Deep Learning Nijmegen Meetup - December 2018

We would like to invite you for the second Deep Learning Nijmegen Meetup, an initiative of the Nijmegen Data Science Centre. There will be two inspiring talks followed by a networking event with drinks and snacks at the Cultuurcafé.

Our first speaker will be Dr. Geert Litjens, Assistant Professor in Computational Pathology at the Department of Pathology, Radboudumc. He studied Biomedical Engineering at the Eindhoven University of Technology and subsequently completed his PhD on computer-aided detection of prostate cancer in multi-parametric MRI at the Radboudumc. He spent 2015 as a postdoctoral researcher at the National Center for Tumor Diseases in Heidelberg, Germany on an Alexander von Humboldt Society Postdoctoral Fellowship. In 2016, he obtained a Young Investigator Award from the Dutch Cancer Society with which he returned to Nijmegen. Geert’s research focusses on computational pathology for improved efficiency and accuracy in oncological diagnostics.

Our second talk will be given by Dr. Mohsen Ghafoorian, Research and Development engineer working on developing machine learning based techniques for autonomous driving map generation at TomTom. Prior to joining TomTom, he obtained a Ph.D. in machine learning for medical image analysis at Radboud University and a Master's degree in artificial intelligence with a thesis on hierarchical reinforcement learning, at Sharif University of Technology, Tehran, Iran. His current research topics are adversarial training and weakly supervised methods.

Schedule:

15.30h: Dr. Geert Litjens talk: "Hurray, the robots are coming: applications of deep learning in medical imaging".
Abstract: In dermatology, pathology, and ophthalmology super-human results have been achieved in image-based diagnostics thanks to deep learning. In this talk, I will discuss these advances and provide some context: how special are these results? Should doctors fear for their jobs? Furthermore, I will highlight some key changes and challenges in applying deep learning algorithms to medical imaging. Last, I will show some of our own work in this field and discuss the implications for the future.

16.15h: Dr. Mohsen Ghafoorian talk: "AI for Mapmaking: Embedding Loss Generative Adversarial Networks for Lane Detection".
Abstract: Generating accurate high definition maps through a scalable process is an important milestone for realizing reliable self-driving cars. This presentation is about our machine learning approach to understanding and semantically modeling the road environment for the purpose of high-definition mapping. In particular, we will go into details on our recent embedding loss driven adversarial training approach, which we use to efficiently impose desired structural consistencies on our semantic dense prediction of road dividers.

17.00h: Networking event with snacks and drinks at the Cultuurcafé, Radboud University (Mercatorpad 1, 6525 HS Nijmegen).

The location of the talks is room CC4 in the Collegezalencomplex, Radboud University (Mercatorpad 1, 6525 HS Nijmegen), see https://goo.gl/maps/fWGT8P2xUpJ2