Local contrastive loss with pseudo-label based self-training for semi-supervised medical image segmentation

Open in new window