AeroPath / README.md
Angelou0516's picture
Upload dataset
fd35e3a verified
metadata
license: cc-by-4.0
task_categories:
  - image-segmentation
modality:
  - CT
language: []
tags:
  - medical-imaging
  - airway-segmentation
  - lung-segmentation
  - thoracic-CT
pretty_name: AeroPath
size_categories:
  - n<100
dataset_info:
  features:
    - name: subject_id
      dtype: int32
    - name: num_slices
      dtype: int32
    - name: ct_middle_slice
      dtype: image
    - name: mask_middle_slice
      dtype: image
    - name: overlay_middle_slice
      dtype: image
  splits:
    - name: train
      num_bytes: 10537061
      num_examples: 27
  download_size: 10544507
  dataset_size: 10537061
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

AeroPath

AeroPath is an airway segmentation benchmark dataset with challenging pathology, containing 27 contrast-enhanced CT scans acquired at St. Olavs Hospital, Trondheim, Norway.

Dataset Summary

Field Details
Modality Contrast-enhanced CT (CTA)
Body Part Chest — airways and lungs
Subjects 27
Labels Airways, Lungs
Total Size ~4.8 GB
License CC-BY 4.0

Data Structure

Each subject folder contains:

  • {N}_CT_HR.nii.gz — CT volume
  • {N}_CT_HR_label_airways.nii.gz — airway segmentation mask
  • {N}_CT_HR_label_lungs.nii.gz — lung segmentation mask

Citation

@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}
}

@article{stoverud2024aeropath,
  title   = {AeroPath: An airway segmentation benchmark dataset with challenging pathology and baseline method},
  author  = {Støverud, Karen-Helene and Bouget, David and Pedersen, André and Langø, Thomas and Hofstad, Erlend Fagertun and others},
  journal = {PLOS ONE},
  volume  = {19},
  number  = {10},
  pages   = {e0311416},
  year    = {2024},
  doi     = {10.1371/journal.pone.0311416}
}