Full
Outline
To investigate the risk factors and causal relationships of coronary heart disease (CHD) by utilizing clinical data and genetic variation data, and construct an interpretable risk prediction model for the disease.
A retrospective cohort analysis was conducted on the clinical data and genetic variation data collected from the Kaggle platform and the IEU OpenGWAS Project database, respectively. SMOTEENN mixed sampling was used to balance the data. Logistic regression models and two-sample Mendelian randomization (MR) were combined to screen risk factors and analyze their causal effects. Then risk prediction models were further constructed using random forest (RF), eXtreme gradient boosting (XGBoost), and artificial neural networks (ANN), and feature importance was assessed with the SHAP algorithm.
Logistic regression analysis showed that gender, age, education level, smoking, hypertension, total cholesterol (TC), systolic blood pressure (SBP), and blood glucose were significantly associated with CHD (<0.05), while body mass index (BMI), diastolic blood pressure (DBP), history of stroke, and diabetes had no such association. Two-sample MR analysis indicated that BMI (=1.014, 95%: 1.011 to 1.017, =1.720×10-22), age (=1.015, 95%: 1.012 to 1.019, =3.455×10-18), number of cigarettes smoked per smoker per day (=1.007, 95%: 1.002 to 1.011, =0.005), hypertension (OR=1.057, 95%: 1.046 to 1.067, =5.870×10-28) were positively associated with CHD; Education level (=0.970, 95%: 0.960 to 0.979, =1.231×10-9) and high-density lipoprotein (=0.990, 95%: 0.986 to 0.993, =0.002) were negatively associated with CHD. In the machine learning models, the SMOTEENN-RF performed best, with an accuracy rate of 0.887 and an AUC value of 0.962. SHAP analysis indicated that age, gender, hypertension, SBP, and cigarettes smoked per day contributed most to the model.
Key risk factors such as age, hypertension, and blood lipid levels are significantly correlated with CHD. A high-performing SMOTEENN-RF prediction model is successfully constructed, providing reliable methodological support and practical tool for CHD risk early warning and precision prevention and treatment.
85
29
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
