title	colorFrom	colorTo	sdk	app_port	emoji	pinned	license	app_file
AeroPath: automatic airway segmentation using deep learning	indigo	indigo	docker	7860	🫁	false	mit	demo/app.py

🫁 AeroPath 🤗

An airway segmentation benchmark dataset with challenging pathology

AeroPath was developed by SINTEF Medical Image Analysis to accelerate medical AI research.

Brief intro

This repository contains the AeroPath dataset described in "AeroPath: An airway segmentation benchmark dataset with challenging pathology". A web application was also developed in the study, to enable users to easily test our deep learning model on their own data. The application was developed using Gradio for the frontend and the segmentation is performed using the Raidionics backend.

The dataset is made openly available at Zenodo and the Hugging Face Hub. Click any of the two hyperlinks to access the dataset.

Dataset

Accessing dataset

The dataset contains 27 CTs with corresponding airways and lung annotations. The folder structure is described below.

The easiest way to access the data is in Python with Hugging Face's datasets package:

from datasets import load_dataset

# downloads data from Zenodo through the Hugging Face hub
# - might take several minutes (~5 minutes in CoLab)
dataset = load_dataset("andreped/AeroPath")
print(dataset)

# list paths of all available patients and corresponding features (ct/airways/lungs)
for d in dataset["test"]:
  print(d)

A detailed interactive demo on how to load and work with the data can be seen on CoLab. Click the CoLab badge to see the notebook or alternatively click here to see it on GitHub.

Dataset structure

└── AeroPath.zip
    ├── README.md
    └──  AeroPath/
        ├── 1/
        │   ├── 1_CT_HR.nii.gz
        │   ├── 1_CT_HR_label_airways.nii.gz
        │   └── 1_CT_HR_label_lungs.nii.gz
        ├── [...]
        └── 27/
            ├── 27_CT_HR.nii.gz
            ├── 27_CT_HR_label_airways.nii.gz
            └── 27_CT_HR_label_lungs.nii.gz

Demo

To access the live demo, click on the Hugging Face badge above. Below is a snapshot of the current state of the demo app.

Continuous integration

Build Type	Status
HF Deploy
File size check
Formatting check

Development

Docker

Alternatively, you can deploy the software locally. Note that this is only relevant for development purposes. Simply dockerize the app and run it:

docker build -t aeropath .
docker run -it -p 7860:7860 aeropath

Then open http://127.0.0.1:7860 in your favourite internet browser to view the demo.

Python

It is also possible to run the app locally without Docker. Just setup a virtual environment and run the app. Note that the current working directory would need to be adjusted based on where AeroPath is located on disk.

git clone https://github.com/raidionics/AeroPath.git
cd AeroPath/

virtualenv -python3 venv --clear
source venv/bin/activate
pip install -r ./demo/requirements.txt

python demo/app.py --cwd ./

Citation

If you found the dataset and/or web application relevant in your research, please cite the following reference:

@misc{støverud2023aeropath,
  title={{AeroPath: An airway segmentation benchmark dataset with challenging pathology}}, 
  author={Karen-Helene Støverud and David Bouget and Andre Pedersen and Håkon Olav Leira and Thomas Langø and Erlend Fagertun Hofstad},
  year={2023},
  eprint={2311.01138},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}

The dataset is hosted at Zenodo, so you should also cite the following:

@dataset{hofstad2023aeropathzenodo,
  title        = {{AeroPath: An airway segmentation benchmark dataset with challenging pathology}},
  author       = {Hofstad, Erlend and Bouget, David and Pedersen, André},
  month        = nov,
  year         = 2023,
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.10069289},
  url          = {https://doi.org/10.5281/zenodo.10069289}
}

The web application is using the Raidionics backend, thus, also consider citing:

@article{bouget2023raidionics,
    title = {Raidionics: an open software for pre-and postoperative central nervous system tumor segmentation and standardized reporting},
    author = {Bouget, David and Alsinan, Demah and Gaitan, Valeria and Holden Helland, Ragnhild and Pedersen, André and Solheim, Ole and Reinertsen, Ingerid},
    year = {2023},
    month = {09},
    pages = {},
    volume = {13},
    journal = {Scientific Reports},
    doi = {10.1038/s41598-023-42048-7},
}

License

The code in this repository is released under MIT license.