Rules

All participants must form teams (even if the team is composed of a single participant), and each participant can only be a member of a single team. 

No additional kidney pathology data is allowed.

No additional glomeruli segmentation annotations are allowed in training and testing.

Pre-trained models are permissible for use as long as they have not been trained with kidney pathology data. This includes widely used datasets like ImageNet or using SAM pre-trained weights.

Generative models are allowed provided that they have not been trained using any kidney pathology data. Generative models trained on other types of data, such as natural images, general radiology data, or non-kidney pathology data, are acceptable.

This challenge only supports the submission of fully automated methods in Docker containers. It is not possible to submit semi-automated or interactive methods. The prompts (e.g., points, boxes, scribbles) are not allowed in the testing phase.

The participant teams will be encouraged to release their training and testing code and explain how they fine-tuned their hyper-parameters. Note that the code can be shared with the organizers only as a way to verify validity, and if needed, NDAs can be signed.

The 3 best-performing methods of both tasks will receive invitations to present papers at the MICCAI 2024 MOVI workshop proceedings. Additionally, the top 3 methods will be requested to provide their code and model to the KPIs organizer, ensuring adherence to reproducible research standards.

More Q&A can be found on the Discussion panel.