Lizard
收藏资源简介:
Lizard数据集是由华威大学计算机科学系和多家医疗机构合作创建的大型核实例分割与分类数据集,专注于结肠组织。该数据集包含近50万标记的核,涵盖多种细胞类型,如上皮细胞、结缔组织细胞、淋巴细胞等。数据集通过多阶段注释流程生成,包括自动和半自动预测的精细调整,以及病理学家的参与,确保注释的准确性和可靠性。Lizard数据集旨在推动计算病理学中基于细胞的模型的开发,特别是在结肠肿瘤微环境的分析和预测中。
The Lizard Dataset is a large-scale nuclear instance segmentation and classification dataset focused on colon tissue, developed through collaboration between the Department of Computer Science at the University of Warwick and multiple medical institutions. This dataset contains nearly 500,000 annotated nuclei, covering a variety of cell types including epithelial cells, connective tissue cells, lymphocytes, and others. The dataset is generated via a multi-stage annotation pipeline, which includes fine-tuning of automated and semi-automated predictions as well as involvement of pathologists, to ensure the accuracy and reliability of the annotations. The Lizard Dataset aims to advance the development of cell-based models in computational pathology, particularly in the analysis and prediction of the colon tumor microenvironment.

- 1Lizard: A Large-Scale Dataset for Colonic Nuclear Instance Segmentation and Classification华威大学计算机科学系 · 2021年



