A Machine Learning Approach for Heart Attack Prediction

A heart attack also known as cardiac arrest, diversify various conditions impacting the and became one of chief-reason for death worldwide over last few decades. Approximately, 31% total deaths globally are due to CVDs. It constitutes pinnacle chronic processes which involve complex interactions between risk factors can cannot be improved. Most instances or cases cardiovascular diseases allocated revisable where most considered preventable. ML enhancing approach evolution predictive models in health care industries was decided test algorithms check what extent their prediction scores estimate ameliorate upon results acquired. Researchers deploy machine learning data mining techniques a set enormous patients attain attacks before occurrence helping healthcare professionals. This research comprises Supervised classifiers like, Gradient Boosting, Decision Tree, Random Forest Logistic Regression that have been used model Myocardial Infarction prediction. uses existing datasets from Framingham database others UCI Heart repository. intends ideate probabilities patients. These deployed pipeline using both ways i.e., without optimizations feature transformations well vice-versa. The impersonate Boosting classifier is achieving highest accuracy score such way by our binary form 1 means chance 0 no chance. Some influential attributes chest pain type among typical angina asymptotic least, cholesterol level greater than 200mg/dl more prone, increased rate, thal, age. concluded premature preventable 80% just healthy diet along with regular exercises not tobacco products person who drinks 5 glasses water daily less likely develop attacks. medical checkup Blood-pressure level, rate on basis meditation help you prevent major