HISTAI
收藏资源简介:
HISTAI数据集是一个开源、大规模的整张幻灯片图像数据集,由HistAI机构创建,旨在为计算病理学研究提供丰富的数据资源。数据集包含超过60,000张来自不同组织类型的幻灯片,每张幻灯片都伴随着详尽的临床元数据,包括诊断信息、患者人口统计学数据、详细的病理学注释和标准化的诊断编码。HISTAI数据集旨在填补现有资源中的空白,促进创新、可重复性和临床相关的计算病理学解决方案的发展。数据集可用于开发诊断模型、研究组织类型和临床背景的泛化、探索多模态病理学模型以及研究数字病理学中的域适应和迁移学习。
The HISTAI dataset is an open-source, large-scale whole-slide imaging dataset created by the HistAI institution, aiming to provide rich data resources for computational pathology research. The dataset contains over 60,000 slides sourced from diverse tissue types, each paired with comprehensive clinical metadata including diagnostic information, patient demographic data, detailed pathological annotations, and standardized diagnostic coding. The HISTAI dataset is designed to fill the gaps in existing resources and advance the development of innovative, reproducible, and clinically relevant computational pathology solutions. It can be used to develop diagnostic models, investigate the generalization of tissue types and clinical contexts, explore multimodal pathological models, and study domain adaptation and transfer learning in digital pathology.

- 1HISTAI: An Open-Source, Large-Scale Whole Slide Image Dataset for Computational PathologyHistAI · 2025年



