We look forward to your innovative contributions to this exciting area of research! This challenge is jointly organized by @JuanEugenioIgl1, @RosenLab, Bene Wiestler, @neuronflow, myself, and the @BraTS_challenge organizing team.
#MICCAIChallenge Synthetic 3D MRI images have enormous potential in medical imaging. Here's an example showing how synthetic MRI can effectively reduce motion artifacts in brain tumor imaging: the algorithm, trained on glioma patients, generalizes robustly to meningioma cases.
A step-by-step tutorial to get you started is available here:
📖: https://t.co/Xi2tzmdauM.
Our evaluation criteria prioritize not only overall image quality (e.g., SSIM), but also importantly assess the clinical utility of synthesized tumor structures in segmentation tasks.
More great news: Our collaborative work with @RolfsMicrobes, Tobi, @HongweiBran, @menze_group, @DQBM_uzh is now in @PLOSONE Lab Protocols! Discover our new automated pipeline for imaging bacterial infections in zebrafish 🐟 Grateful for this collaboration!
https://t.co/fPDi710MsR
@ja_schnabel gave a talk on “Resolving MR motion artefacts using deep learning” showcasing their works by @oksuzilkay@KingsImaging@SmartHeartUK and @hannah_eichhorn and @spieker_vj @HelmholtzMunich 👏. Check out their recent review on motion correction in MRI: https://t.co/ECCc0ojpXa
Also… neural implicit representation could be used for k-space interpolation (IPMI2023): https://t.co/v0Tl4KWXoj , and multi-view reconstruction (MICCAI 2023) https://t.co/QpjBDx2JHj
Yes, it seems that we gain more robustness in k-space interpolation when using masked auto-encoding by a vision transformer. @PeterPanJZ will tell you more!
🥃 After GRAPPA, RAKI, Caipirinha, SAKE etc., We have one more alcoholic beverage in MR Reconstruction😉
I will present our new work, k-GIN at #MICCAI2023 poster W-06-042. 🤗 You can also check our project page: https://t.co/NErZUvtB6b