Extracting effective factors Incidence of coronary artery disease using association rules

Nowadays cardiovascular diseases are one of the main causes of death in the world. Among the major types of these diseases, correct and in-time diagnosis of coronary artery disease is essential. The most reliable method for CAD diagnosis is angiography, but it is costly and hazardous. In this study the non-invasive methods such as Association rule mining has been used to detect factors which contribute to coronary artery disease which are becoming more popular by day. In this study, Apriori and Predictive Apriori algorithms were applied on 303 patients of Z-Alizadeh Sani Dataset to obtain the rules; and the important factors that are more relevant to the disease were chosen and were used for the two mentioned algorithms. Also, by evaluating the two algorithms we reach the conclusion that the Predictive Apriori gives more applicable rules in predicting the Risk factors of the disease of the unseen samples for new patients.

کلیدواژه ها:

Coronary Artery Disease ، Association Rules Mining ، Apriori algorithm ، Predictive Apriori algorithm