儿童白血病骨髓数据集
收藏资源简介:
本数据集由德国弗劳恩霍夫数字医学研究所MEVIS的研究团队创建,旨在为儿童白血病的诊断提供支持。数据集包含246名患有急性髓细胞性白血病(AML)、慢性髓细胞性白血病(CML)或急性淋巴细胞白血病(ALL)的儿童的骨髓涂片图像,以及这些儿童的诊断、临床和实验室信息。数据集包含超过40000个细胞的边界框注释,以及超过28000个细胞的高质量细粒度类别标签,这些标签是通过五位血液学专家的一致性方法获得的。这是首个集成细胞检测、细胞分类和诊断预测三个关键步骤的大型数据集。数据集的创建旨在促进人工智能辅助诊断的研究与发展,为更精确的诊断和改善患者预后做出贡献。
This dataset was developed by the research team from MEVIS, Fraunhofer Institute for Digital Medicine, Germany, to support the diagnosis of childhood leukemia. It comprises bone marrow smear images from 246 children diagnosed with acute myeloid leukemia (AML), chronic myeloid leukemia (CML), or acute lymphoblastic leukemia (ALL), alongside their corresponding diagnostic, clinical, and laboratory information. The dataset includes bounding box annotations for over 40,000 cells, as well as high-quality fine-grained category labels for more than 28,000 cells, which were obtained via a consensus approach involving five hematology experts. This is the first large-scale dataset that integrates three key steps: cell detection, cell classification, and diagnostic prediction. The development of this dataset aims to promote research and development of AI-assisted diagnosis, and contribute to more accurate diagnostics and improved patient prognosis.

- 1通过德国弗劳恩霍夫数字医学研究所MEVIS · 2025年



