🧠 Cancer Prediction API

A production-ready Machine Learning API that predicts cancer risk using patient data. Built with FastAPI and a trained Random Forest model.

---

🚀 Features

• 🧠 Machine Learning model (Random Forest)
• ⚡ FastAPI backend
• ✅ Input validation (strict feature check)
• 📊 Confidence score & probability output
• 📝 Prediction logging system
• 🏥 Cancer risk classification (High / Low)
• 🔍 Swagger UI for testing

---

🛠 Tech Stack

• Python
• FastAPI
• Scikit-learn
• NumPy

---

📁 Project Structure

cancer-prediction-api/ │ ├── app.py ├── train.py ├── model.pkl ├── requirements.txt ├── README.md │ ├── cancer-api-output.png │ ├── logs/ │ └── predictions.log │ └── data/

---

⚙️ Installation

pip install -r requirements.txt

---

▶️ How to Run

python train.py
uvicorn app:app --reload

---

🌐 API Access

http://127.0.0.1:8000/docs

---

🧪 Example Request

{ "features": [14.0, 20.0, 90.0, 600.0, 0.1, 0.2, 0.15, 0.1, 0.2, 0.07, 0.5, 1.2, 3.5, 40.0, 0.005, 0.02, 0.03, 0.01, 0.02, 0.003, 15.0, 25.0, 100.0, 700.0, 0.15, 0.25, 0.3, 0.15, 0.3, 0.08] }

---

📸 Swagger Output

---

📊 Example Response

{ "prediction": "Low Cancer Risk", "confidence": "53.00%", "probabilities": { "low_risk": 0.53, "high_risk": 0.47 } }

---

⚠️ Notes

• Model requires exactly 30 input features
• Logs stored in logs/predictions.log

---

🚀 Future Improvements

• Model deployment (Docker)
• Frontend integration
• Authentication system

---

👨‍💻 Author

Sivaguru Arumugam

---

⭐ Support

If you like this project, give it a ⭐ on GitHub!