❤️ Cardiac Attack Risk Prediction by Deep learning

A production‑ready full‑stack mini project that predicts cardiac attack risk using patient vitals and lifestyle factors. The system is powered by Deep Learning, secured with Firebase Authentication, and deployed online for real‑world use.

🚀 Live Backend: https://cardiac-backend-xe0p.onrender.com

---

🔍 Project Overview

This application allows users to sign up, log in, and predict heart attack risk using a trained neural network model. Predictions are returned instantly through a FastAPI backend. The project is designed for college mini‑projects, ML demonstrations, and full‑stack learning.

---

✨ Key Features

🧠 Deep Learning Prediction

• Multi‑Layer Perceptron (MLP) Neural Network
• Trained on UCI, synthetic, and lifestyle datasets
• Returns probability‑based cardiac risk score

🔐 Secure Authentication

• Firebase Authentication (Signup / Login)
• JWT‑protected API routes
• Only authenticated users can access predictions

⚡ Fast & Interactive

• FastAPI backend with async endpoints
• React frontend with Dark Mode and responsive UI
• Real‑time predictions (no page reloads)

☁️ Cloud Deployment

• Backend hosted on Render
• Frontend deployable via Firebase Hosting
• Firebase Admin SDK for secure server‑side access

---

🏗️ Tech Stack

Layer	Technology
Frontend	React, Vite, Tailwind CSS
Backend	FastAPI, Uvicorn
ML	Scikit‑Learn (MLPClassifier)
Auth	Firebase Authentication
Database	SQLite / Firestore
Hosting	Render (Backend), Firebase (Frontend)

---

📂 Project Structure

cardiac-prediction/
│
├── backend/
│   ├── main.py              # FastAPI entry point
│   ├── auth/                # Authentication routes & utils
│   ├── profile_routes.py    # Protected user APIs
│   ├── database.py          # Firebase / DB initialization
│   ├── models/              # Trained ML models (.pkl)
│   └── requirements.txt
│
├── client/
│   ├── src/
│   │   ├── components/      # Login, Signup, Predictor UI
│   │   ├── context/         # Auth Context
│   │   └── pages/
│   └── package.json
│
├── train_model.py           # Model training
├── generate_data.py         # Dataset generation
├── cardiac.db               # Local SQLite DB
└── README.md

---

▶️ Local Development

1️⃣ Backend Setup

pip install -r backend/requirements.txt
python -m uvicorn backend.main:app --reload --host 0.0.0.0 --port 8000

Backend runs at:

http://localhost:8000

---

2️⃣ Frontend Setup

cd client
npm install
npm run dev

Frontend runs at:

http://localhost:5173

---

🌐 Production Deployment

Backend (Render)

• Build Command:

pip install -r backend/requirements.txt

• Start Command:

uvicorn backend.main:app --host 0.0.0.0 --port $PORT

• Environment Variable:

FIREBASE_CREDENTIALS=<Firebase service account JSON>

---

Frontend (Firebase Hosting)

npm run build
firebase deploy

Add deployed frontend domain to:

Firebase Console → Authentication → Authorized Domains

---

🔐 Authentication Flow

1. User signs up or logs in
2. Firebase issues JWT
3. JWT sent in API requests
4. Backend verifies token
5. Prediction endpoint unlocked

---

---

⭐ Future Improvements

• Doctor/Admin dashboard
• Email & SMS alerts
• Model retraining pipeline
• Improved risk explainability

---

⚠️ Disclaimer: This project is for educational purposes only and should not be used for real medical diagnosis.