Deep Learning-based algorithms showed great performance in many fields of image procession and pattern recognition and compete with technologies such as compressive sensing and iterative optimization. The basis for the success of these algorithms is the availability of large amounts of data (big data) for training and of high computing power (typically GPUs). In this seminar we try to answer the question how typical problems in medical imaging might profit from deep learning concepts. In particular, we will aim at the data annotation problem. We have recently started a non-profit organization that is accepting medical data donations ( http://www.medicaldatadonors.org ). The donations also allow annotation via gamification over the Internet. Aim of this project seminar will be to develop a game for data annotation. In order to do so, we will look into the topics of deep learning, game design, programming in unity, and related topics.
Dates & Rooms:
Monday, 8:15 - 9:45; Room: KH 1.021