Dong Wu

Individuals with chronic obstructive pulmonary disease (COPD) experience a significant decline in their quality of life owing to cardiovascular disease (CVD). This study aimed to develop a predictive framework for evaluating CVD risk in patients with COPD. Data from 1070 COPD patients participating in the 2015 China Health and Retirement Longitudinal Study (CHARLS) were analyzed. To ensure robust feature selection, Least Absolute Shrinkage and Selection Operator (LASSO) regression and the Boruta algorithm were utilized. Subsequently, the predictive performance of six distinct Machine learning (ML) models (Logistic Regression, Random Forest, Support Vector Machine (SVM), Gradient Boosting Machine, XGBoost, and Multi-Layer Perceptron) was comprehensively compared. The Synthetic Minority Oversampling Technique-Nominal Continuous (SMOTE-NC) was applied to the training set to combat class imbalance. An interpretable risk assessment tool was developed using SHapley Additive exPlanations (SHAP). 305 participants (28.50%) had CVD. Seven variables were used to build the six models. The SVM model showed comparatively better performance than the others, with a training Area Under the Receiver Operating Characteristic curve (AUROC) of 0.819 (95% Confidence Interval (CI) 0.793–0.844), accuracy of 74.42%, sensitivity of 75.56%, precision of 74.18%, specificity of 73.26%, and F1 score of 74.86%. In the test set, the AUROC was 0.719 (95% CI, 0.670–0.760), with an accuracy of 68.63%, sensitivity of 64.20%, precision of 66.53%, specificity of 64.96%, and F1 score of 69.36%. This study identified seven key predictors—sex, body weight, hypertension, dyslipidemia, disability, self-rated health, and vision status—that are significantly associated with cardiovascular risk in Chinese patients with COPD. Among the six machine-learning algorithms evaluated, the SVM model demonstrated the most robust performance; however, its predictive capacity remains moderate, reflecting the inherent limitations of cross-sectional survey data and the reliance on self-reported diagnoses. Future prospective studies and rigorous external validation in independent cohorts are essential to refine these predictors and translate this machine-learning approach into reliable clinical decision-support systems for the personalized management of COPD patients.

Introduction: Brain metastases (BM) represent a common and fatal progression in small cell lung cancer (SCLC), yet prognostic tools for this population remain underdeveloped. This study aimed to establish and externally validate a machine learning–based model to predict overall survival (OS) in SCLC patients with BM. Methods: We extracted clinical data from 2392 SCLC patients with BM from the SEER database to construct prognostic models using Cox regression, AJCC staging, and four machine learning algorithms: Random Survival Forest (RSF), Extreme Gradient Boosting (XGB), Elastic Net (Enet), and Artificial Neural Network (ANN). Key features were selected via Lasso-Cox regression. Model performance was evaluated using time-dependent AUC, calibration curves, Brier scores, precision-recall (PR) curves, and decision curve analysis (DCA). SHAP and partial dependence plots were applied for model interpretability. External validation was conducted using an independent hospital-based cohort of 85 patients, with comparability to the SEER cohort addressed through inverse probability of treatment weighting (IPTW). Results: Among all models, the RSF algorithm demonstrated the best overall performance. In the training cohort, it achieved AUCs of 0.738 and 0.809 for 1-year and 2-year OS, respectively. In the internal validation cohort, AUCs were 0.718 and 0.748, and in the external validation cohort, 0.686 and 0.802, respectively. The RSF model also showed favorable calibration and the lowest Brier scores across datasets. SHAP analysis ranked chemotherapy, liver metastasis, N stage, and age as the most influential prognostic features. A web-based calculator was developed to enable real-time individualized risk prediction. Conclusions: This study presents a robust, interpretable, and externally validated RSF-based model for predicting OS in SCLC patients with BM. The model offers clinically relevant insights and is accessible via an online tool, supporting its potential integration into personalized treatment planning.

Objective: This study aims to evaluate the predictive potential of D-dimer levels and the systemic immune-inflammatory index (SII) for identifying concurrent pulmonary thromboembolism (PTE) in patients with acute exacerbations of chronic obstructive pulmonary disease (AECOPD).Methods: We conducted a case–control study involving 75 patients with AECOPD and concurrent PTE, admitted to the Affiliated Hospital of Guangdong Medical University between June 2017 and December 2020. A control group comprising 76 AECOPD patients without PTE was included for comparison. Clinical characteristics and laboratory findings were compared between the two groups. Multivariate logistic regression was employed to identify independent risk factors for PTE in AECOPD patients. The predictive accuracy of these risk factors was assessed using receiver operating characteristic (ROC) curves, while Spearman correlation analysis evaluated associations between variables. Model robustness was validated internally using bootstrap techniques, and decision curve analysis (DCA) was applied to determine the clinical utility of the predictive model.Results: Multivariate logistic regression identified D-dimer [odds ratio (OR) 1.17, 95% confidence interval (CI) 1.03–1.38] and the SII (OR 1.003, 95% CI 1.000–1.006) as independent risk factors for PTE in patients with AECOPD. The area under the ROC curve (AUC) values for predicting PTE in AECOPD patients were 0.758 (95% CI 0.682–0.834) for D-dimer, 0.757 (95% CI 0.677–0.838) for SII, and 0.834 (95% CI 0.768–0.900) for their combination, respectively. The Bootstrap results demonstrated that the model combining D-dimer and SII had an AUC of 0.8409 (95% CI: 0.7649, 0.9156), indicating robust performance across resampled datasets. Both D-dimer and SII showed a positive correlation with hospital stay duration (r = 0.289, p = 0.047; r = 0.235, p = 0.043). DCA demonstrated that the combination of D-dimer and SII provides significant net benefit in clinical decision-making for predicting PTE in AECOPD patients.Conclusion: D-dimer and SII are independent risk factors for predicting concurrent PTE in AECOPD patients. The combination of these two markers demonstrates robust predictive accuracy, and their levels may be positively correlated with disease severity. These findings suggest that D-dimer and SII could play a valuable role in clinical decision-making and risk stratification in AECOPD patients.

