Objective To analyze the characteristics of clinical indicators in patients with type 2 diabetes,and to establish a simple and effective risk prediction model for type 2 diabetes complicated with coronary heart disease by screening risk prediction indicators with machine learning.Methods A retrospective study was conducted,and 217 patients diagnosed with coronary artery disease combined with type 2 diabetes mellitus who were hospitalized in the Hospital from January 2022 to November 2023 were selected.Additionally,214 patients diagnosed with T2DM during the same period in the outpatient department were selected as the con-trol group.Their routine laboratory test data were recorded.The Least Absolute Shrinkage and Selection Op-erator(Lasso)algorithm was used to select features,and the models were built by using seven machine learn-ing algorithms:Random Forest,Decision Tree,Support Vector Machine,eXtreme Gradient Boosting,Logistic Regression,K-Nearest Neighbor,and Artificial Neural Network.The diagnostic efficacy of different models through receiver operating characteristic curve(ROC),area under curve(AUC),calibration curve,specificity,sensitivity,F1 value,and other indicators were evaluated.Results Twenty key factors,including age,gender,systolic blood pressure,diastolic blood pressure,heart rate,C-reactive protein and blood glucose were selected using Lasso regression.When incorporated into various models,the SVM model exhibited the highest sensitiv-ity(88.37％),negative predictive value(82.14％),and area under curve(0.845).The Random Forest model had the highest accuracy(76.47％),positive predictive value(76.74％),and F1 score(0.77).Meanwhile,the XGBoost algorithm demonstrated relatively good specificity(80.95％).After introducing the SHAP model,it was inferred that blood glucose had a significant positive impact on the occurrence of coronary heart disease in individuals with type 2 diabetes.Conclusion Machine learning can serve as an effective tool for assessing the risk of coronary heart disease in patients with type 2 diabetes.In this study,SVM,Random Forest,and XG-Boost models all demonstrate good predictive performance,indicating promising clinical application prospects.

Objective:To explore the association between occupational noise perception and cardiovascular disease (CVD), depression symptoms, as well as their comorbidity in occupational population and provide evidence for the prevention and control of physical and mental illnesses.Methods:A cross-sectional survey design was adopted, based on baseline data in population in 28 prefectures in Sichuan Province and Guizhou Province, and 33 districts (counties) in Chongqing municipality from Southwest Occupational Population Cohort from China Railway Chengdu Group Co., Ltd. during October to December 2021. A questionnaire survey was conducted to collect information about noise perception, depressive symptoms, and the history of CVD. Latent profile analysis model was used to determine identify noise perception type, and multinomial logistic regression analysis was conducted to explore the relationship between different occupational noise perception types and CVD, depression symptoms and their comorbidity.Results:A total of 30 509 participants were included, the mean age was (36.6±10.5) years, and men accounted for 82.0%. The direct perception of occupational noise, psychological effects and hearing/sleep impact of occupational noise increased the risk for CVD, depressive symptoms, and their comorbidity. By using latent profile analysis, occupational noise perception was classified into four levels: low, medium, high, and very high. As the level of noise perception increased, the association with CVD, depressive symptoms, and their comorbidity increased. In fact, very high level occupational noise perception were found to increase the risk for CVD, depressive symptoms, and their comorbidity by 2.14 (95% CI: 1.73-2.65) times, 8.80 (95% CI: 7.91-9.78) times, and 17.02 (95% CI: 12.78-22.66) times respectively compared with low-level occupational noise perception. Conclusions:Different types of occupational noise perception are associated with CVD and depression symptom, especially in the form of CVD complicated with depression symptom. Furthermore, the intensity of occupational noise in the work environment should be reduced to lower the risk for physical and mental health.

Objective:To understand the influence of unhealthy lifestyle on diabetic dyslipidemia and the key influencing factors in occupational population and provided scientific evidence for the prevention of diabetic dyslipidemia.Methods:Based on baseline data and follow-up data of Southwest Occupational Population Cohort from China Railway Chengdu Group Co., Ltd. during 2021. Diabetic dyslipidemia was defined as diabetes plus one or more forms of dyslipidemia, and unhealthy lifestyle factors included smoking, alcohol consumption, unhealthy dietary patterns, low physical activity, and abnormal BMI. Multivariate logistic regression model was used to analyze the relationship between unhealthy lifestyle scores and diabetic dyslipidemia, network analysis was used to find and explore the key lifestyles influencing glycolipid metabolism.Results:A total of 25 631 subjects were included. People with unhealthy lifestyle score 2 and 3 were 1.93 (95% CI: 1.31-2.86) times and 2.37 (95% CI: 1.60-3.50) times more likely to have diabetes with ≥1 forms of dyslipidemia than those with scores of 0; People with unhealthy lifestyle score 1, 2 and 3 were 1.98 (95% CI: 1.08-3.61) times, 2.87 (95% CI: 1.60-5.14) times and 3.95 (95% CI: 2.22-7.06) times more likely to have diabetes with ≥2 forms of dyslipidemia than those with score 0. Network analysis found that abnormal BMI and HDL-C were the "bridge nodes" that link unhealthy lifestyles with diabetic dyslipidemia. Conclusion:The higher the score of unhealthy lifestyle, the higher the risk for diabetic dyslipidemia, abnormal BMI and HDL-C are key factors influencing the association between unhealthy lifestyle and diabetic dyslipidemia.

