A Holistic User Centric Acute Myocardial Infarction Prediction System With Model Evaluation Using Data Mining Techniques

摘要

Acute Myocardial Infarction (Heart Attack), a Coronary Heart Disease (CHD) is one of the major killers worldwide. Around one thousand data has been collected from AMI patients, people are at risk of maybe a heart attack and individuals with the significant features closely related to heart attack. The sophistication in mobile technology, health care applications offers remarkable opportunities to improve our health, safety and in some sense preparedness to common illnesses. The excess delay time between the onset of a heart attack and seeking treatment is a major issue which may lead to permanent blockage or even die often. So, a proficient mobile application approach is projected in this paper that may predict the possibilities of a attack once an individual is bearing the noticeable symptoms of chest pain. Random forest predicts the result of the user input features and the automated result is shown on the smartphone application. The application categorizes the prediction of the user’s input as a heart attack, maybe heart attack and no heart attack. The experimental results showed that the accuracy of the proposed technique is 92%, whereas the precision is 95%, 92%, 87% respectively for heart attack, maybe heart attack and no heart attack prediction. Our research target is to raise heart attack awareness on time in an innovative way through available and accessible medium to mass people.