Datasets:
metadata
dataset_info:
features:
- name: image
dtype: image
- name: pathols
sequence: uint8
- name: structs
sequence:
sequence:
sequence: uint8
splits:
- name: train
num_bytes: 17021730614.839962
num_examples: 23094
- name: test
num_bytes: 4255801185.160039
num_examples: 5774
download_size: 732030045
dataset_size: 21277531800
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
license: apache-2.0
task_categories:
- image-classification
- feature-extraction
Dataset Structure
This dataset contains vision data for chest X-ray pathology identification.
Data Fields
- image: The PIL image of the chest X-ray. These images are size (224,224) by default.
- pathols: A binary-valued (14)-shaped array that indicates whether each of the 14 pathologies is present.
- structs: A binary-valued array of shapes (14,224,224) that gives the segmentation for each of the 14 anatomical structures.
The 14 pathologies are:
- Atelectasis
- Cardiomegaly
- Consolidation
- Edema
- Effusion
- Emphysema
- Fibrosis
- Hernia
- Infiltration
- Mass
- Nodule
- Pleural Thickening
- Pneumonia
- Pneumothorax
The 14 anatomical structures are:
- Left Clavicle
- Right Clavicle
- Left Scapula
- Right Scapula
- Left Lung
- Right Lung
- Left Hilus Pulmonis
- Right Hilus Pulmonis
- Heart
- Aorta
- Facies Diaphragmatica
- Mediastinum
- Weasand
- Spine
Data Splits
- train: 23094 samples
- test: 5774 samples
Usage
from datasets import load_dataset
train_dataset = load_dataset("BrachioLab/chestx", split="train")
test_dataset = load_dataset("BrachioLab/chestx", split="test")