Thrilled to announce the start of the Shifts Challenge 2022 on uncertainty estimation under distributional shift! Medical imaging is one of the tracks with segmentation of white matter lesions in brain MRI scans. Have look and join our efforts to create a vibrant community 🔥🔥🔥
Plastic injection molding is a technology for manufacturing single-use medical devices. Continuous control during a manufacturing batch helps reduce potential scrap. In task 5, you aim to build a mapping between manufacturing parameters and product quality. Mentor:Elias Rüfenacht
On 3rd/4th September it is again time for some hacking! Join us for the third edition of the MICCAI Hackathon. We have prepared 5 exciting tasks. Visit https://t.co/kDJHNdklXD to register and find out more!
#MICCAI#MICCAIHackathon@MiccaiStudents@MICCAI_Society
The black-box nature of deep learning models inhibits their translation into the clinical environment. In task 4, you perform an associative analysis of features extracted by neural networks and semantic features annotated by clinicians. Mentors: Francisco, Tânia, Hélder.
Images are not the only sources of clinical information available. Combining complementary diagnostic data might yield more robust predictive models. In task 3, you investigate integrated diagnostics for missing and longitudinal data. Mentors: Zuhir, @S_Trebeschi, @Sean1987
Together with @AndreyMalinin, and @MeriBach we are prepping up interesting challenges for you.
Checkout the hackathon repo: https://t.co/o6WeYCYfJO and https://t.co/HlpPwwcXuC. [3/3]
Join us on 🗓️03.09 for talks, discussion and technical support👩💻.
Even more exciting, this event is part of the longer term @ShiftsProject, on studying model robustness and uncertainty in real-world data with domain shifts.
[2/3]
Any plans for this weekend? Let me challenge you. Register to @miccaihackathon and try Task 1. We worked hard to gather data and create baselines that will let you study domain shifts among clinical annotators.
How much is model uncertainty influenced by annotator bias? [1/3]
Researchers often face limited training data when translating medical image computing methods to a new clinical application. In task 2, you aim at localizing target structures in head CT to facilitate manual annotation. Mentor: Daniel Erpenbeck
Annotators performing manual segmentation may follow slightly different annotation procedures, introducing bias in the learning process. In Task 1, you will investigate these domain shifts in white matter lesion segmentation. Mentors: @mormontre and @henmueller1
Congratulations to the winning team HIT for their contribution "Show Me Consistency! Increasing the consistency and quality of annotations", see https://t.co/D68mbHZJQB
The MICCAI Hackathon started, we are looking forward to exciting two days! Check out the welcome session if you are interested. Registration is still possible at https://t.co/Gr6wRvNpCg
https://t.co/gP4kVwX169
Above all, the aim is to have fun and exchange with others. But good effort deserves reward. Therefore, we have awards for the best contributions. We are looking forward to seeing you and are excited about your contribution to the MICCAI Hackathon.
Some spare time on Oct. 23/24? Join us for the MICCAI hackathon weekend and work on tasks related to ‘bridging the gap to the clinics’. Best contribution wins 500 USD. Visit https://t.co/3yXzGVeuP6 to register and for more info.
While working on your task, within a team or individually, you will get inputs and guidance from mentors: 1) @KlausKades, 2) @c_f_baumgartner, 3) @conjeti, 4) Miguel José Monteiro, 5) @mckinley_scan.
Did we get your attention? Join our kick-off event at https://t.co/EXA6ByG85l and look for more information at our website on how to register for the hackathon weekend https://t.co/Gr6wRvvOdG
The second keynote speaker will be Prof. Dr. Klaus Maier-Hein @maierhein, Head of Medical Imaging Computing at @mic_dkfz. He will discuss the development and deployment of the JIP (joint imaging platform) for federated data analytics – also known as the kaapana platform.