Background: Inflammatory response and the coagulation system are pivotal in the pathogenesis of clinical deterioration in chronic obstructive pulmonary disease (COPD), prompting us to hypothesize that the systemic coagulation-inflammation (SCI) index is associated with clinical deterioration in COPD. Methods: A cohort of 957 COPD patients (mean age: 68.4 ± 7.8 years; 74.4% male) from January 2018 to December 2021 was analyzed. Six machine learning models (XGBoost, logistic regression, Random Forest, elastic net (ENT), support vector machine (SVM), and K-nearest neighbors (KNN)) were evaluated using accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (AUC-ROC). Results: Our study encompassed 957 patients, out of which 171 were classified in the clinical deterioration of COPD (cd-COPD) cohort. Significant disparities in age, comorbidities like respiratory failure, C-reactive protein, lymphocyte count, red blood cell distribution width (RDW), SCI, procalcitonin (PCT), and D-dimer were depicted between the cd-COPD and non-cd-COPD groups. Concerning machine learning and model comparison, the SVM model showcased consistent performance and strong generalization capabilities on both the training and testing sets compared to the other five machine learning (ML) models. The SCI index, as the most influential predictor, demonstrated a median of 93.08 in cd-COPD compared to 81.67 in non-cd-COPD patients. Conclusion: The SCI is markedly elevated in cd-COPD patients compared to COPD patients, and SVM demonstrates reliable performance in cd-COPD prediction.

Three-vessel coronary artery disease (TVD) is a severe subtype of coronary heart disease, strongly associated with inflammation and metabolic dysfunction. The C-reactive protein-triglyceride glucose index (CTI), an integrated measure of inflammation and metabolism, may serve as a critical predictor of TVD risk. This study evaluates the predictive utility of CTI for TVD. This single-center retrospective study compared clinical characteristics, laboratory indices, and inflammation- and metabolism-related markers between TVD and non-TVD groups. Feature selection using the Boruta algorithm and LASSO regression identified key predictors for constructing a TVD risk model. SHAP analysis provided model interpretability and evaluated the relative importance of predictive factors. Predefined subgroup analyses by sex and diabetes status were conducted using multivariable logistic regression, with interaction terms tested to assess the consistency of CTI's predictive effect across subgroups. Traditional risk factors, including gender, age, hypertension, diabetes, and smoking, were significantly more prevalent in the TVD group. The CTI index was markedly elevated in this group and emerged as a critical predictor. Boruta and LASSO identified CTI, age, and gender as primary predictors. SHAP analysis confirmed the prominent role of CTI, with age and gender also contributing significantly to risk prediction. CTI demonstrated the strongest positive impact on TVD risk assessment. Subgroup analyses showed CTI was significantly associated with TVD in men and diabetic patients, with no significant interaction by sex or diabetes, indicating a generally stable predictive effect. CTI is a reliable predictor of TVD with immediate value for improving current risk assessment and patient stratification. Beyond its present clinical relevance, CTI also holds potential for future applications in early screening and long-term management.

Coronary heart disease (CHD) and ischemic stroke (IS) share several pathophysiological mechanisms and risk factors, such as hypertension, hyperlipidemia, and diabetes. Investigating novel markers, such as the glucose-to-albumin ratio (GAR), for predicting the risk of IS in CHD patients holds significant clinical value. We retrospectively enrolled 1,885 patients diagnosed with CHD who were treated at our hospital from January 1, 2022, to July 31, 2024. Feature selection was conducted using the Boruta algorithm, and a multilayer perceptron (MLP) model was employed to predict the risk of IS in CHD patients. The performance of the model was evaluated using ROC curves and calibration plots. SHAP values and partial dependence plots (PDP) were used to interpret the model's predictions. The study showed that patients in the IS group were older and had significantly higher rates of hypertension and diabetes compared to those without AIS. Additionally, the AIS group had a higher prevalence of triple-vessel disease and right coronary artery lesions. GAR was significantly elevated in the IS group compared to the non-IS group. Key features identified by the Boruta algorithm included GAR, hyperlipidemia, and a history of hypertension. SHAP analysis indicated that GAR was significantly associated with IS risk, and PDP analysis further confirmed GAR as an independent predictor of IS. GAR is a significant independent predictor of IS risk in CHD patients, with elevated GAR levels being strongly associated with an increased risk of IS.

