---
license: mit
library_name: pytorch
tags:
- Medical Vision Language Pre-Training
- Foundation Model
---
# UrFound Checkpoint Model Card

UrFound: Towards Universal Retinal Foundation Models via Knowledge-Guided Masked Modeling

## Model Details
- **Paper**: https://arxiv.org/abs/2408.05618
- **Framework**: https://github.com/yukkai/UrFound

## Datasets
- **Pretrain dataset**: [FLAIR](https://github.com/jusiro/FLAIR) for more details
<div align="center">
    <img src="urfound_p2.jpg" style="border: 1px solid black; margin: 10px; margin: 10px; width: 50%; height: auto;">
</div>

- **Finetune dataset**: [RETFound](https://github.com/rmaphoh/RETFound_MAE) for more details
<div align="center">
    <img src="urfound_p3.jpg" style="border: 1px solid black; margin: 10px; margin: 10px; width: 50%; height: auto;">
</div>

## Citation

```bibtex
@inproceedings{yu2024urfound,
  title={UrFound: Towards Universal Retinal Foundation Models via Knowledge-Guided Masked Modeling},
  author={Yu, Kai and Zhou, Yang and Bai, Yang and Soh, Zhi Da and Xu, Xinxing and Goh, Rick Siow Mong and Cheng, Ching-Yu and Liu, Yong},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={753--762},
  year={2024},
  organization={Springer}
}
```