🛡️ OncoShield – Non-Invasive Tumor Detection System

OncoShield is a non-invasive, AI-based Tumor Detection System that integrates Thermography, Near-Infrared Spectroscopy (NIR), and Electrical Impedance Spectroscopy (EIS) to provide early-stage Tumor detection. It uses a hybrid machine learning model and is built with a modern web interface to make early diagnosis more accessible, affordable, and scalable.

---

🚀 Features

• 🌡️ Combines multiple non-invasive sensing technologies (Thermal, NIR, EIS)
• 🤖 AI-based prediction using CNN and LSTM
• 💻 Frontend built with React.js for real-time interaction (under development)
• 📊 Time-windowed analysis for greater diagnostic accuracy
• ☁️ Backend and cloud-hosted ML model (planned)

---

🧬 ML Model Pipeline (currently icludes core components)

[ NIR/Infrared Image ]      [ EIS Time-Series ]
↓ ↓                               ↓ ↓
CNN Model                     LSTM Model
↓ ↓                               ↓ ↓
→→→→→→→→→→→ Ensemble Learning →→→→→→→→→→→→
                   ↓
            Final Prediction

🤝 Contributing

Contributions are welcome!

• 1.Fork the repo
• 2.Create a new branch
• 3.Commit your changes
• 4.Submit a pull request

📬 Contact

Developer: Sourav Pati 📧 Email: k0259.mpsbls@gamil.com 🌐 GitHub: sourav-625

Disclaimer:
This project is intended solely for educational and research purposes.
It has not been reviewed, tested, or approved by any certified medical professionals or regulatory bodies.
The system is not intended for clinical diagnosis and should not be used to make medical decisions.
Additionally, the current model may not be capable of detecting all tumor types or conditions, especially deep internal tumors or cancers.