Prediction of Heart Attack

This project uses the Heart Disease Data Set from the UCI Machine Learning Repository to identify major contributors of Heart Attack and to predict patients with a high chance of having Heart Attack.

Survey

We surveyed 84 people to find out what they thought were the major contributors to Heart Attack and here is what they said:

72.6% said Cholesterol whereas 48.8% said Age

But are these really the biggest contributors to Heart Attack? This project aims to explore exactly that

Models Used:

1. Logistic Regression
2. StepAIC
3. Decision Tree

Major Contributors Identified

1. Values for Thalassemia (Normal/Fixed Defect/Reversable Defect)
2. The number of major vessels colored by Fluoroscopy (0 - 3)
3. Maximum heart rate achieved
4. Chest pain type (Typical Angina/Atypical Angina/Non-Anginal Pain/Asymptomatic)

We achieved an accuracy of 86.67% using the Decision Tree Model