Parametric Spine Model Viewer

A machine learning-powered interactive visualization tool for generating and manipulating 3D spine models based on clinical parameters. This project combines statistical shape modeling with an interactive web interface to allow real-time manipulation of spine morphology.

🌟 Features

• Real-time 3D spine visualization
• Interactive parameter adjustment via sliders
• Multiple viewing angles (Front, Side, Top)
• Model export functionality
• Parameter validation
• Consistent camera positioning
• STL file generation

📊 Parameters

The model uses the following clinical parameters:

• PI (Pelvic Incidence): 30-70°
• PT (Pelvic Tilt): 5-35°
• SS (Sacral Slope): Derived from PI - PT
• LL (Lumbar Lordosis): -60 to -20°
• GT (Global Tilt): 20-40°
• LDI (Lumbar Distribution Index): 80-120
• TPA (T1 Pelvic Angle): 15-35°
• Cobb Angle: 0-30°

Additional derived parameters:

• LL-PI
• RPV (Relative Pelvic Version)
• RLL (Relative Lumbar Lordosis)
• RSA (Relative Spinopelvic Alignment)
• GAP (Global Alignment and Proportion)

🛠️ Technical Stack

• Frontend: Streamlit
• 3D Visualization: PyVista, VTK
• Machine Learning: scikit-learn (PCA, Ridge Regression)
• 3D Processing: Trimesh
• Data Management: NumPy, Pandas
• Model Persistence: Joblib

🚀 Installation

1. Clone the repository:

git clone [repository-url]
cd parametric-spine-model

2. Create and activate virtual environment:

python -m venv env
source env/bin/activate  # Linux/Mac
# or
.\env\Scripts\activate  # Windows

3. Install requirements:

pip install -r requirements.txt

4. For Linux users, increase inotify watch limit:

echo fs.inotify.max_user_watches=524288 | sudo tee -a /etc/sysctl.conf
sudo sysctl -p

📦 Project Structure

parametric-spine-model/
├── trained_models/           # Trained statistical shape models
│   └── spine_model.joblib
├── stl/                     # Reference mesh files
│   └── 1.stl               # Reference topology
├── exported_spines/         # Generated models
├── requirements.txt         # Project dependencies
├── spine_app.py            # Main Streamlit application
└── README.md               # This file

🎮 Usage

1. Start the application:

streamlit run spine_app.py

2. Access the web interface at http://localhost:8501

3. Use the sliders to adjust spine parameters

4. Change view angles using the radio buttons

5. Export models using the "Export Model" button

🔬 Model Architecture

The system uses a two-stage approach:

1. Statistical Shape Model: PCA-based model capturing spine shape variations
2. Parameter Mapping: Ridge regression mapping between clinical parameters and PCA space

🎯 Features in Development

• Parameter presets for common spine configurations
• Advanced measurement tools
• Cross-sectional visualization
• Parameter relationship visualization
• Animation between states
• Additional view angles
• Export of parameter reports

📄 License

[Your chosen license]

🙏 Acknowledgments

Based on research and development from:

• [Your Institution/Lab]
• Statistical Shape Model methodology from [Reference]
• Clinical parameter definitions from [Reference]

📧 Contact

[Your Contact Information]

📚 Citation

If you use this software in your research, please cite: