tejaswinishelke18/Alzheimers_Disease_Prediction

🧠 Alzheimer's MRI Detection System

An AI-powered web application for early detection of Alzheimer's disease using brain MRI scans. Built with Streamlit and TensorFlow Lite for fast, accurate predictions.

  

🌟 Features

• —🔐 User Authentication: Secure login/register system with password hashing
• —🤖 AI-Powered Analysis: Ensemble of three deep learning models (MobileNet, DenseNet, ResNet)
• —📱 Responsive UI: Clean, modern interface optimized for all devices
• —📊 Detailed Results: Individual model predictions + ensemble voting
• —📈 History Tracking: Save and review all previous analyses
• —🚀 Fast Inference: Optimized TensorFlow Lite models for quick predictions
• —☁️ Cloud Ready: Deployable on Streamlit Cloud, Heroku, or Hugging Face Spaces

🎯 Model Performance

Our ensemble approach combines three state-of-the-art models:

🚀 Quick Start

Local Installation

1. 1.Clone the repository

   git clone https://github.com/yourusername/alzheimer-mri-detection.git
   cd alzheimer-mri-detection

1. 1.Install dependencies

   pip install -r requirements.txt

1. 1.Add your model files

   # Place your .tflite model files in the models/ directory:
   # - models/mobilenet.tflite
   # - models/densenet.tflite  
   # - models/resnet.tflite

1. 1.Run the application

   streamlit run app.py

1. 1.Open your browser to http://localhost:8501

🌐 Online Demo

Try the live demo: **Alzheimer's MRI Detection App**

📁 Project Structure

alzheimer-mri-detection/
│
├── app.py                 # Main Streamlit application
├── requirements.txt       # Python dependencies
├── README.md             # This file
├── setup.sh              # Streamlit Cloud setup
├── .gitignore           # Git ignore rules
├── runtime.txt          # Python version
│
├── models/              # TensorFlow Lite models
│   ├── mobilenet.tflite
│   ├── densenet.tflite
│   └── resnet.tflite
│
├── data/                # Sample MRI images
│   ├── sample_alzheimer.jpg
│   └── sample_normal.jpg
│
└── .streamlit/          # Streamlit configuration
    └── config.toml

🔧 Configuration

Model Requirements

• —Format: TensorFlow Lite (.tflite)
• —Input Shape: (1, 224, 224, 3)
• —Output Shape: (1, 2) - [Alzheimer's probability, Normal probability]
• —Preprocessing: Images are resized to 224x224 and normalized to [0,1]

Environment Variables

For production deployment, you can set:

# Optional: Custom database path
DB_PATH=/path/to/database.db

# Optional: Models directory
MODELS_DIR=/path/to/models/

# Optional: Enable debug mode
DEBUG=True

📊 Usage

1. Register/Login

• —Create a new account or login with existing credentials
• —All data is stored locally in SQLite database

2. Upload MRI Scan

• —Supported formats: JPG, JPEG, PNG
• —Optimal resolution: 224x224 pixels
• —Try sample images to test the system

3. Get AI Analysis

• —View individual model predictions
• —See ensemble result with confidence score
• —Results are automatically saved to your history

4. Review History

• —Access all previous analyses
• —Compare results over time
• —Export data for medical consultations

🏥 Medical Disclaimer

⚠️ IMPORTANT: This application is for educational and research purposes only. - NOT a medical diagnosis tool - NOT a substitute for professional medical advice - Always consult healthcare professionals for medical decisions - Results should be verified by qualified radiologists

🚀 Deployment

Streamlit Cloud

1. 1.Fork this repository
2. 2.Connect to Streamlit Cloud
3. 3.Deploy directly from GitHub
4. 4.Add your model files to the models/ directory

Hugging Face Spaces

1. 1.Create a new Space on Hugging Face
2. 2.Choose Streamlit as the SDK
3. 3.Upload your files including model files
4. 4.Your app will be live at https://huggingface.co/spaces/USERNAME/SPACE_NAME

Heroku

1. 1.Install Heroku CLI
2. 2.Create a new Heroku app
3. 3.Deploy using Git:

   git add .
   git commit -m "Deploy to Heroku"
   git push heroku main

🛠️ Development

Adding New Models

1. 1.Train your model using TensorFlow/Keras
2. 2.Convert to TensorFlow Lite:

   converter = tf.lite.TFLiteConverter.from_keras_model(model)
   tflite_model = converter.convert()
   
   with open('new_model.tflite', 'wb') as f:
       f.write(tflite_model)

1. 1.Add to the ensemble in app.py:

   model_files = {
       "MobileNet": MODELS_DIR / "mobilenet.tflite",
       "DenseNet": MODELS_DIR / "densenet.tflite", 
       "ResNet": MODELS_DIR / "resnet.tflite",
       "NewModel": MODELS_DIR / "new_model.tflite",  # Add here
   }

Running Tests

# Install test dependencies
pip install pytest pytest-streamlit

# Run tests
pytest tests/

📈 Performance Optimization

• —TensorFlow Lite: 5-10x faster inference than full TensorFlow
• —Model Quantization: Reduces model size by 75% with minimal accuracy loss
• —Caching: Models and database connections are cached for faster response
• —Batch Processing: Support for multiple image analysis (coming soon)

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines.

1. 1.Fork the repository
2. 2.Create a feature branch (git checkout -b feature/amazing-feature)
3. 3.Commit your changes (git commit -m 'Add amazing feature')
4. 4.Push to the branch (git push origin feature/amazing-feature)
5. 5.Open a Pull Request

📄 License

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

🙏 Acknowledgments

• —Dataset: Augmented Alzheimer's MRI Dataset
• —Models: Based on MobileNet, DenseNet, and ResNet architectures
• —Framework: Built with Streamlit and TensorFlow

📞 Support

• —Issues: GitHub Issues
• —Discussions: GitHub Discussions
• —Email: your.email@example.com

<div align="center">

🧠 Early Detection Saves Lives | Built with ❤️ for Healthcare

🌟 Star this repo • 🐛 Report Bug • 💡 Request Feature

</div>