---
title: Healthcare Stroke Prediction System
emoji: 🧠
colorFrom: red
colorTo: blue
sdk: streamlit
sdk_version: 1.35.0
app_file: app.py
pinned: false
license: mit
short_description: An ML-powered web dashboard to predict the probability of a stroke.
---

# 🧠 Healthcare Stroke Prediction System

This is a Machine Learning-powered web application that predicts the likelihood of a patient experiencing a stroke based on various clinical and demographic factors (such as age, hypertension, heart disease, smoking status, and BMI).

The application is built using **Streamlit** for the frontend interface and **Scikit-Learn** for the underlying Machine Learning pipeline.

---

## 🚀 Live Demo

You can interact with the live application directly on Hugging Face Spaces.

---

## 🛠️ Features

- **Real-time Prediction:** Input patient details to get instant stroke risk probability.
- **User-Friendly Dashboard:** Clean and intuitive UI built with Streamlit.
- **Robust ML Pipeline:** Features automated scaling and categorical encoding using saved preprocessing artifacts (`scaler.pkl`, `columns.pkl`).

---

## 🗂️ Project Structure

The repository structure for this Hugging Face Space is as follows:

```text
├── app.pytext              # Main Streamlit application script
├── requirements.txt       # Python dependencies
├── model.pkl              # Trained Machine Learning Model (Trained using Jupyter Notebook)
├── scaler.pkl             # Fitted Scaler for numerical features
├── columns.pkl            # Saved feature columns list for consistency
└── README.md              # Project documentation and HF configuration
```