This workshop demonstrates how to:
1. Download publicly available MRI data from OpenNeuro
2. Run BabySeg locally using Docker
3. Visualise and inspect segmentation outputs
Python IDE (recommended: VS Code)
Participants should install a Python development environment (IDE) before the workshop. We recommend Visual Studio Code (VS Code), which provides an integrated terminal, notebook support, and a convenient interface for editing and running Python code.
Download VS Code:
Windows, Mac & Linux: https://code.visualstudio.com/
After installation, launch VS Code and verify that you can:
Open a folder on your computer Open a terminal within VS Code Run Python scripts and Jupyter notebooks
Other Python IDEs are also suitable (e.g., PyCharm, Spyder, JupyterLab), but the workshop materials and screenshots will use VS Code.
Docker
BabySeg is run inside a Docker container. Docker must be installed before the workshop.
Install Docker Desktop:
- Windows & Mac: https://www.docker.com/products/docker-desktop/
- Linux: https://docs.docker.com/engine/install/
After installation, open a terminal (for windows, open a terminal in Docker Desktop), and verify Docker is working using:
```bash
docker run hello-world
```
You should see a message beginning with:
Hello from Docker!
In VS Code (or an alternative interface), clone the repository by typing the following in a terminal. This should be in the directory where you want the workshop installed. For windows users type clone . in the Docker terminal to open VS Code.
```bash
git clone <REPOSITORY_URL>
cd OpenInfantMRI_workshop
```
or download the ZIP file from GitHub and extract it.
In VS Code (or an alternative interface), open a terminal and install the Python packages for the workshop:
```bash
pip install -r requirements.txt
```
Or have VS Code make the environment for you by pointing to the repo.
Follow the guides in the notebooks:
1. 01_Download_data.ipynb
2. 02_install_babyseg.ipynb
3. 03_run_babyseg_on_opendata.ipynb