Federated Learning for Heart Disease Prediction

This project demonstrates a federated learning system for predicting heart disease risk using the Flower framework. Federated learning allows multiple clients (e.g., clinics) to collaboratively train a machine learning model without sharing sensitive patient data.

Features

• Privacy-Preserving: Data remains decentralized across clients, ensuring patient privacy.
• Heart Disease Prediction: Utilizes the UCI Heart Disease dataset to train a binary classification model.
• Scalable: Easily extendable to include more clients or additional privacy techniques.

Technologies Used

• Flower: Federated Learning framework
• TensorFlow: Neural network implementation
• Scikit-learn: Data preprocessing
• Pandas: Data manipulation

Installation

1. Clone the repository:

   git clone https://github.com/ay0788/federated-heart-disease
   cd federated-heart-disease