Contrastive self-supervised learning from 100 million medical images with optional supervision

Florin C. Ghesu, Bogdan Georgescu, Awais Mansoor, Youngjin Yoo, Dominik Neumann, Pragneshkumar Patel, R. S. Vishwanath, James M. Balter, Yue Cao, Saša Grbić, Dorin ComaniciuJournal of Medical Imaging 2022 · DOI

Abstract

The proposed approach enables large gains in accuracy and robustness on challenging image assessment problems. The improvement is significant compared with other state-of-the-art approaches trained on medical or vision images (e.g., ImageNet).
BibTeX
@article{ghesu2022contrastive,
  author = {Florin C. Ghesu and Bogdan Georgescu and Awais Mansoor and Youngjin Yoo and Dominik Neumann and Pragneshkumar Patel and R. S. Vishwanath and James M. Balter and Yue Cao and Saša Grbić and Dorin Comaniciu},
  title = {{Contrastive self-supervised learning from 100 million medical images with optional supervision}},
  journal = {Journal of Medical Imaging},
  year = {2022},
  doi = {10.1117/1.jmi.9.6.064503},
}

Loading related content...

Figure