🔦 General Rules¶

To participate, you have to agree to the following rules:

• As we can’t allow access to the test images during the competition, the final model and code must be submitted in a docker container. You will be given a template docker container to use for this.

• Participants are requested to publish a (brief) description of their method and results on ArXiv (or similar pre-print platforms) that is linked to their final submission. There is no page limit for that description, but it has to include a conclusive description of the approach of the participating team.

• Training data is licensed as CC BY, i.e. everyone (also non-participants of the challenge) are free to use the training data set in their respective work, given attribution in the publication.

• In award category 1,  

  • participants are allowed to use all publicly available data for training and select any network architecture.
  • participants are allowed to combine as many models as needed to solve the task.  

• For participation in award category 2 (data-centric challenge), participants are not allowed to use any additional (publicly available) data, and the reference network implementation must be used. However, it is permitted to use AI models in the data pipeline under the following conditions:

  • the AI models ONLY alter/augment the data or synthetically generate new data but they don’t generate the prediction;
  • the AI models can be pretrained but have to be as such publicly available, but it is NOT allowed to train them on any ADDITIONAL data (only the challenge dataset is allowed for training).

• Algorithms must be made publicly available as e.g. GitHub repository with a permissive open source license (Apache Licence 2.0, MIT Licence, GNU GPLv3, GNU AGPLv3, Mozilla Public Licence 2.0, Boost Software Licence 1.0, The Unlicence)

• To avoid potential conflicts of interest, researchers affiliated with the institutes of the organizers are not allowed to participate.

• The provided baselines will hold their respective positions but without eligibility for prize money.

• Resubmissions of the provided baseline models in award category 1 are not eligible for prizes.

• Rules for inference on test set:

  • During inference, no metadata is provided, i.e., no information on whether a sample belongs to FDG or PSMA will be given.
  • Please be aware that the inference time is limited to 5 minutes per case.

• For the final test set submission, teams/participants are allowed to submit two models in total. The category in which the models fall is up to you. This means you can submit two models in category 1, two models in category 2, or one model in category 1 and one model in category 2. Each user can make two submissions. Since we can't prevent teams from submitting 2*n submissions, where n refers to the number of team members, we ask teams to coordinate their submissions and only submit two models. If teams submit more than the allowed number of models, only the two worst submissions for the associated team will be counted. This ensures a fair balance between teams and individual participants.

Publication policy

• Participants may publish papers including their official performance on the challenge data set, given proper reference of the challenge and the dataset. There is no embargo time in that regard. Please use the following references:Database:  , Challenge:
• We aim to publish a summary of the challenge in a peer-reviewed journal (further information). The first and last author of the submitted arxiv paper will qualify as authors in the summary paper. Participating teams are free to publish their own results in a separate publication after coordination with the authors of the challenge to avoid significant overlap with the summary paper.