Talk abstracts and speaker biographies
Industry Trends, Technology and Ethics: an Overview
Malte Kremer, Strategy& “#FutureOfHealth – a Strategist‘s Perspective”
The #FutureOfHealth is people-driven, preventative, personalized, datarized and integrated into daily live – 96% of PharmaExecutives agree, hardly any doubts left. This will a) present massive challenges to the entire HC ecosystem (e.g. HCPs to redefine their role and education, regulators to patent health data solutions, payer to account for prevention) and b) lead to tectonical shifts of $3.4tr – $4.4tr across existing and new value pools by 2030. Gross budget per patient will decline by 6.5% to 27.5% (particularly in care and drugs & appliances) – Pharma needs to react by improving efficiency and / or capture share via differentiation and / or tapping into new value pools. The biggest opportunity will arise for those players who best converge traditional and digital health capabilities – who will be the smartest?
According to Pharma, Tech will be the frontrunner already ramping up their healthcare activities (e.g. Amazon + JPM + Berkshire, Apple Watch with 400k people ECG study) but chances exist for Pharma if they substantially disrupt themselves. While Tech will need to develop selective medical capabilities to succeed, Pharma will need to focus on building data & analytics capabilities, securing access to data and relentlessly focus on people-centric, preventative health offerings – while facing, among others, cultural, structural and regulatory hurdles.
Dr. Kremer’s short bio:
– Born 1983 in Germany, married, one kid, lives in Berlin
– BSc Molecular Biotechnology – TU Munich and University of Oxford (Visiting Scientist)
– MSc Neuroscience – LMU Munich
– PhD Biochemistry and Neuroscience – LMU Munich and International Max-Planck-Research-School Munich
– Since 2014 – various Roles in Healthcare Consulting, currently Senior Manager at Strategy&, the strategy consultancy within PWC
– Passionate Sailor & Kitesurfer and loves Photography & Contemporary Design
Sebastian Schaal, Luminovo “Human-AI Hybrids – Best of both Worlds”
The media is full of news about the dawn of artificial intelligence and how it will take all our jobs. In this talk, Sebastian will give his perspective why AI is more about automating tasks than killing jobs. He will motivate that humans and machines are good in very different things and how we can use this to our advantage, building human-AI hybrids. Using AI to augment human intelligence is the key driver of his company luminovo.ai and in his opinion the way to think about AI in healthcare.
Sebastian Schaal is Founder at Luminovo, where he is responsible for Sales and Product. Luminovo focuses on B2B deep learning projects and building tools to automate repetitive workflows as classifying images or documents. He graduated top of his class with an M.Sc. in Electrical and Computer Engineering from TU Munich and holds an Honours Degree from the CDTM. He obtained his second M.Sc. from Stanford University where he focused on Management Science and Machine Learning. He worked for Intel and researched on NILM algorithms for the energy sector before transitioning to Deep Learning, working for the deep tech startups NavVis and Magazino on Computer Vision problems. Besides his technical experiences, Sebastian also worked as a consultant at McKinsey, focusing on IT and data strategy.
Yu Wang, Leibniz Supercomputing Center “Current Trends in AI Computing”
The current success of AI largely depends on the convergence of big data and high performance computing systems. The availability of big data, combined with HPC systems with AI accelerators (GPU, TPU, NNP, etc), makes training of insanely large deep neural networks possible. These bigger networks in general have superior performance in various fields, including visual, audio application, and to certain extend, cognitive applications, e.g. AlphaZero. In this talk, I will give an overview about current trends in AI computing, including automatic machine learning (AutoML), network architecture search (NAS), parallel and distributed deep learning. I will conclude my talk with a demo on why model size matters in the modern AI.
Dr. Yu Wang studied Artificial Intelligence and Neural Networks at Katholieke Universiteit Leuven in Belgium. He later moved to Munich to join MIPS, institutes for bioinformatics genomics and bioinformatics from Technical University Munich in 2011. He is currently a senior AI scientist at Leibniz Supercomputing Centre and a Nvidia University Ambassador, working on automated AI computing on HPC systems.
Big Data and Diagnostics
Nicolas Brieu, Definiens “Deep learning for automated PD-L1 scoring in lung cancer”
PD-L1 expression measured by immunohistochemistry helps identify Non-Small-Cell-Lung-Cancer (NSCLC) patients that may respond to anti-PD-1/PD-L1 immunotherapies. We will introduce how the use of state-of-the-art computer vision algorithms based on deep fully supervised and deep semi-supervised learning enables the automated estimation of PD-L1 Tumor Cell (TC) scoring. In a comprehensive study, we quantify the concordance between the automated scores and scoring by pathologists and show the predictive value of the automated score.
