🔬 AI Skin Disease Detection System

An AI-powered skin disease detection system that analyzes skin lesion images and classifies them into 8 categories. Features an interactive chatbot assistant to explain results, treatments, and causes.

📁 Model Files

Important: The trained model files are too large (>100MB) for GitHub, so they are hosted on Google Drive:

📥 Download Models: Google Drive Link

Please download the model files and place them in the models/final_model_optimized/ folder before running the application.

👥 Project Team

• Paresh Suva - Team Leader
• Pranay Suthar - Developer
• Krish Zalavadiya - Developer
• Samarth Patel - Developer

🎯 Features

• 8-Class Skin Disease Classification

  • Actinic Keratosis (precancerous)
  • Basal Cell Carcinoma (skin cancer)
  • Benign Keratosis (harmless)
  • Dermatofibroma (benign)
  • Melanocytic Nevus (common mole)
  • Melanoma (dangerous cancer)
  • Squamous Cell Carcinoma (skin cancer)
  • Vascular Lesion (blood vessel abnormality)

• AI Chatbot Assistant - Ask questions about your diagnosis

• Detailed Information - Causes, treatments, home care tips

• Educational Resources - YouTube videos, medical articles

• Downloadable Reports - Save analysis results

🚀 Live Demo

Run the application locally using the instructions below.

�️ Local Setup

1. Clone the Repository

git clone https://github.com/Pranay-Suthar/Skin-disease-analysis-using-Deep-Learning.git
cd Skin-disease-analysis-using-Deep-Learning

2. Download Model Files

Download the trained model from Google Drive and extract to:

models/final_model_optimized/
├── config.json
├── model.safetensors
└── preprocessor_config.json

3. Create Virtual Environment

python -m venv venv

# Windows
venv\Scripts\activate

# macOS/Linux
source venv/bin/activate

4. Install Dependencies

# For running the app
pip install -r requirements-streamlit.txt

# For full development (includes training dependencies)
pip install -r requirements.txt

5. Set Up Environment (Optional - for chatbot)

cp .env.example .env
# Edit .env and add your GROQ_API_KEY

6. Run the App

streamlit run skin_app.py

Open http://localhost:8501 in your browser.

📁 Project Structure

skin-disease-detection/
├── skin_app.py              # Main Streamlit application
├── train_cpu.py             # Training script for CPU
├── train_optimized_colab.py # Training script for GPU (Colab)
├── requirements.txt         # Full dependencies (dev + training)
├── requirements-streamlit.txt # Minimal dependencies (deployment)
├── .streamlit/
│   └── config.toml          # Streamlit configuration
├── .env.example             # Environment template
├── .gitignore               # Git ignore rules
├── README.md                # This file
├── LICENSE                  # MIT License
└── models/
    └── final_model_optimized/ # Our trained model

🖥️ Usage

1. Upload Image: Click "Browse files" and select a skin lesion image (JPG, PNG)
2. Analyze: Click "🚀 Analyze Image" to get AI prediction
3. View Results: See the predicted condition, confidence score, and severity
4. Chat: Ask the AI assistant questions about your results
5. Learn More: Check the resources section for treatments and articles
6. Download Report: Save your analysis as a text file

🛠️ Tech Stack

• Frontend: Streamlit
• ML Model: Custom trained Swin Transformer (offline)
• Inference: PyTorch + Transformers
• Chatbot: Groq API with LLaMA 3.1
• Dataset: ISIC (International Skin Imaging Collaboration)

📊 Model Performance

Metric	Score
Validation Accuracy	85-92%
Top-3 Accuracy	96-98%

🔄 Using Your Own Model

If you have trained your own skin disease classification model, you can easily integrate it:

Option 1: Replace Existing Model

1. Place your model files in models/final_model_optimized/
2. Ensure files are named: config.json, model.safetensors, preprocessor_config.json
3. Update the LABEL_MAP in skin_app.py if your classes are different

Option 2: Add New Model Path

1. Update LOCAL_MODEL_PATH in skin_app.py:

   LOCAL_MODEL_PATH = "models/your_model_folder"

2. Place your model files in the new folder
3. Restart the application

Supported Model Formats

• HuggingFace Transformers format (recommended)
• PyTorch models with AutoImageProcessor
• Models with 8-class skin disease classification

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

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

📝 License

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

🙏 Acknowledgments

• ISIC Archive for the skin lesion dataset
• HuggingFace for model hosting
• Groq for free LLM API access
• Medical resources from Mayo Clinic, AAD, and Skin Cancer Foundation

---

⭐ If you found this project helpful, please give it a star!