CRC100K (100,000 histological images of human colorectal cancer and healthy tissue)

这是一组 100,000 个非重叠图像块，来自人类结肠直肠癌 (CRC) 和正常组织的苏木精和伊红 (H&E) 染色组织学图像。
所有图像均为 224x224 像素 (px)，每像素 0.5 微米 (MPP)。
用于组织分类；类别是：脂肪（ADI），背景（BACK），碎片（DEB），淋巴细胞（LYM），粘液（MUC），平滑肌（MUS），正常结肠粘膜（NORM），癌症相关基质（STR），结直肠腺癌上皮（TUM）。
这些图像是从 NCT 生物库（德国海德堡国家肿瘤疾病中心）和 UMM 病理档案（大学医学中心）的福尔马林固定石蜡包埋 (FFPE) 样本中手动提取的 N=86 H&E 染色的人类癌症组织切片曼海姆，曼海姆，德国）。组织样本包含 CRC 原发性肿瘤切片和来自 CRC 肝转移灶的肿瘤组织；胃切除标本中的非肿瘤区域增加了正常组织类别，以增加变异性。

This is a set of 100,000 non-overlapping image patches derived from hematoxylin and eosin (H&E)-stained histology images of human colorectal cancer (CRC) and normal tissues.
All images are 224×224 pixels (px) in size, with a spatial resolution of 0.5 micrometers per pixel (MPP).
This dataset is intended for tissue classification tasks, with the following categories: adipose (ADI), background (BACK), debris (DEB), lymphocytes (LYM), mucus (MUC), smooth muscle (MUS), normal colonic mucosa (NORM), cancer-associated stroma (STR), and colorectal adenocarcinoma epithelium (TUM).
These image patches were manually extracted from N=86 H&E-stained human cancer tissue slides prepared from formalin-fixed paraffin-embedded (FFPE) samples sourced from the NCT Biobank (National Center for Tumor Diseases, Heidelberg, Germany) and the UMM Pathology Archives (University Medical Center Mannheim, Mannheim, Germany).
The tissue samples include CRC primary tumor sections and tumor tissue from CRC liver metastases; non-tumor regions from gastrectomy specimens were added to the normal tissue category to increase dataset variability.

搜集汇总

背景与挑战

背景概述

CRC100K是一个包含10万个非重叠图像块的组织学图像数据集，专门针对人类结直肠癌和健康组织，图像尺寸为224x224像素，分辨率为每像素0.5微米。该数据集用于组织分类任务，涵盖9个类别，包括癌症相关组织和正常组织，数据来源于德国研究机构的FFPE样本，旨在支持癌症病理学研究。