Objective: To develop and validate an interpretable machine learning–based model for the early identification of the risk of clinical deterioration during hospitalization in newly diagnosed lung cancer patients. Methods: Clinical data of patients with newly diagnosed lung cancer admitted between July 2023 and December 2025 were retrospectively collected in this single-center study. A hybrid feature selection strategy integrating least absolute shrinkage and selection operator (LASSO) regression and the Boruta algorithm was employed to identify key predictors. Logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM) models were constructed. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC-ROC), calibration curves, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) values were further applied to interpret the model's decision-making process. Results: A total of 1,001 newly diagnosed lung cancer patients were included, comprising 121 patients (12.1%) in the deterioration group and 880 patients (87.9%) in the non-deterioration group. Through the hybrid feature selection approach, five core predictive variables were identified, including Activities of Daily Living (ADL) score, lymphocyte count, C-reactive protein (CRP), respiratory rate, and neutrophil count. Among the models, the SVM model achieved the highest observed AUC of 0.822 (95% CI: 0.737–0.899) in the internal validation set, with a sensitivity of 0.750 (95% CI: 0.594–0.893) and a specificity of 0.792 (95% CI: 0.740–0.838). After Platt scaling calibration, the model achieved a Brier score of 0.083 (Hosmer-Lemeshow p = 0.649), with a calibration intercept of 0.141 and a slope of 1.126, indicating satisfactory agreement between predicted and observed risks. DCA suggested a potential net clinical benefit. SHAP analysis, in conjunction with multivariable regression results, consistently demonstrated that ADL score and lymphocyte count were associated with reduced risk of clinical deterioration, whereas CRP and neutrophil count were associated with increased risk. Conclusion: The SVM model developed in this study may help identify patients at risk of clinical deterioration among hospitalized lung cancer patients and may assist healthcare professionals in achieving early risk identification and potential clinical decision support.

Objective: To characterize the link of the C-reactive protein-albumin-lymphocyte index (CALLY index) to preserved ratio impaired spirometry (PRISm), and to determine its incremental contribution to PRISm risk prediction in clinical practice. Methods: A total of 189 participants were included in this study and were classified into the PRISm group ( n = 70) and the non-PRISm group ( n = 119) according to the presence or absence of PRISm. Baseline demographic characteristics, comorbidities, and laboratory parameters were collected. Multivariable Logistic regression models with sequential adjustment for confounding factors were used to assess the independent association between the CALLY index and PRISm. Restricted cubic splines (RCS) and subgroup analyses were performed to evaluate the robustness of this association. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) were used to assess the incremental clinical predictive value of the CALLY index. Results: The CALLY index was significantly lower in the PRISm group than in the non-PRISm group (1.4 ± 2.5 vs. 4.7 ± 8.1, P < 0.001). In the fully adjusted model controlling for 11 confounding factors (adjusted model II), each one-unit increase in the CALLY index as a continuous variable was associated with 16% lower odds of PRISm (OR = 0.84, 95% CI: 0.74–0.95, P = 0.007). After dichotomizing the CALLY index, the high-level group showed 73% lower odds of PRISm than the low-level group (OR = 0.27, 95% CI: 0.13–0.57, P < 0.001). The RCS curve demonstrated a significant inverse trend between the CALLY index and the odds of PRISm. Subgroup analyses indicated that this inverse association remained robust across populations stratified by age, gender, smoking history, and comorbidities. In addition, after the CALLY index was incorporated into a model containing baseline clinical characteristics, the area under the ROC curve (AUC) increased significantly from 0.611 to 0.703 ( P = 0.003), and DCA further confirmed a higher net clinical benefit. Conclusion: A high CALLY index is independently associated with lower odds of PRISm. Integrating the CALLY index into routine clinical assessment can significantly improve the predictive performance for PRISm and provide net clinical benefit.