Heart Attack Prediction Using Machine Learning — A Complete Project Overview

Heart disease is one of the leading causes of death globally. Early detection can save lives. To address this, I developed a Heart Attack Prediction System using Machine Learning, integrated into a beautiful, user-friendly interactive web interface using Streamlit.

✅ Project Objective

The primary goal of this project is to predict the likelihood of a heart attack based on patient data. It analyzes various health parameters using machine learning models trained on a comprehensive heart disease dataset.

🧠 Tools & Technologies Used

• Python
• Pandas, NumPy
• Scikit-learn
• Random Forest Classifier
• Streamlit
• HTML/CSS (for styling)
• Pickle

📊 Dataset Details

I used a health dataset containing various medical features like age, cholesterol levels, blood pressure, and more, categorized by the presence or absence of heart disease.

🛠️ Steps Involved

1. Data Preprocessing

• Dropped unnecessary identifiers like id.
• Converted the target variable 'Heart Disease' (Presence/Absence) into binary format (1 or 0).
• Sampled the dataset for efficient training.
• Split data into training and testing sets (80/20 split).

2. Feature Selection

Extracted the most significant features and saved them as features.pkl for consistent prediction in the deployment phase.

3. Model Training

• Implemented Random Forest Classifier with 100 estimators and max depth of 10.
• Achieved high accuracy on both training and test datasets safely avoiding overfitting.

The Random Forest model proved to be highly effective and robust.

4. Web App with Streamlit

Built a modern user interface where users can enter health metrics. On submission, the application uses the trained model to instantly predict the likelihood of a heart attack.

5. Model Deployment

Deployed the interactive web application seamlessly using Streamlit.

🌐 User Interface Preview

A highly responsive and clean dashboard. Users input medical details and get instant prediction feedback.

🧪 Try It Live

Click here to try the Heart Attack Predictor App

📌 Final Thoughts

This project not only strengthened my understanding of machine learning algorithms like Random Forest, but also gave me hands-on experience with Streamlit app deployment. I believe predictive models in healthcare can serve as an excellent preliminary diagnostic tool.

🔗 Connect with Me
🌐 www.tauqueeralam.com
📱 LinkedIn | GitHub

View a live demo below: