🚀 Cardiac Forecasting System (HeartGuard)

🔗 Repository: https://github.com/anasrobo/Cardiac_Forecasting_System/tree/main

---

💡 Project Overview

HeartGuard is a Flask‑powered web app that predicts your heart disease risk using 11 non‑invasive clinical features.
By excluding the angiogram‑based “ca” feature, we help you dodge unnecessary invasive tests and save ₹20,000–₹80,000 per procedure.

---

📊 Key Features

• 🏷️ Binary Classification: High vs. low chance of heart disease
• 🔍 11 Core Metrics: Age, sex, chest pain, BP, cholesterol, ECG, FBS, HR, angina, ST‑depression, ST‑slope
• 🚫 No “ca” Feature: Avoids reliance on invasive angiogram data
• 🤖 Multi‑Model Soft Ensemble: Logistic Regression, SVM, Decision Tree, Random Forest, XGBoost
• 🌐 Flask Web App: Slick UI in templates/index.html
• 📄 PDF Reports: Downloadable summary with trend charts

---

🗂️ Dataset

1. Primary: Cleveland + Hungary Final Dataset
2. Excluded:
   • “Heart Disease Data” (contains “ca” feature)
   • heart.csv (723 duplicates)
   • A4_Cardiac_Disease.csv (too many duplicates)
   • Tiny datasets (< 400 rows)

---

⚙️ Installation

git clone https://github.com/anasrobo/Cardiac_Forecasting_System.git cd Cardiac_Forecasting_System

python3 -m venv venv #macOS/Linux source venv/bin/activate

#Windows .\venv\Scripts\activate

pip install -r requirements.txt No requirements.txt? pip install flask pandas scikit-learn joblib fpdf matplotlib

🚀 Usage

Explore & Train jupyter notebook Cardiac_Forecasting_Tool.ipynb Run Flask App

#macOS/Linux export FLASK_APP=app.py

#Windows set FLASK_APP=app.py

Flask run

– Open 👉 http://127.0.0.1:5000 – Fill out your clinical metrics – Get instant risk % & per‑model breakdown – Hit “Generate Report” to snag a PDF

📸 Screenshots

1. Hero Section & Overview

2. Input Form UI

3. Features and Benefits

4. Testimonials & Footer (Adjust file names/paths if you rename your images.)

5. Predictions

6. Risk Forecasting

7. PDF Report Predictions page-1

8. PDF Report Predictions page-2

🌟 Roadmap

🎨 Responsive UI revamp with React

🧪 Unit tests + CI/CD pipeline

☁️ Deploy to Heroku / Render / AWS

📦 Dockerize for one‑click launch

🤝 Contributing

Fork & clone Create branch: git checkout -b feature/your-awesome-feature Commit & push: git commit -m "✨ Add awesome feature" git push origin feature/your-awesome-feature Open a PR — let’s collab! 🚀

📄 License

Licensed under MIT. See LICENSE for details.

“The best way to predict the future is to create it.” – Peter Drucker

Made with ❤️ by Anas