OpenMind
收藏资源简介:
OpenMind数据集是一个大规模的头颈3D MRI图像数据集,包含114k个MRI图像。该数据集旨在提供大量三维医学成像数据,以加速用于三维医学成像的自监督学习方法的发展。数据来源于OpenNeuro平台的800个数据集,包含来自30多台不同扫描仪的23种不同的MRI模态/技术,是一个高度多样化的预训练数据集。此外,数据集还提供了分层元数据、匿名化掩码和解剖学掩码,以及按照修改后的BIDS格式组织的结构化数据。
The OpenMind dataset is a large-scale 3D head-and-neck MRI image dataset containing 114k MRI images. It is designed to provide abundant 3D medical imaging data to accelerate the development of self-supervised learning methods for 3D medical imaging. The data is sourced from 800 datasets hosted on the OpenNeuro platform, covering 23 distinct MRI modalities and techniques from more than 30 different scanners, rendering it a highly diverse pre-training dataset. Furthermore, the dataset provides hierarchical metadata, anonymization masks, anatomical masks, as well as structured data organized in the modified BIDS format.