RNA-binding proteins (RBPs) regulate the generation of circular RNAs (circRNAs) by participating in the reverse splicing of circRNA and thereby influencing circRNA function in cells and diseases, including cancer. Increasing evidence has demonstrated that the circRNA-RBP network plays a complex and multifaceted role in tumor progression. Thus, a better understanding of this network may provide new insights for the discovery of cancer drugs. In this review, we discuss the characteristics of RBPs and circRNAs and how the circRNA-RBP network regulates tumor cell phenotypes such as proliferation, metastasis, apoptosis, metabolism, immunity, drug resistance, and the tumor environment. Moreover, we investigate the factors that influence circRNA-RBP interactions and the regulation of downstream pathways related to tumor development, such as the tumor microenvironment and N6-methyladenosine modification. Furthermore, we discuss new ideas for targeting circRNA-RBP interactions using various RNA technologies.

Background: Asthma is a heterogeneous group of diseases. The mechanism by which dysregulated circRNAs affect severe asthma by regulating macrophage polarization remains unclear.Methods: High-throughput RNA sequencing technology was used to analyze circRNA expression in peripheral blood mononuclear cells (PBMCs) from patients with severe asthma. RT-qPCR and ELISA were used to analyze the expression of inflammatory factors in a mouse model of severe asthma induced by ovalbumin-lipopolysaccharide. The effect of circ-0001875 on macrophage activation and the underlying mechanism were analyzed by RT-qPCR, Western blot, and ELISA. Subsequently, the regulatory relationships among circ-0001875, miR-31-5p, and SP1 were examined through dual luciferase reporter gene assay, and the mechanism by which they regulate macrophage polarization was analyzed by Western blot.Results: Compared with the healthy control group, 420 circRNAs were differentially expressed in PBMCs from patients with severe asthma. Among them, circ-0001875, which was mainly expressed in the cytoplasm of monocytes, was significantly downregulated in asthmatics, especially those with severe disease. circ-0001875 overexpression inhibited M1 macrophage activation in vitro and alleviated lung inflammation in a mouse model of severe asthma. Mechanistically, circ-0001875 promoted SP1 translation by competitively binding to miR-31-5p, thereby reducing its inhibitory effect on SP1 translation; SP1 then inhibited M1 macrophage polarization, which is associated with severe asthma, through the NF-κB signaling pathway.Conclusions: We found that circ-0001875 plays an important role in regulating M1 macrophage polarization, which is associated with a severe pro-inflammatory response.

The phylogeny of the vertebrate pathogen Histoplasma capsulatum and its varieties was analyzed on the basis of GenBank data, comparing preceding papers that distinguished lineages on the basis of a much smaller dataset, partly dating back two decades. The aim was to establish the predictive value of individual research papers on biodiversity, which eventually may lead to altered nomenclature with large clinical consequences. A total of 1985 sequences of ITS, ARF, OLE and H-anti were downloaded. ITS showed insufficient resolution, and the sequences of the H-anti gene were too short to provide reliable conclusions. Ten major lineages from the seven reports were selected for comparison. Compared to the currently available global data, several earlier studies applied somewhat skewed datasets, biased towards the Americas. Possible separation of Indian and Indonesian lineages were consequently overlooked. Previously distinguished lineages were again recognized, but because of low bootstrap values and extensive genetic exchange, several of these do not deserve species status. No recombination was observed with North American H. mississippiense and H. ohiense. An African clade (var. duboisii) was individualized. Despite its position in close association with South American clades, histopathology and clinical course of this entity underlines that it has other evolutionary drivers. This might also hold true for the North African donkey disease caused by var. farciminosum, although strains analyzed thus far are indistinguishable from South American strains. On the basis of phylogenetic data, Indian and Indonesian clades are separate, but more clinical data are needed to establish their value as individual species.

Background: Although studies have explored the factors influencing the occurrence of disability, predictive models for disability risk in the chronic respiratory diseases (CRD) patient population remain inadequate.Methods: This study employed baseline data from the 2015 China Health and Retirement Longitudinal Study (CHARLS) to select 803 CRD patients without disabilities, who were then followed for 3 years to observe the emergence of new disabilities. Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis was applied to identify risk factors associated with the onset of disability. Ultimately, multivariable logistic regression analysis pinpointed four critical predictive factors: marital status, self-perceived health, depressive symptoms, and age, which were subsequently incorporated into a nomogram model. The model’s predictive efficacy was evaluated using the receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA).Results: During the 3-year follow-up, 196 patients developed new disabilities, yielding an incidence rate of 24.41%. The model evaluation results revealed that area under the curve (AUC) for the training set was 0.724 (95% confidence interval [CI]: 0.676-0.771), and the AUC for the test set was 0.720 (95% CI: 0.641-0.799), demonstrating high accuracy, sensitivity, and specificity. The calibration curve confirmed that the predicted results aligned closely with the actual outcomes, while the DCA analysis illustrated that the model provided substantial net benefits in clinical decision-making, effectively identifying high-risk patients.Conclusion: The nomogram model developed in this study effectively predicts the risk of new disability occurrence in CRD patients within 3 years. By identifying high-risk patients at an early stage, this model provides scientific evidence for early intervention and health management in CRD patients.