The paper: "Dual Contrastive Learning with Anatomical Auxiliary Supervision for Few-shot Medical Image Segmentation"
More contents will be added later.
CUDA/CUDNN
torch >= 1.4.0
torchvision
numpy
- CHAOS-T2: A abdominal MRI dataset from CHAOS - Combined (CT-MR) Healthy Abdominal Organ Segmentation
- MAALC: A abdominal CT dataset from Multi-Atlas Labeling Beyond the Cranial Vault - Workshop and Challenge
Part of the code about the local prototype is referenced from SSL-ALPNet.
The contrastive loss is based on info-NCE
TO DO