Important Updates: Co-Authorship Criteria & Docker Containers for Test Submission Phase

Dear Challenge Participants,

We hope your model development is progressing well. We have some important updates to share:

Co-Authorship Opportunity: If you are interested in being a co-author on the challenge conclusion paper, please note that you must make your submission to the model development phase before August 15th to be considered. Additionally, to qualify for co-authorship, your submission needs to achieve an average Dice score above 0.7064, which is the performance of the baseline U-Net model.

Test Submission Phase - Docker Containers: We are currently working on the Docker containers that will be used for the test submission phase of the challenge. More details will follow soon, but please rest assured that we are working hard to ensure a smooth and efficient process.

If you have any questions or need further clarification, feel free to reach out. Good luck with your submissions, and thank you for your continued hard work!

Best regards, AortaSeg24 Team

Does our model already have to achieve a average dice score of >0.7064 in the develement phase? We assumed that we would have time until the test phase ends.

Hello,

Also, no chance for a little bit of deadline extension, I guess?

Adrian

Hello Dommi,

Yes, that is correct! Participants looking to co-author on the challenge paper, must beat this score.

Thanks, AortaSeg Team

Hello Agaldran,

We have chosen to extend the deadline for the development phase submissions till 8/23 at 11:59PM EST, as mentioned in our latest announcement.

This will be the final extension we make for this phase.

Thanks, AortaSeg Team

Hello Alexander,

Just to mention that what Dommi asked is a meaningful question. In our case, we are still training and will need several more days to finish. We can already submit something to the development leaderboard, and make sure our container works as expected, but we'll be replacing our current weights by better ones next week. Therefore, we won't be beating the baseline score on the development phase but we could very well beat it in the final phase.

Wouldn't it make more sense to include in the challenge publication those teams who beat the baseline in the test set rather than in the validation set? More so given that the test set is a more realistic assessment of performance?

Cheers,

Adrian