CardioRisk: Predictive Application for Myocardial Infarction Incident Risk Assessment Based on Blood Pressure-Glucose-Lipid Patterns

Background: Myocardial Infarction (MI) is a severe cardiovascular disease, typically characterized by ischemic necrosis of myocardial tissue due to sudden occlusion of coronary arteries. MI has a high incidence worldwide, and there is abundant evidence showing a close correlation between MI and factors such as Blood Pressure, Blood Glucose, and Blood Lipid. There is still insufficient in-depth research on the relationship between the Blood Pressure-Glucose-Lipid Patterns and MI. This study aims to explore the association between the Blood Pressure-Glucose-Lipid Patterns and MI, and to establish a corresponding warning model. Develop an application software to provide effective support for the early warning and prevention of MI. Methods: A case-control design was used to study 901 valid samples from a hospital in Sichuan. Principal Component Analysis (PCA) was used to identify Blood Pressure-Glucose-Lipid Patterns in each study subject. Logistic regression models were used to evaluate the association between different patterns and MI. The optimal prediction model was used to develop an application software. Results: Through PCA, three patterns of Blood Pressure-Glucose-Lipid were identified. Results from the logistic regression indicated that the association be-tween the patterns and MI was statistically significant. After evaluation by the test set, the accuracy of the prediction model is 0.821 (95% CI: 0.748–0.866), and the area under the curve is 0.710 (0.606−0.814). Conclusion: Different Blood Pressure-Glucose-Lipid Patterns may associate with the risk of MI. A user-friendly and easy-to-use predictive application: CardioRisk, specifically designed and developed to enhance the efficiency of detecting the incidence risk of MI.