🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Image synthesis could revolutionize clinical practice. Hadrien Reynaud developed a method to generate video sequences from single images in echocardiograms. 🚀🏥 Better by 38 points! Department AIBE (FAU Erlangen-Nürnberg) BioMedIA Imperial Borderless research group #MICCAI2023 arxiv.org/abs/2303.12644 poster W-06-037

Addressing motion in abdominal MRI, especially in IBD patients, we propose a deep adversarial super-resolution approach that corrects motion artefacts without needing detailed intestine knowledge or paired data. 🏥🖼️ Department AIBE (FAU Erlangen-Nürnberg) BioMedIA Imperial Borderless research group #MICCAI2023 W-06-062

I'm very happy about the outstanding area chair award! Unfortunately, I couldn't be there to receive it in person. #MICCAI2023

Please join us today either online or offline for the discussion with distinguished speakers and oral presenters! Great thanks to our lead organizers: Yuan Xue Chao Chen

Athira Jacob presenting her work on deep conditional shape models for cardiac segmentation at the STACOM workshop at #MICCAI2023

Bye #MICCAI2023

Leo Feiner presenting his work on uncertainty propagation in medical imaging pipelines at the UNSURE workshop at #MICCAI2023

Self-supervised single-class strategies with histogram-equalised gradient upscaling are showing promise. Johanna P Mueller ‘s approach surpasses existing benchmarks. 🧠🔍 #MICCAI23 UNSURE arxiv.org/abs/2303.13227 Department AIBE (FAU Erlangen-Nürnberg) BioMedIA Imperial Borderless research group #MICCAI2023

Johannes C. Paetzold speaking on topology aware segmentation in the topology driven image analysis tutorial at #MICCAI2023

Histopathology is tough to analyse. Sarah developed a diffusion model approach to boost datasets with realistic examples, paving the way for better ML training in histopathology. 🔬🧠 Arxiv.org/abs/2304.09534 Department AIBE (FAU Erlangen-Nürnberg) BioMedIA Imperial Borderless research group #MICCAI2023

Matthew Baugh presenting our #MICCAI223 MOOD challenge submission

Very successful #MICCAI MOOD challenge medicalood.dkfz.de/web/ with our Team SubmitSomething winning all categories 🎉🎉 Department AIBE (FAU Erlangen-Nürnberg) BioMedIA Imperial Borderless research group #MICCAI2023 x.com/bernhardkainz1… x.com/bernhardkainz1…

For the November edition of the RSIP Vision computer vision magazine, I interviewed Robbie Holland from BioMedIA Imperial about his work on temporal biomarker detection for age-related macular degeneration in OCT scans: rsipvision.com/ComputerVision…

Fantastic work about interpretability constraints in large multi-modal language-vision #MachineLearning models by Mischa Dombrowski presented AAAI this week (oral).

Very nice idea to illustrate what we are doing Borderless research group during a talk. Thanks a lot to Jens Kleesiek and the excellent etim.uk-essen.de event for this!

❤️ New Job ❤️ An exciting opportunity to join our team BioMedIA Imperial Imperial DSI Imperial Department of Brain Sciences funded by British Heart Foundation New Horizons Grant. Develop machine learning and 3D shape analysis techniques for population-level cardiac imaging data. Advert: imperial.ac.uk/jobs/search-jo…

💨 We explored Geoffrey Hinton's forward-forward ⏩⏩ learning concept and discovered some new methods to deliver rapid and resource-efficient adaptable #MachineLearning for numerous tasks in medical image analysis. More to come... ⏩⏩

This work wouldn’t have been possible without the guidance of my supervisor Bernhard Kainz and the contributions of all my co-authors, with a special thanks to Qingjie Meng and Mischa Dombrowski ! @AI4HealthCentre Ultromics BioMedIA Imperial

Fully synthetic datasets are becoming a reality. We show how to make them in Hadrien Reynaud et al.'s latest #MICCAI24 work. Such datasets promise to solve significant data-sharing challenges in medicine by ensuring complete patient privacy, as it contains no data from real patients

🔎 New method in #MICCAI24 for distilling image data, analysing latent graphs, reducing the need for extensive datasets in histopathology. This allows for the creation of a small set of synthetic images that encapsulate the most critical information from large datasets.