🧠 Stroke Prediction using Machine Learning

Course Project: 191ROE051T - Machine Learning for Robotics

Predicting stroke risk using patient health metrics and machine learning

📋 Table of Contents

• Overview
• Features
• Installation
• Usage
• Project Structure
• Results
• Contributing
• License
• Acknowledgements

🌟 Overview

This project implements a machine learning pipeline to predict the likelihood of a patient having a stroke based on various health parameters. The model helps in early detection of stroke risk, enabling timely medical intervention.

Key Metrics (on test set):

• Accuracy: 95.3%
• Precision: 0.72
• Recall: 0.52
• F1-Score: 0.60
• AUC-ROC: 0.86

✨ Features

• Comprehensive EDA with interactive visualizations
• Feature Engineering with domain-specific transformations
• Multiple ML Models including Random Forest, XGBoost, and LightGBM
• Hyperparameter Tuning using Optuna
• Model Explainability with SHAP values
• Deployment-ready API using FastAPI

🚀 Installation

1. Clone the repository

   git clone https://github.com/theaathish/stroke-prediction.git
   cd stroke-prediction

2. Create and activate virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Download the dataset

   • Get the dataset from Kaggle
   • Place healthcare-dataset-stroke-data.csv in the data/ directory

💻 Usage

1. Run the full pipeline

# Preprocess data
python -m src.data_preprocessing

# Train model
python -m src.model

# Start the web app
python -m src.app

2. Explore the notebooks

Check out the Jupyter notebooks in the notebooks/ directory for detailed analysis and experimentation.

📁 Project Structure

stroke-prediction/
├── data/                    # Raw and processed data
│   ├── raw/                 # Original dataset
│   └── processed/           # Processed datasets
│
├── notebooks/               # Jupyter notebooks
│   └── Stroke_Prediction_Analysis.ipynb
│
├── src/                     # Source code
│   ├── __init__.py
│   ├── data_preprocessing.py
│   ├── feature_engineering.py
│   ├── model.py
│   ├── train.py
│   └── app.py
│
├── models/                  # Trained models
│   └── stroke_model.pkl
│
├── reports/                 # Reports and visualizations
│   └── figures/
│
├── tests/                   # Unit tests
│   └── test_*.py
│
├── .gitignore
├── requirements.txt
└── README.md

📊 Results

Feature Importance

Confusion Matrix

ROC Curve

🤝 Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📜 License

Distributed under the MIT License. See LICENSE for more information.

🙏 Acknowledgements

• Kaggle for the dataset
• Scikit-learn for ML tools
• XGBoost and LightGBM
• SHAP for model interpretability

---

Developed with ❤️ by @theaathish

📅 August 2025