zeyugao/RCCNucSeg

该数据集设计用于透明细胞肾细胞癌（ccRCC）的细胞核分级。它包含1000张H&E染色的图像块，分辨率为512×512，共标注了70,945个细胞核。每个细胞核都有实例分割掩码和分类掩码的标注。原始全切片图像（WSIs）来自TCGA（癌症基因组图谱）数据库。数据集分为训练、测试和验证三个子集，分别包含700、200和100张图像块。每个样本包含两部分：ROIs（图像块）和标注。图像块保存为.png文件，标注保存为.mat文件。标注文件包含instance_map、class_map和TCGA_file_name三个键。数据集包含四类细胞核：内皮细胞核、1级肿瘤细胞核、2级肿瘤细胞核和3级肿瘤细胞核。数据集仅用于研究目的，禁止商业使用。

This dataset is designed for nuclei grading in clear cell renal cell carcinoma (ccRCC). It consists of 1000 H&E stained image patches with a resolution of 512 × 512, containing 70,945 labeled nuclei. Each nucleus is annotated with instance segmentation masks and classification masks. The original Whole Slide Images (WSIs) are sourced from the TCGA (The Cancer Genome Atlas) database. The dataset is divided into three subsets: training, testing, and validation, containing 700, 200, and 100 patches respectively. Each sample in this dataset comprises two parts: ROIs (image patches) and annotations. The image patches are saved as .png files, and the annotations are saved as .mat files. The annotation files contain three keys: instance_map, class_map, and TCGA_file_name. The dataset contains four classes of nuclei: endothelial nuclei, grade 1 tumor nuclei, grade 2 tumor nuclei, and grade 3 tumor nuclei. The dataset is for research purposes only, and commercial use is strictly prohibited.