Abstract

Heart disease is one of the most dangerous and common diseases at the international and local levels, especially in recent years, and is considered one of the main causes of death. This study aims to identify the most important factors associated with the risk of heart disease through a methodology that relies on discriminant analysis and logistic analysis as two important and effective models in identifying risk factors resulting from heart disease and correct prediction of it, by subjecting a sample of individuals with heart disease. And healthy people, after obtaining their consent in a group of local clinics and hospitals, taking into account the inclusion criteria that include adults who are between the ages of 40 and 70 and have been diagnosed with heart disease and comorbidities such as high blood pressure, cholesterol, diabetes, smokers, and taking into account the exclusion criteria, which include excluding pregnant and breastfeeding wome patients who have infections or any bacterial infection, or suffer from chronic diseases, as well as patients who take Anticoagulants (such as warfarin) and some psychiatric medications. The results indicated that the two models were close, as the accuracy of the logistic analysis was 83% compared to 79% for the discriminant analysis, the sensitivity of the logistic analysis was 81% and the discriminant analysis was 78%, and the specificity was 84% compared to 80% for the discriminant analysis. As for the American University in Cairo analysis, the results of the logistic analysis reached 89% and 85% for the discriminant analysis. The results also indicated that logistic analysis has greater flexibility in interpreting and identifying infected cases compared to uninfected cases. The most influential variables were similar in the two models: age, smoking, diabetes, and physical activity.
Keywords:( logistic analysis, discriminant analysis, heart disease, accuracy, sensitivity, specificity)

Download Paper

Download Cover