LungScan-AI

🔬 Revolutionizing Lung Cancer Detection with AI

LungScan-AI is a state-of-the-art medical imaging platform that harnesses the power of deep learning to detect lung cancer with 97.03% accuracy. By analyzing chest X-rays in seconds, it provides healthcare professionals with rapid, reliable insights for early cancer detection and improved patient outcomes.

✨ Key Highlights

• High Accuracy: 97.03% accurate in detecting lung cancer variants
• Instant Analysis: Get comprehensive results in seconds
• Smart Reporting: Detailed medical reports with actionable insights
• User-Friendly: Intuitive interface for healthcare professionals

✨ Key Highlights

• High Accuracy: 97.03% accurate in detecting lung cancer variants
• Instant Analysis: Get comprehensive results in seconds
• Smart Reporting: Detailed medical reports with actionable insights
• User-Friendly: Intuitive interface for healthcare professionals

🔗 Try the Live Demo

UI Preview

Here's a preview of the LungScan-AI application interface:

User Registration

User registration interface for secure access

Analysis Interface

X-ray upload and analysis interface showing probability distribution

Detailed Results

Detailed analysis results with key findings and recommendations

Features

• Upload and analyze chest X-ray images
• Detect three types of conditions:
  • Normal
  • Squamous Cell Carcinoma
  • Adenocarcinoma
• Detailed medical report generation
• Interactive visualization of results
• User registration and history tracking

Technical Stack

• Python 3.11+
• Streamlit for web interface
• PyTorch for deep learning (ResNet18)
• SQLite for database

Deployment

Local Deployment

1. Follow the installation steps above
2. Run the application using streamlit run src/app.py
3. Access the application at http://localhost:8501

Streamlit Cloud Deployment

1. Push your code to GitHub
2. Visit share.streamlit.io
3. Connect your GitHub repository
4. Select src/app.py as the main file
5. Deploy

Project Structure

├── src/
│   ├── app.py              # Main Streamlit application
│   ├── config.py           # Configuration settings
│   ├── data_preprocessing.py # Data preprocessing utilities
│   ├── train.py            # Model training script
│   └── database.py         # Database operations
├── models/
│   └── lung_cancer_detector.pth  # Trained model
├── data/
│   └── train/              # Training data directory
├── requirements.txt        # Python dependencies
└── README.md              # This file

Model Performance

• Accuracy: 97.03%
• Supports detection of three classes
• Based on ResNet18 architecture

Security and Privacy

• User data is stored securely in a local database
• X-ray images are processed locally
• No data is shared with external services

Security Notes

• Never commit .streamlit/secrets.toml to version control
• Use environment variables or secrets management for sensitive data
• For Streamlit Cloud deployment, configure secrets in the Streamlit Cloud dashboard
• Regularly rotate passwords and access credentials
• Monitor access logs for suspicious activity

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Copyright 2024 LungScan-AI

<<<<<<< HEAD