This dataset is the largest 3D medical image segmentation dataset currently built by the Shanghai Jiao Tong University team in 2024. It brings together 72 public datasets, 22K+ images from three modalities of CT, MR and PET, 302K+ segmentation annotations, covering 497 segmentation targets in 8 major parts of the human body, and realizing a universal medical segmentation model for radiological images through text prompts. Related research is based onOne Model to Rule them All: Towards Universal Segmentation for Medical Images with Text Prompts" was published on the academic platform arXiv.org.

Citation

@article{zhao2025large, title={Large-vocabulary segmentation for medical images with text prompts}, author={Zhao, Ziheng and Zhang, Yao and Wu, Chaoyi and Zhang, Xiaoman and Zhou, Xiao and Zhang, Ya and Wang, Yanfeng and Xie, Weidi} journal={NPJ Digital Medicine}, volume={8}, number={1}, pages={566}, year={2025}, publisher={Nature Publishing Group UK London} }