Skip to content

Hausdorff distance loss support #6993

Closed

Description

@imadtoubal

Hausdorff distance is widely used in evaluating medical image segmentation methods. Adding an objective/loss function directly to optimize this distance can be instrumental in optimizing this score [1].

Hausdorff Distance

An existing implementation is publicly available on GitHub by Patryk Rygiel, although it might take some engineering work to integrate within the MONAI framework.

I was able to train a 3D segmentation model successfully using MONAI's SwinUNETR with this implementation and I would like to contribute the code for this loss so that it's more widely available and easier to integrate with MONAI-based repositories.

References:
[1] Karimi, D., & Salcudean, S. E. (2019). Reducing the Hausdorff distance in medical image segmentation with convolutional neural networks. IEEE Transactions on medical imaging, 39(2), 499-513.

Activity

  1. added a commit that references this issue on Sep 19, 2023
    e5f9337
  2. imadtoubal commented on Oct 11, 2023

    @imadtoubal
    ContributorAuthor

    Looks like I have forgotten to include the parameter alpha in the constructor method docstring. Do you recommend using this issue to make a PR to fix that?

  3. wyli commented on Oct 11, 2023

    @wyli
    Contributor

    sure, thanks, please feel free to create PRs, if that's done in the next 10 hours we can still include it in v1.3 otherwise it'll be released in the future versions

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions