Urinary Sediment Dataset

Dataset Overview

Dataset Name

• Urinary Sediment Dataset

Dataset Content

• Contains 5,376 annotated images covering seven urinary sediment particle categories:
  • cast
  • cryst (crystals)
  • epith (epithelial cell)
  • epithn (epithelial nuclei)
  • eryth (erythrocyte)
  • leuko (leukocyte)
  • mycete

Dataset Format

• Uses the PASCAL VOC format.

Dataset Structure

/VOCdevkit
└── Urinary Sediment Dataset
    ├── Annotations
    ├── ImageSets
    │   └── Main
    │       ├── test.txt
    │       ├── train.txt
    │       └── val.txt
    └── JPEGImages

Dataset Split

• Training set: 4,256 images
• Validation set: 852 images
• Test set: 268 images

Citation Information

• If you use this dataset, please cite:
  • Liang, Yixiong, et al. "Object detection based on deep learning for urine sediment examination." Biocybernetics and Biomedical Engineering 38.3 (2018): 661-670.
  • Liang, Yixiong, et al. "An End-to-End System for Automatic Urinary Particle Recognition with Convolutional Neural Network." Journal of Medical Systems 42.9 (2018): 165.
  • Yan, Meng, et al. "A Bidirectional Context Propagation Network for Urine Sediment Particle Detection in Microscopic Images." ICASSP 2020 – 2020 IEEE International Conference on Acoustics, Speech and Signal Processing.