|
|
--- |
|
|
language: |
|
|
- en |
|
|
license: cc-by-4.0 |
|
|
size_categories: |
|
|
- n<1K |
|
|
task_categories: |
|
|
- image-segmentation |
|
|
dataset_info: |
|
|
features: |
|
|
- name: series_uid |
|
|
dtype: string |
|
|
- name: center_irc |
|
|
dtype: string |
|
|
- name: ct_path |
|
|
dtype: string |
|
|
- name: mask_path |
|
|
dtype: string |
|
|
splits: |
|
|
- name: train |
|
|
num_bytes: 11429 |
|
|
num_examples: 100 |
|
|
download_size: 6716 |
|
|
dataset_size: 11429 |
|
|
configs: |
|
|
- config_name: default |
|
|
data_files: |
|
|
- split: train |
|
|
path: data/train-* |
|
|
--- |
|
|
# Dataset Card for LUNA16 Segmentation Dataset |
|
|
|
|
|
This dataset contains CT image slices retrieved from the LUNA16 challenge, which focuses on the automatic detection of lung nodules in CT scans. |
|
|
|
|
|
The dataset is organized into two main folders: `ct/` and `mask/`. |
|
|
|
|
|
The `ct/` folder contains the CT scna images, while the `mask/` folder contains the corresponding segmentation masks. |
|
|
|
|
|
Each CT scan and mask are linked through a `metadata.jsonl` file, which provides metadata for each pair. |
