News¶

🥳 SEG.A. challenge - Thank You for Being with Us 🎁

Dear Participants of the SEG.A. Challenge,

Hello there!

We're thrilled by the fantastic response to this challenge and are over the moon to announce our three incredible winners. A huge shout-out to each and every one of you for your fantastic work and dedication throughout this journey. We hope you had as much fun as we did - your participation means the world to us!

We must acknowledge that managing this challenge presented some difficulties and apologize for any inconvenience caused. We appreciate your patience. We are all learning together, and our goal was to ensure that things went as smoothly as possible.

We are thrilled to share that the upcoming MICCAI conference in Vancouver later this year is highly anticipated. It's going to be an amazing opportunity to learn and exchange ideas. We're looking forward to sharing what we've learned from this challenge and gaining insights from others. Let's get ready for a great knowledge-sharing experience!

It is important also to acknowledge the remarkable community that we have cultivated together. Your energy and passion have driven our journey, and we're truly grateful for your contributions. Together, we've accomplished remarkable feats that will have a lasting impact on the scientific world.

With heartfelt thanks and lots of anticipation,

Antonio Pepe, Gian Marco Melito, Jan Egger

📢 Exciting News: Refinement in Evaluation Metrics!

Dear Participants,

We hope you're just as excited as we are about the SEG.A. challenge. We want to share some updates regarding the metrics we'll be working with in Phase 3 - P3, P4, and PJ. While these changes might come as a surprise, we're here to assure you that they're designed to enhance the competition and make it even more engaging!

We understand that change can sometimes be unexpected, but please rest assured that these modifications won't affect the optimization process you're embarking on for Phase 2. In fact, these adjustments will provide a fresh perspective on performance assessment.

The revised metrics are now based on a weighted ranking formula. This formula considers essential statistical factors: the mode, variance, and skewness coefficients. These adjustments aim to provide a more comprehensive perspective of your solutions. This will allow you to demonstrate your skills from various angles. Although we understand that these adjustments may cause some inconvenience, we are very eager about the possibilities they offer.

Our utmost priority is to guarantee a just and fulfilling experience for all users, and these modifications align with that objective. For more details on these adjustments and how they'll be implemented, please refer to the updated ranking methods page.

We appreciate your support and commitment to this competition. Our goal is to provide an exceptional experience for all participants, and we are grateful for your cooperation. We look forward to seeing the innovative solutions you create and the positive impact they will have. Thank you for being a part of this fantastic journey.

🚨 PHASE 2 submission 🚨

We were able to fix the issues. We changed the weights in the formulations of P3, P4, and PJ to fix the disproportionate calculation of the metrics. The new weights are described on the ranking method page.

However, the phase 2 will be ranked based on the MEDIAN and STANDARD DEVIATION of the Hausdorff distance and Dice Score. The P3, P4 and PJ metrics will be used for phase 3 prior to further investigation.

We thank you for your understanding. Keep up the good work :)

PHASE 2 submission: TEMPORARY CLOSED

We are sorry to inform you that due to technical problems found in the evaluation metrics of Phase 2, we are temporarily closing (just some hours) the submission to this phase. The updated results of the leaderboard to today (August 2nd, 2023) are the following:

• 坤坤kk: 1,1364
• IWM: 0,5941
• tpvagenas: 0,4768

• IWM: 1082,13
• tpvagenas: 376,15
• 坤坤kk: 306,984

The leaderboard will be updated as soon as possible. The results above are considered as valid. We thank you for your understanding. Keep up the good work :)

🗳️ New Hausdorff distance metric 🗳️

Given some issues recorded on the Hassdorf distance score, we changed the evaluation container. The new Haussdorf distance considered for ranking all phases is considered the 95th percentile of its values. 💡

If you have already submitted it, we encourage you to submit it again to check its changes and to update your score on this metric for phase 1 - container testing. Also, we modified to 6 the max number of submissions for phase 1.

📣 Time limit exceed problem

UPDATE (07/07/2023): the new time limit is 10 minutes.

Due to several issues with the submissions regarding the time limit, we are working with the grand-challenge system to find a solution that suits everybody. At the moment, the time limit is set to 5 minutes. Stay tuned for any updates.

🎉 Exciting News: Introducing the Challenge Prize! 🎉

We are thrilled to announce that the winners of our eagerly awaited challenge will be rewarded with a monetary prize! In this way, we can recognize and celebrate the participants' remarkable achievements and talents.

The prize has been made possible through the generous support of our sponsor, XYZ company. Their commitment to fostering innovation and supporting aspiring individuals leads this challenge to a resounding success. We are deeply grateful for their sponsorship, which has allowed us to offer this prestigious monetary prize. Their support has made it possible for us to reward and recognize the outstanding achievements of our challenge participants.

We understand that you are eager to know more about the prize breakdown. Our team currently works to ensure a fair and equitable distribution among the winners.

Stay tuned for further announcements as we unveil comprehensive information about the prize distribution process. 🏆✨

Challenge Polls opening on May 15, 2023

How is your preparation going? Let's share it! 🤩

To create a preview of the challenge outlook, we welcome the communication of temporary results (Dice Score and Hausdorff distance) obtained with a 5-fold cross-validation.

There is no limit to the number of updates. Send us a nickname (or team name) and your current performance via ✉️ e-mail at sega.grandchallenge@gmail.com or write a post on the forum 💬.

☝️ These results are not final! Only the results on the hidden test set will be used for the actual ranking.