<![CDATA[Diabetes and heart disease are leading causes of global death, and if not properly managed, can be fatal. Previous studies have explored the factors influencing these two diseases through correlation analysis, multivariate analysis, machine learning, and deep learning. However, these approaches generally only identify associations without being able to predict causal relationships. This study aims to model the causal relationships between factors in two clinical datasets: heart disease (13 parameters) and diabetes (9 parameters), in order to support early diagnosis and prevention. The Greedy Equivalence Search (GES) algorithm is used to determine the direction of the causal relationship between parameters. The results show that heart disease exhibits three directional relationships: between blood pressure and age, between Maximum Heart Rate (MHR) and age, and between age and cholesterol. Then, diabetes exhibits two bidirectional relationships: blood pressure and BMI, then BMI and Diabetes Pedigree Function. In addition, diabetes also exhibits three directional relationships: Diabetes Pedigree Function and Glucose, BMI and Glucose, and Blood Pressure and Glucose. Thus, it can be concluded that the algorithm can identify the causal relationship between diabetes and heart disease.]]>
Copyrights © 2025
