Jerry-Master/lung-tumour-study
收藏Hugging Face2024-03-28 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/Jerry-Master/lung-tumour-study
下载链接
链接失效反馈官方服务:
资源简介:
该数据集用于结合图神经网络和计算机视觉方法对肺组织中的细胞核进行分类。数据集包含来自9名不同患者的85张1024x1024像素的H&E染色全切片图像,主要分为两类:肿瘤细胞(2)和非肿瘤细胞(1)。由于问题的复杂性,153个细胞被标记为不确定,这些细胞在训练和验证集中被移除,测试集中不包含不确定细胞。数据集总共包含21255个训练集细胞、4114个验证集细胞和5533个测试集细胞。数据集的结构包括原始图像、带有细胞覆盖的图像、原始GeoJSON文件以及训练、测试和验证集的文件夹。此外,还提供了示例预测和裁剪后的图像。
This dataset is intended for classifying cell nuclei in lung tissue by combining graph neural networks and computer vision approaches. It contains 85 1024×1024 pixel H&E-stained whole-slide images from 9 distinct patients, which are primarily classified into two categories: tumor cells (label 2) and non-tumor cells (label 1). Owing to the complexity of the task, 153 cells were marked as uncertain, which were excluded from both the training and validation sets, and no uncertain cells are included in the test set. In total, the dataset consists of 21255 training set cells, 4114 validation set cells, and 5533 test set cells. The dataset structure includes original images, cell-overlaid images, raw GeoJSON files, as well as folders for the training, test, and validation sets. Additionally, sample predictions and cropped images are provided.
提供机构:
Jerry-Master
原始信息汇总
数据集概述
数据集描述
该数据集包含85张1024x1024像素的H&E染色全切片图像(WSIs)的补丁,来自9名不同患者。主要分为两类:肿瘤性(2)和非肿瘤性(1)。由于问题的复杂性,有153个细胞被标记为不确定。在训练和验证集中剔除了这些不确定细胞,并在测试集中仔细选择,确保不包含不确定细胞。总共有21255个细胞在训练集,4114个在验证集,5533个在测试集。确保了患者数据在不同分组中没有重复,避免了数据泄露。
数据结构
数据以多种方式提供:
orig文件夹:包含未标注的图像。overlay文件夹:包含带有细胞覆盖的相同图像,红色表示健康细胞,绿色表示肿瘤细胞,用于可视化目的。raw_geojson文件夹:包含从QuPath软件提取的原始geojson文件,可能包含重复和不确定细胞。train、test、validation文件夹:包含准备使用的数据,结构与tumourkit包文档中指定的一致。
其他信息
pred和hov文件夹:包含我们模型的预测示例,若要训练自己的模型,应删除这些文件夹。npy文件夹:包含原始图像的518x518大小的裁剪,可以使用Tumourkit库提供的代码修改训练形状。
引用
@article{PerezCano2024, author = {Jose Pérez-Cano and Irene Sansano Valero and David Anglada-Rotger and Oscar Pina and Philippe Salembier and Ferran Marques}, title = {Combining graph neural networks and computer vision methods for cell nuclei classification in lung tissue}, journal = {Heliyon}, year = {2024}, volume = {10}, number = {7}, doi = {10.1016/j.heliyon.2024.e28463}, }



