bdzyubak / torch-control

A top-level repo for evaluating natively available models
MIT License
2 stars 0 forks source link

Set up nnUnet #1

Closed bdzyubak closed 8 months ago

bdzyubak commented 8 months ago

nnUnet is a state of the art image segmentation architecture that automatically configures hyperparameters to the GPU memory and dataset properties. I was able to obtain a major out-of-the-box organ segmentation accuracy compared to established Unet-based models at one of the companies I worked for. So, it makes sense to have nnUnet model available in the toolchain to compare to other state-of-the-art models within medical imaging, and in other fields.

Paper: Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2), 203-211.

bdzyubak commented 8 months ago

Linked pull request to achieve this is merged!

bdzyubak commented 8 months ago

nnUnet training curves (placing files here allows them to be linked in markdown docs): image image