---
dataset_info:
  features:
  - name: label
    dtype: string
  - name: image
    dtype: image
  - name: original_path
    dtype: string
  - name: segmentation_map
    dtype: image
  splits:
  - name: train
    num_bytes: 14066101161
    num_examples: 52742
  - name: test
    num_bytes: 7365684485
    num_examples: 27172
  - name: validation
    num_bytes: 605894733
    num_examples: 2122
  download_size: 22020516738
  dataset_size: 22037680379
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---
This dataset is only a restructure of the original BMAD dataset. Please cite their work if you use this dataset:

@INPROCEEDINGS{10678042,
  author={Bao, Jinan and Sun, Hanshi and Deng, Hanqiu and He, Yinsheng and Zhang, Zhaoxiang and Li, Xingyu},
  booktitle={2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)}, 
  title={BMAD: Benchmarks for Medical Anomaly Detection}, 
  year={2024},
  volume={},
  number={},
  pages={4042-4053},
  keywords={Computer vision;Benchmark testing;Video surveillance;Retina;Pattern recognition;Medical diagnosis;Medical diagnostic imaging;Anomaly detection;medical images;benchmark},
  doi={10.1109/CVPRW63382.2024.00408}}