Objective:To understand the relationship between unhealthy lifestyle and hyperuricemia, as well as the modification effects of hypertension and dyslipidemia in occupational population and provide a theoretical basis for the prevention of hyperuricemia.Methods:A cross-sectional survey design was adopted, based on baseline data from the Southwest Occupational Population Cohort from China Railway Chengdu Group Co., Ltd., which included the population in 28 prefectures from Sichuan Province and Guizhou Province, and 33 districts (counties) from Chongqing Municipality between October and December 2021. This study collected the information about the demographics characteristics, lifestyles, and prevalence of chronic non-communicable diseases of the study subjects through questionnaire, physical measurement and laboratory biochemical test. The unhealthy lifestyle score was scored based on smoking, alcohol consumption, dietary patterns, physical activity, and low weight or overweight, with higher scores being associated with more unhealthy lifestyles. The multivariate logistic regression model was used to analyze the relationship between unhealthy lifestyle score, smoking, alcohol consumption, other factors and hyperuricemia, and the stratified analysis was used to explore the modification effect of hypertension and other diseases on the relationship between unhealthy lifestyle and hyperuricemia.Results:A total of 11 748 participants were included in this study, the prevalence of hyperuricemia was 34.4%. Multivariate logistic regression model showed that current/previous smoking, current/previous alcohol consumption and BMI abnormality were risk factors for hyperuricemia, and the unhealthy lifestyle score showed a "cumulative" effect on the risk for hyperuricemia, with higher score increasing the risk of hyperuricemia, and the OR increased from 1.64 (95% CI: 1.34-2.00) to 2.89 (95% CI: 2.39-3.50). Stratified analysis showed that unhealthy lifestyles had a greater impact on the risk for hyperuricemia in people with hypertension and dyslipidemia. Conclusions:The coexistence of multiple unhealthy lifestyles might increase the risk of hyperuricemia, and this effect was stronger in participants with hypertension and dyslipidemia. Timely correction of unhealthy lifestyles, and control of hypertension and dyslipidemia might reduce the risk for hyperuricemia.

Objective:To investigate the association between alcohol consumption and hypertension and SBP, DBP and the mediating effects of body mass index (BMI) and lipid level in occupational population, and provide reference for the intervention and prevention of hypertension.Methods:Based on the data of Southwest Occupational Population Cohort from China Railway Chengdu Group Co., Ltd., the information about the demographic characteristics, behavior and lifestyle, blood pressure and lipids level of the participants were collected through questionnaire survey, physical examination and blood biochemical test. Logistic/linear regression was used to analyze the association between alcohol consumption and hypertension, SBP and DBP. The individual and joint mediating effects of BMI, HDL-C, LDL-C, TG, and TC were explored through causal mediating analysis. A network analysis was used to explore the correlation between alcohol consumption, BMI and lipid levels, and hypertension.Results:A total of 22 887 participants were included, in whom 1 825 had newly detected hypertension. Logistic regression analysis found that current/former drinkers had a 33% increase of risk for hypertension compared with never-drinkers ( OR=1.33, 95% CI:1.19-1.48). Similarly, alcohol consumption could increase SBP ( β=1.05, 95% CI:0.69-1.40) and DBP ( β=1.10, 95% CI:0.83-1.38). Overall, BMI and lipid levels could mediate the associations between alcohol consumption and hypertension, SBP and DBP by 21.91%, 28.40% and 22.64%, respectively. BMI and TG were the main mediators, and they were also the two nodes with the highest edge weight and bridge strength centrality in the network of alcohol consumption, BMI, lipid levels and hypertension. Conclusions:Alcohol consumption was associated with increased risk for hypertension, and BMI and TG were important mediators and key nodes in the network. It is suggested that paying attention to the alcohol consumption, BMI and TG might help prevent hypertension in occupational population.