Evaluating semi-supervision methods for medical image segmentation: applications in cardiac magnetic resonance imaging

Sarah Hooper, Sen Wu, Rhodri Davies, Anish Bhuva, Erik B. Schelbert, James Moon, Peter Kellman, Hui Xue, Curtis P. Langlotz, Christopher RéJournal of Medical Imaging 2023 · DOI

Abstract

We evaluate semi-supervision for medical image segmentation using heterogeneous datasets and clinical metrics. As methods for training models with little labeled data become more common, knowledge about how they perform on clinical tasks, how they fail, and how they perform with different amounts of labeled data is useful to model developers and users.
BibTeX
@article{hooper2023evaluating,
  author = {Sarah Hooper and Sen Wu and Rhodri Davies and Anish Bhuva and Erik B. Schelbert and James Moon and Peter Kellman and Hui Xue and Curtis P. Langlotz and Christopher Ré},
  title = {{Evaluating semi-supervision methods for medical image segmentation: applications in cardiac magnetic resonance imaging}},
  journal = {Journal of Medical Imaging},
  year = {2023},
  doi = {10.1117/1.jmi.10.2.024007},
}

Loading related content...

Figure