openGTN workshop “Realistic MR image simulation for deep learning training”
The openGTN project investigates methods to simulate realistic magnetic resonance images (MRI) that can be used for optimizing, benchmarking and validating medical image analysis algorithms.
The project invites you to a one-day workshop on “Realistic magnetic resonance image simulation”, to be held on Friday 27 March 2020 in the “Zwarte Doos” at the Eindhoven University of Technology (TU/e), Eindhoven, The Netherlands, from 9:00 – 17:00. Attendance is free of charge, please register via below registration form (limited seating available!). Your registration will be confirmed by email.
Note: The number of registrations is already approaching the maximum capacity. Please register asap, you may be put on the waiting list.
About the workshop
At the workshop we will present and discuss how to simulate MR images with as much as possible similar appearances, anatomical variation and artefacts as scanned human MRI data. Special attention will be paid at the use of such data for optimizing deep-learning based image segmentation methods.
The three openGTN early-stage researchers (PhD researchers) will present the approaches used so far to generate simulated MRI data, and the use of these data to improve deep-learning based MRI segmentation. Thereafter four renowned, invited speakers will give their view on the subject, from different perspectives. The workshop will close with a discussion about how to make MRI simulations even more realistic.
|9:30||Introduction (prof. Marcel Breeuwer, TU/e)|
|9:45||MRI simulation approaches and segmentation optimization, with focus on the brain, spine & heart
(MSc Sina Amirrajab, Aymen Ayaz and Yasmina Al Khalil, TU/e)
|13:00||MRI simulation approaches: from full spin to analytical
(Prof. Tony Stoecker, Deutsches Zentrum Neurologische Erkrankungen, Bonn, Germany)
|13:50||Realistic MRI simulation: incorporation of motion, noise, artefacts, …
(Prof. Sebastian Kozerke, ETH Zuerich, Switserland)
|14:40||Short break (coffee/tea/soft drinks)|
|15:00||Image analysis algorithm optimization using simulated MRI data
(Prof. Daniel Rueckert, Imperial College London, United Kingdom)
|15:50||Medical image simulation using Generative Adversarial Networks
(Dr. Daniel Truhn, RWTH Aachen, Germany)
|16:20||Discussion on “Realism in MRI simulation”
Where are we, what to improve, how?