Winter newsletter 2025

Published 30 Jan. 2025

🛠️ New features of Grand Challenge

Welcome to the Winter Edition of Grand Challenge Insights!

As winter's chill embraces us and the world is draped in a glistening layer of frost, we are thrilled to bring you the latest updates and achievements from our vibrant Grand Challenge community. This season, we celebrate remarkable progress, innovative breakthroughs, and inspiring stories that warm the heart despite the cold. Together, let’s reflect on our journey and continue striving towards excellence as we welcome the opportunities that this season of renewal and perseverance brings!

This image was generated with the assistance of DALL-E.

Simplified Challenge Pricing Policy!

We’ve introduced a new pricing model to make budgeting for challenges easier and more transparent. This updated policy splits costs into an up-front invoice for minimal compute and storage and a post-challenge invoice for additional usage, capped at a maximum you set. Stay in control of your budget while enjoying a stress-free planning process!

More Powerful Compute Instances!

Challenge organizers can now request NVIDIA A10G GPUs (g5 instances), offering 24GB of GPU memory—a significant increase from the previous 16GB on NVIDIA T4s. This enables participants to develop and test more resource-intensive algorithms while enjoying faster runtimes overall.

This image was generated with the assistance of DALL-E.

First Steps Toward Genomics!

Grand Challenge now supports the Newick and BIOM file formats, key to genomics research. These formats enable the use of graph-theoretical trees and sparse contingency tables as inputs and outputs, with validation for data quality. Viewing tools for these formats will follow soon!

As we step into the serene beauty and frosty embrace of winter, we’re inspired by the resilience and creativity blooming within our community, much like the warmth of an igloo amidst the snow. Let’s make this winter a season of innovation and connection, as enduring and brilliant as the winter landscape around us!

This image was generated with the assistance of DALL-E.

💡 Blogposts¶

"December 2024 Cycle Report"¶

Read all about the platform improvements from our RSE team in December, including new pricing policies and more powerful compute instances.

🔦 Highlighted algorithms¶

Tibia bone and implant segmentation

This model uses 3D CT scans of knee implants to segment the tibial TKA component and the tibial cortex bone. Try out this algorithm with your data!

Automated stroke lesion segmentation

DeepISLES is a deep learning-based algorithm for automated stroke lesion segmentation. It leverages an ensembling strategy, combining leading algorithms from the ISLES'22 challenge to enhance robustness and accuracy. The algorithm generates binary lesion masks and visualizations to support advanced stroke research. Try out this algorithm with your data!

🚀 Upcoming and running challenges¶

Grand Challenge currently offers two types of submissions: prediction submission and algorithm container submission. The algorithm container submission type has the advantage of producing reproducible algorithms that remain accessible to the research community long after the challenge has ended. This allows for continued use and exploration of the algorithms by the community. Therefore, it should be noted that we are phasing out the prediction submission procedure in favour of the algorithm submission procedure to ensure that challenges always produce reproducible algorithms. To learn more about hosting a challenge on our platform, go to our documentation!

🥅 Goal: The goal of the Fuse My Cells challenge is to advance new methods for 3D image-to-image fusion using deep learning in the fields of biology and microscopy.
✍️ Register: Registration is now open!
⏰ Deadline: Accepting AI algorithm submissions to the testing leaderboard until the 28th of February.
🏆 Prize:
The organizers of the Fuse My Cells challenge invite the winner to be a co-author an academic paper and invite the winner to publish their methods in the proceedings of the IEEE International Symposium on Biomedical Imaging 2025s.

🥅 Goal: The goal of the SynthRAD2025 challenge is to provide the first platform offering public data and evaluation metrics to compare the latest developments in synthetic CT generation methods. It exists of two tasks
Task 1: MRI-to-sCT generation to facilitate MRI-only and MRI-based adaptive radiotherapy.
Task 2: CBCT-to-sCT generation to facilitate CBCT-based adaptive radiotherapy.
✍️ Register: Registration is opens March 1st, 2025!
⏰ Deadline: Accepting AI algorithm submissions to the testing leaderboard until the 15th of August.
🏆 Prize:
Task 1: MRI-to-CT
1. €875,-
2. €550,-
3. €325,-
Task 2: CBCT-to-CT
1. €875,-
2. €550,-
3. €325,-

🏆 Leaderboard of finished challenges¶

We would like to extend our heartfelt congratulations to all the winners of the MICCAI 2024 challenges! A huge shoutout as well to all participants for their dedication and innovative solutions that continue to push the boundaries of AI development in healthcare. Your contributions are driving progress and shaping the future of medical AI—well done! 🎉