Nicolas Brieu is Principal Research Scientist at Definiens. He has more than 10 years’ experience in the field of medical image analysis. His main area of research is in machine and deep learning for computer vision and data mining. He joined Definiens in 2011 and has since been working on the research and development of novel algorithms to automate the analysis of digital pathology images and the identification of immune-related prognostic factors.
Precision Medicine
Jan Baumbach, TUM “Computational Systems Medicine”
One major obstacle in current medicine and drug development is inherent in the way we define and approach diseases. We discuss the diagnostic and prognostic value of (multi-)omics panels. We have a closer look at breast cancer survival and treatment outcome, as case example, using gene expression panels – and we will discuss the current “best practice” in the light of critical statistical considerations. In addition, we introduce computational approaches for network-based medicine. We discuss novel developments in graph-based machine learning using examples ranging from Huntington’s disease mechanisms via Alzheimer’s drug target discovery back to where we started, i.e. breast cancer treatment optimization – but now from a systems medicine point of view. We conclude that multi-scale network medicine and modern artificial intelligence open new avenues to shape future medicine. We will also have a short glimpse on novel approaches for privacy-aware sensitive medical data sharing. We quickly introduce the concept of federated machine learning and blockchain-based consent management to build a Medical AI Store ensuring privacy by design and architecture.
Jan Baumbach studied computer science at Bielefeld University in Germany. His research career started at Rothamsted Research in Harpenden (UK) where he worked on computational methods for the integration of molecular biology data. He returned to the Center for Biotechnology in Bielefeld for his PhD studies where he developed CoryneRegNet. Afterwards, at the University of California at Berkeley, he worked in the Algorithms group of Richard Karp on Transitivity Clustering, a novel clustering framework for large-scale biomedical data sets. From March 2010, Jan was head of the Computational Systems Biology group at the Max Planck Institute for Informatics in Saarbrücken, Germany. In October 2012, he moved to the University of Southern Denmark as head of the Computational BioMedicine group. His research concentrated on systems and network biomedicine. He was study program coordinator of the Computational BioMedicine program from 2015 to 2017. In January 2018 he moved to the Technical University of Munich as chair of the Experimental Bioinformatics. In Munich, he develops computational methods for systems medicine and novel federated AI approaches ensuring privacy by design.
Medical Imaging
Claudio von Schacky, UCSF “Artificial Intelligence (AI) in Radiology: Where does AI in Radiology stand and what does AI mean for Medical Imaging?”
Recent progress in artificial Intelligence (AI) provide promising approaches for a variety of applications in radiology. First, this presentation will seek to unravel existing AI techniques and their implications for radiology. Then, it will take a closer look at a research project involving an artificial intelligence that assesses x-rays of the hip as a radiologist would. Finally, this presentation will offer an outlook on future developments of AI in radiology.
Claudio von Schacky, MD, is a radiologist-in-training at Technische Universität München (TUM) and a Postdoctoral Research Scholar in the Musculoskeletal and Quantitative Imaging Research (MQIR) group at the Department of Radiology and Biomedical Imaging at the University of California (UC), San Francisco. He is part of the Big Data in Radiology (BDRad) and the Center for Intelligent Imaging research network, both corporate efforts of UC San Francisco and UC Berkeley. Claudio von Schacky holds a double degree in Medicine and Electrical Engineering and Information Technology from TUM and completed a PhD program in molecular imaging including a research stay at Stanford University. Current research projects revolve around developing new clinically applicable machine learning approaches for image analysis in musculoskeletal radiology using state-of-the-art machine learning libraries such as PyTorch and TensorFlow.
AI Driving Management and Commercial Changes in Pharma
Michael Streit, Novartis “Change Management in Pharma: Why should we see AI as Augmented Intelligence”
Novartis, like many others, is also investing earlier to help seed, accelerate and scale promising digital health technologies. For example, we invested in Aktana, which provides AI-enabled insights to help companies deliver information to physicians, based on their needs. We’re partnering with Aktana to help us create a dynamic and simple ‘personal assistant’ for our sales representatives so their 100,000 daily interactions with doctors are more personalized and meaningful.
Michael has developed a data-driven mindset with over ten years of experience in the digital world, entrepreneurship, technology and moving organizations, SMEs and big enterprises.
More than ten years ago my time in the digital world began. After getting to know the e-commerce world, while still being a student, he founded a startup for web development and Facebook ads as a co-founder. After his graduation and two stays abroad in Mexico and Taiwan, his entrance to Germany’s most famous price comparison website was due. Here Michael got the chance to purse his startup spirit and founded an in-house spin-off for coupon marketing. During that time he did my Master’s degree in Berlin as well. After all that it was time for a change. So he joined healthcare and have been in love since then. Over time, however, Michael’s priorities have changed, away from digital strategy and online marketing, to data strategies and analytics projects.