AGGC(自动格里森分级挑战赛)数据集是一个旨在开发前列腺癌自动分级算法的竞赛数据集。该数据集提供了由经验丰富的病理学家标注的前列腺切除和活检标本的H&E染色全切片图像(WSI)数据。数据集中的每个病例包含五个类别的二值掩膜:格里森模式3、格里森模式4、格里森模式5、正常组织和间质组织。WSI图像包含105例训练数据和45例测试数据,而活检图像包含37例训练数据和16例测试数据。
腹部维度 | 2D |
模态 | pathology |
任务类型 | segmentation |
解剖结构 | 前列腺癌 |
解剖区域 | 前列腺 |
类别数 | 5 |
数据量 | 203 |
文件格式 | .tif |
AGGC22
│
├── AGGC2022_test
│ ├── Subset1_Test_annotation
│ ├── Subset1_Test_image
│ ├── Subset2_Test_annotation
│ ├── Subset2_Test_image
│ ├── Subset3_Test_annotation
│ └── Subset3_Test_image
│
└── AGGC2022_train
├── Subset1_Train_annotation
│ ├── Subset1_Train_1
│ │ ├── G3_Mask.tif
│ │ ├── G4_Mask.tif
│ │ ├── Normal_Mask.tif
│ │ └── Stroma_Mask.tif
│ ├── Subset1_Train_2
│ │ └── ...
│ └── Subset1_Train_3
│ └── ...
├── Subset1_Train_image
├── Subset2_Train_annotation
├── Subset2_Train_image
├── Subset3_Train_annotation
└── Subset3_Train_image
统计类型 | 间距 (mm) | 尺寸 |
---|---|---|
最小值 | - |
7920*4800 |
中位值 | - |
9360*11520 |
最大值 | - |
19440*21120 |
@article{huo2022comprehensive,
title={Comprehensive AI model development for Gleason grading: From scanning, cloud-based annotation to pathologist-AI interaction},
author={Huo, Xinmi and KokHaur, ONG and Lau, Kah Weng and Gole, Laurent and Young, David and Tan, Char Loo and Zhang, Chongchong and Zhang, Yonghui and Zhu, Xiaohui and Li, Longjie and others},
year={2022}
}