostic biomarker of cardiovascular events among CKD patients. Its regular evaluation can help in the initial classification of risks and selective cardiovascular preventive interventions.