Needle Segmentation For Real-time Guidance of Minimally Invasive Procedures Using Handheld 2D Ultrasound Systems 🩺💉

A machine learning system for real-time guidance of minimally invasive procedures.

📖 Table of Contents

• Installation
• Usage
• Project Structure
• Data
• Notebooks
• License

💻 Installation

git clone https://github.com/ColinCoderUmich/needle_segmentation.git
cd needle_segmentation

# Create virtual environment
python -m venv nseg
nseg\Scripts\activate  # Windows

# Install dependencies
pip install -r requirements.txt

# Obtain data and copy to folders
python src/obtain_image_data.py --data_path {path-to-imgs-and-csv-files}

🛠️ Usage

Train Model

python src/train_pytorch.py --arch "UNet" -sd '/content/needle_segmentation/saved_models' --batch_size 8 --ep 5 -sz 256 --name 'UNet_orig'

📂 Project Structure

.
├── data/                       # Data storage
│   ├── raw/                    # Original CSVs and images
│   ├── processed/masks         # Processed data including generated masks
│   ├── interim/
│   └── external/          
│
├── results/                    # CSV Results (performances)
│   └── segmentation/           # Performance of segmentation models
│
├── notebooks/                  # Research notebooks
│   └── needle_segmentation.ipynb  # Final modular notebook analysis
│
├── src/                        # Core code
│   ├── models_definitions/     # Model architecture definitions
│   ├── detect.py               # Data ingestion
│   ├── general_utils.py        # Feature pipeline
│   ├── obtain_model.py         # Model classes
│   ├── train_pytorch.py        # Training script
│   ├── train_arch.py           # Testing model definitions
│   └── test_model.py           # Inference script
│
├── saved_models/               # Saved model/weights are saved here
├── requirements.txt            # Dependencies
└── README.md                   # This file

📊 Data

Source: Local and from Rutgers University
Format: Images and csv files detailing needle delineation

📓 Notebooks

The notebooks/ directory contains the notebooks:

notebooks/
└── needle_segmentation.ipynb  # training and testing

📜 License

MIT License - See LICENSE for details

❤️ Acknowledgments

• Makerere Research and Innovations Fund for funding
• Rutgers University for data access