MedVision Logo MedVision: Benchmarking Quantitative Medical Image Analysis

Yongcheng Yao1 , Yongshuo Zong1 , Raman Dutt1 , Yongxin Yang3 ,
Sotirios A Tsaftaris2 , Timothy Hospedales1
1School of Informatics, University of Edinburgh
2School of Engineering, University of Edinburgh
3Queen Mary University of London
arXiv GitHub Docs Data Models Demo Project Data Explorer

📝 Blog

Stories behind the MedVision project: from dataset releases to leaderboard updates.

  1. MedVision v1.4.0: 50× More Tumor/Lesion Measurements and Their Annotation Recall Dataset v1.4.0 selects tumor and lesion measurements by physical size in millimetres instead of pixel count, growing published measurements from 75,840 to 3,801,540, and quantifies annotation recall by cluster length and by cluster area. Read post

Acknowledgements

This work was supported by UK Research and Innovation (grant EP/S02431X/1) through the UKRI Centre for Doctoral Training in Biomedical AI at the School of Informatics, University of Edinburgh. We also acknowledge the support of the Edinburgh International Data Facility (EIDF) and the Data-Driven Innovation Programme at the University of Edinburgh.

BibTeX

@misc{yao2026medvisionbenchmarkingquantitativemedical,
    title={MedVision: Benchmarking Quantitative Medical Image Analysis}, 
    author={Yongcheng Yao and Yongshuo Zong and Raman Dutt and Yongxin Yang and Sotirios A Tsaftaris and Timothy Hospedales},
    year={2026},
    eprint={2511.18676},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2511.18676}, 
}