AugDiff

AugDiff: Diffusion based Feature Augmentation for Multiple Instance Learning in Whole Slide Image

Installation

Please use the instructions in latent diffusion to set up your environment. If you're missing any packages, you can find them in my environment (requirements.txt).

Image augmentation, feature extraction

please refer to Augmentation/README.md

Diffusion training

please refer to LatentDiffusion/README.md

MIL training and test

please refer to MIL/README.md

Acknowlegement

Thanks for the opensource project at latent diffusion and Augmentation-PyTorch-Transforms.

Citation