Roeterseiland Campus, Roetersstraat, Amsterdam, Netherlands

ADS Deep Dive: Deep Learning in Medical Imaging

ADS Deep Dive: Deep Learning in Medical Imaging

Date: Tuesday 09 January 2018

Time: 14:00-16:00

Location: UvA Roeterseiland Campus (Roetersstraat)- Building A – Room A1.02 (map here)

The ADS Deep Dive sessions are an opportunity to delve into a data science related topic in more depth, specifically on 09 January we will highlight cutting-edge research on Deep Learning in Medical Imaging.


Introduction & Chair: Max Welling, Research Chair in Machine Learning, Informatics Institute UvA 

We will have 3 speakers, each presenting for 30 minutes plus 10 minutes discussion time:

14:00-14:40: Jörn-Henrik Jacobsen, Informatics Institute UvA on “Inverting Deep Networks – The Why, How and Implications”

Inverting deep neural networks has been proposed as a potential tool to investigate the information discarded with depth. One way to obtain such inverses is to make use of priors that regularize the inversion. However, such priors are typically based on simple heuristics, or come in the form of inflexible deep networks themselves. 
I will discuss how to invert arbitrary deep networks with simple learned priors that are useful for other downstream tasks as well. Further, I will introduce a class of deep networks that are invertible by construction and highlight the implications of such architectures.

14:40-15:20: Matthan Caan, Brain Imaging Center, Academic Medical Center (AMC) on “Deep Learning for accelerated MRI reconstruction”

High resolution Magnetic Resonance Imaging (MRI) of the human brain is a timely procedure. In order to accelerate MRI, k-space can be incoherently undersampled beyond the Nyquist-criterion. Compressed Sensing (CS) reconstructs images using a predefined transform, such as the Wavelet transform. Deep learning applies multi-layered neural networks as universal function approximators and is able to find its own compression implicitly. This allows to further accelerate image acquisition. We propose a Recurrent Inference Machine (RIM), which can acquire great network depth, while retaining a low number of parameters. We demonstrate its performance on accelerating MRI-scans of the human brain.

15:20-16:00: Joost Batenburg, Scientific Staff Member & Group Leader, Department of Computational Imaging CWI and Professor, Leiden University on “Real-Time 3D Tomography” 

Just two decades ago, taking photographs was an elaborate endeavor, involving not just taking a picture and viewing it, but also various time-consuming steps to process the analog film into a printed result. Now that everyone has digital cameras providing instant access to our pictures, we see that this has not just resulted in more convenience, but it has dramatically changed the way we using imaging in ways that no one imagined before. In comparison, CT scanning is still a time-consuming process, where reconstructions are usually made only after the scan has been finished, and analyzed even later at a different location.

In this lecture I will discuss the various challenges and opportunities involved in speeding up the tomography pipeline towards real-time tomography, with a specific focus on how machine learning techniques can play a role in making the complete process automated, accurate, and computationally efficient.  

16:00: Coffee, snack & networking


It is free to attend but please register in advance.

Please only register if you plan to come as space is limited. 


Amsterdam Data Science (ADS) accelerates data science research by connecting, sharing and showcasing world-class technology, expertise and talent from Amsterdam on a regional, national and international level. Our research enables business and society to better gather, store, analyse and present data in order to gain valuable insights and make informed decisions.


Find out more about ADS at