Adversarial Networks: From Architecture to Practical Training

Level: Intermediate Length: 4 hours Format: In-Person Lecture Intended Audience: This intermediate-level course assumes basic knowledge in deep learning on the level of the Basic Course, SC1235 [i]Introduction to Medical Image Analysis using Convolutional Neural Networks[/i]. We also assume basic programming skills in Python, as we will show code examples that participant will obtain for later review and self-learning. Description: This half-day deep dive course will guide researchers with some background knowledge, e.g. from the introductory course, SC1235 Introduction to Medical Image Analysis using Convolutional Neural Networks, through the most important concepts of generative adversarial networks (GANs) and show example applications to medical data. GANs are powerful appearance models, but GANs can also be used to map between different domains (such as between CT and MRI) or to help training better segmentation models. Adversarial training can be introduced into several learning tasks in medical image analysis. It has been shown to help make image analysis algorithms more robust to variability in the data and to reduce the probability of failure on unseen cases. GANs in their initial implementation have been known to be hard to configure and train, but recent advances have helped them catch ground in applications of classification and segmentation. We will introduce GANs conceptually and from a Variational Inference perspective, give an overview of their development towards the state of the art, and explain specific architectural decisions and developments that have been proposed to stabilize their training. We will show code examples and illustrate the course content with live demonstrations on example data, so that the participants gain some first-hand experience on the subject. The course is not designed as a hands-on workshop, though. Learning Outcomes: This course will enable you to: - explain adversarial training in general - understand the basic problem statement of GANs - identify potential applications of GANs in medical image analysis - summarize how to implement at least one specific GAN architecture - describe typical problems in the training and how to mitigate them Instructor(s): Markus Thorsten Wenzel works on machine learning methods for medical applications since 2005 and has published more than 30 conference and journal papers on the subject. He received his PhD for his work on decision support systems for breast care. At Fraunhofer MEVIS, he is a senior scientist for cognitive medical computing. He is a funded member of the Fraunhofer Society research class "Cognitive Machines" and is experienced in teaching and lecturing for academia and industry. He has acquired and led several international research projects. Florian Weiler has been working on the field of medical image processing since 2006. His research interests focus on both image-analysis and computer-visualization of medical imaging data, with a special interest in clinical applicability. He received his PhD in the field of neuro-image analysis, but has previously also been active in the context of liver-surgery planning and software support for radiation therapy. At Fraunhofer MEVIS, he currently works in the decision-support team, with a focus on deep-learning based methods for predicting therapy response in neurologic diseases. Event: SPIE Medical Imaging 2020 Course Held: 16 February 2020