We accepted the challenge from UCSF to perform knee segmentations on accelerated raw MRI data (from k-space to segmentation). Accelerating MRI is essential to reduce patient discomfort in the scanner and decrease healthcare costs. We submitted a fully automated end-to-end pipeline for generating high-resolution multi-class knee segmentations from undersampled k-space. We showed that a state-of-the-art deep learning model could accelerate knee MRI 8x for task-based segmentation and are happy to finish third in this international K2S challenge of MICCAI 2022!
By joining forces between the UMCG (Groningen, The Netherlands), RadboudUmc (Nijmegen, The Netherlands), University of Twente (Enschede, The Netherlands) and Siemens Healthineers (Erlangen, Germany), the acceleration of MRI with AI will be investigated. The research focuses on new Artificial Intelligence (AI) techniques for faster and more accurate MRI for diagnostic and interventional purposes in prostate cancer.
A collaboration between Derya Yakar (UMCG), Thomas Kwee (UMCG), Henkjan Huisman (RadboudUmc), Jurgen Futterer (RadboudUmc/UT), Frank Simonis (UT), Jelmer Wolterink (UT) and Wouter Nijhof (Siemens Healthineers) receives a grant for a project called FastMRI. The goal of the project is to develop new artificial intelligence (AI) techniques that will make MRI for the diagnosis and interventions in prostate cancer targeted, and therefore faster and more accurate. The FastMri collaboration project is co-funded by the PPP Allowance made available by Health~Holland, Top Sector Life Sciences & Health, to stimulate public-private partnerships and by Siemens Healthineers.