Summer newsletter 2024
Published 5 July 2024
🛠️ New features of Grand Challenge
Welcome to the Summer Edition of Grand Challenge Insights!
As the sun shines brighter and days grow longer, we are excited to bring you the latest updates and highlights from our community. This season, we're celebrating remarkable achievements, innovative projects, and inspiring stories from participants around the world. Let's make this summer a season of progress and collaboration!
Play through time!
Firstly, participants can now automatically 'play' through the time dimension of 4D images; much like a video clip! The controls include play direction and playback speed and are automatically enabled for all viewing of 4D images in CIRRUSImproved overlay viewing parameters
Additionally, it is now possible to set overlay parameters for each item of the segmentations individually, changing e.g. alpha values, look-up tables etc.
Each Challenge Phase Paves the Way for the Next!
Up until now challenge phases were always independent from one another, meaning that submissions to one phase did not affect submissions to another. This cycle, we added the option to link two or more phases of a challenge together. With this option, you could for example enforce that participants need to have successfully submitted to the preliminary development phase of your challenge before being able to submit to the final test phase.Unveiling Cutting-Edge Upgrades for Pathology Image Viewers
Furthermore, we extended the pathology image viewer with support for point annotations, 2D bounding boxes, (interpolated) spline and line annotations and improved polygon annotations.
Streamlined Container Updates: Separate Uploads for Algorithm Models and Ground Truths
Previously, algorithm models and ground truths had to be included in their respective container images, requiring a new container for every update, even for minor changes. Now, algorithm models and evaluation ground truths can be uploaded and updated separately from their containers, streamlining the process.
As we dive into the warmth and vibrancy of summer, we're excited to see the blossoming of new ideas and collaborations within our community. Let's make this summer as bright as the sunniest day!
💡 Blogposts¶
"April 2024 Cycle Report"¶
Read all about the platform improvements from our RSE team in April, including new cirrus features, enabling 4D cine play and initial slice selection configurations.
"May 2024 Cycle Report"¶
In the May cycle, the RSE team worked on offering the dependency linkage between two or more phases of a challenge. Until now, challenge phases were always independent, but now you could force participants to first successfully submit to preliminary development phases before submitting to the final test phase. Read more about it in this blog post!
"June 2024 Cycle Report"¶
Improved user insight into the operational status of our platform, new feature additions for annotations, and smoothening of algorithm submissions were the main topics of interest for the RSE team in the June cycle of this year. This blog post provides more details on the progress and more!
🔦 Highlighted algorithms¶
DataScientX (N. Debs, A. Routier, et al.; France) algorithm trained on PI-CAI: Private and Public Training Dataset
This algorithm represents the submission from the DataScientX team (N. Debs, A. Routier, et al.; France) to the PI-CAI challenge. This algorithm will perform two tasks: localize and classify each lesion with clinically significant prostate cancer (if any) using a 0–100% likelihood score and classify the overall case using a 0–100% likelihood score for clinically significant prostate cancer diagnosis. Try out this algorithm with your data!
Fully automatic PDAC detection on CECT - PANORAMA challenge baseline
This is the baseline algorithm for the PANORAMA challenge. This algorithm produces a likelihood score for the presence of pancreatic ductal adenocarcinoma (PDAC) in an input venous-phase contrast-enhanced computed tomography scan (CECT), together with a detection map with the lesion localization. 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 LEOPARD challenge is to yield top-performance deep learning solutions to predict the time to biochemical recurrence from H&E-stained histopathological tissue sections.
✍️ Register: Registration is now open!
⏰ Deadline: Accepting AI algorithm submissions to the testing leaderboard until the 1st of August.
🏆 Prize:
1. €2.000,-
2. €1500,-
3. €1000,-
🥅 Goal: The goal of the ACOUSTIC-AI challenge is to develop and benchmark AI models for the automated measurement of fetal abdominal circumference on this specific data type, to broaden the accessibility of prenatal care in areas with limited resources.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 31th of July.
🏆 Prizes:
Teams whose algorithms rank in the top three of the Final Leaderboard (Final Test Phase) will be invited to co-author the challenge paper, with a maximum of three members per team.
🥅 Goal: The PENGWIN segmentation challenge is designed to advance the development of automated pelvic fracture segmentation techniques in both 3D CT scans (Task 1) and 2D X-ray images (Task 2), aiming to enhance their accuracy and robustness.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 16th of August.
🏆 Prizes:
1. $1.000,-
2. $500,-
3. $200,-
🥅 Goal: The CURVAS challenge goal is focused, mainly, on evaluating how to deal with multi-rater annotations and how to exploit them.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close in July.
🏆 Prizes:
1. $1.000,-
2. $500,-
3. $200,-
🥅 Goal: COSAS 2024 contains two tasks for participating teams to contribute their findings: Cross-Organ Adenocarcinoma Segmentation, and Cross-Scanner Adenocarcinoma Segmentation in digital pathology slides.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on the 31st of August.
🥅 Goal: Participants of the AortaSeg24 challenge will have the opportunity to develop innovative algorithms for accurate, automated, and multi-class segmentation of this intricate vascular structure.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 31th of August.
🏆 Prizes:
The top five teams, based on their algorithm’s performance, will be invited for an oral presentation. Participants will have the opportunity to discuss their approaches, challenges, and results. The top five teams will be announced.
🥅 Goal: The TOOTHFAIRY2 challenge aims to push the development of deep learning frameworks to segment anatomical structures in CBCTs.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 8th of August.
🏆 Prizes:
1. $1.500,-
2. $1.000,-
3. $500,-
🥅 Goal: The SurgVU24 challenge consists of two goals; the first goal is to train a model to localize surgical tools and their corresponding key points in videos, using tool presence data as weak labels, the second goal is to segment videos into different surgical steps being performed.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 10th of September.
🏆 Prizes for first goal:
1. $4.000,-
2. $2.000,-
3. $1.000,-
Best methodology report receives $500,-
🏆 Prizes for second goal:
1. $3.000,-
2. $2.000,-
3. $1.000,-
Best methodology report receives $500,-
🥅 Goal: In the DENOISING challenge, we want to focus on an unsupervised denoising task, which is particularly challenging and key for the bioimage community.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 31st of July.
🥅 Goal: The ISLES24 challenge aims to segment the final stroke infarct from pre-interventional acute stroke data.
✍️ Register: Registration is now open!
⏰ Deadline: The submissions close on 15th of August.
🏆 Leaderboard of finished challenges¶
The task of the DREAMING grand challenge is to implement inpainting-based DR methods in oral and maxillofacial surgery. Algorithms had to fill regions of interest concealed by disruptive objects with a plausible background, such as the patient's face and surroundings. The winner of the Final Test Phase was Ryosuke Seshimo. Congratulations! Top challenge participants will be invited as co-authors on an upcoming high-impact journal paper summarizing the findings of this challenge.
The task of the Light My Cells grand challenge is to predict the best-focused output images of several fluorescently labelled organelles from label-free transmitted light input images. The winner of the Final Test Phase was Trang Le. Congratulations! Top challenge participants will be invited to an upcoming high-impact journal paper summarizing the findings of this challenge.
