Abstract: How can we apply technology to improve healthcare? What is the best approach to deal with all digital data from healthcare that we obtain every day? What can we learn from that data? These and many more questions are inspiring the researchers and engineers worldwide.        Remarkable advancements are made in data analysis for healthcare, making medical diagnosis and interventions more objective, safe and efficient. Yet, despite the growing amounts of data and AI technologies, there are still challenges. In this keynote talk we will consider examples of them: content and context in data, explainability of AI outcomes, and future developments..

Bio:  Dr. Sveta Zinger received her MSc in computer science in 2000 from the Radiophysics faculty of the Dnepropetrovsk State University, Ukraine. In 2004, she received a PhD from the Ecole Nationale Superieure des Telecommunications Paris, France, for her thesis on interpolation and resampling of 3D data. Sveta held several postdoc positions in France and the Netherlands in the area of digital image understanding and content based image retrieval. Currently Sveta is an Associate Professor in the Eindhoven University of Technology. Her research concerns digital image analysis in healthcare. Sveta published 180 papers in her research domain. In 2020 she established a new Elsevier journal “Computer Methods and Programs in Biomedicine Update” in which she is currently an editor-in-chief.

Abstract: Data science is a booming business. Every year thousands of algorithms (if not more) are developed for all kinds of purposes, ranging from social media to grocery shopping. The same is happening in healthcare; data science is booming there as well in healthcare research. Here technologies such as artificial intelligence (AI) can be used for many purposes that can generally be classified in three aspects of healthcare: 1) prevention, 2) diagnosis, and 3) treatment. AI can aid in prevention by providing tailored advice to improve an individual’s lifestyle and thus reducing risks of disease. In diagnosis, AI can help by automatically analyzing medical images to identify disease. And in treatment AI help physicians offer patients personalized treatments or help them find the most suitable treatment for a specific disease. Despite these many opportunities, it remains a challenge to actually implement AI into the clinic. Even when patients would clearly benefit from the algorithm. In this presentation, I will describe scenarios in which AI could make a big difference in healthcare and will discuss with the audience which challenges we still face.  

Bio:  Dr. Rianne Fijten is an assistant professor in the Clinical Data Science department of Maastricht University. She defended her PhD thesis in 2017, which revolved around building AI models to diagnose disease based on biomarkers. In 2017 she moved into the healthcare field and has since been actively working at implementing clinically relevant models into the clinic. Her work includes developing clinically relevant AI models, developing decision aids and decision support systems for patients and doctors, implementing those in a clinical setting, and finally evaluating the effectiveness of these tools.

Abstract: In this talk, we will investigate cancer treatment as a game-theoretic contest between the physician’s therapy and the cancer cells’ resistance strategies. This game has two critical advantages for the physician: (1) Only the physician can play rationally. Cancer cells, like all evolving organisms, can only adapt to current conditions; they can neither anticipate nor evolve adaptations for treatments that the physician has not yet applied. (2) It has a distinctive Stackelberg (leader-follower) structure; the “leader” oncologist plays first and the “follower” cancer cells then respond and adapt to therapy.  We will learn how the physician can exploit their advantages in this game. This approach leads to evolutionary cancer therapies, i.e.  therapies that anticipate and steer treatment-induced resistance in cancer cells. We will illustrate how patient data can be utilized for that purpose and show evidence for superiority of the evolutionary therapy against standard of care from ongoing clinical trials.  

Bio: Katerina is since September 2021 an associate professor and Delft Technology Fellow at the Faculty of Technology, Policy and Management, where she is responsible for research in healthcare, a new emerging topic of the faculty.  She has a doctorate from the Delft Institute of Applied Mathematics in Stackelberg game theory and obtained postdoctoral experience at the French Institute for Research in Computer Science and Automation in system identification and at the Delft Center for Systems and Control in game-theory based optimal control. Before joining TU Delft, she was assistant and associate professor at Maastricht University, where she set up and led its Dynamic Game Theory team. Katerina leads a number of national and international projects, such as the European Training Network EvoGamePlus and the Netherlands Research Foundation (NWO) project “Understanding cancer through evolutionary game theory and dynamic systems theory”. She is recipient of the 2020 NWO Stairway to Impact award for designing novel cancer therapies and bringing them to clinical trials.

Abstract: Data driven innovation is increasingly a key success in bringing integral solutions in the healthcare sector. In the recent years, the abundance of data and in particular in the consumer sectors has led to a rapid development of machine learning techniques enabling the study of consumer behaviours for the sake of improved and personalized services. While the healthcare sector is enjoying rich data, extracting value to serve patients’ health is not straightforward. Above all, patient privacy and safety must be maintained and governed with regulatory processes entailing the use of data and extracted value to serve patients’ outcome with measured cost and burden on the healthcare system and professionals. Therefore, by taking the human central approach in our daily work, we seek meaningful innovation in data science enabling professionals to deliver the best to their patient at lower cost. Philips has a long history in developing successful healthcare solutions. In this presentation, I will follow the path of data-to-value focusing on a medical use case and demonstrating how a human centric technological solution is bringing value in healthcare.

Bio: Dr. Calina Ciuhu grew up in a small danubian town in Romania, and her passion for mathematics began in her childhood, enjoying solving problems and math puzzles, as well as reading about famous mathematicians and history of science. She studied physics and graduated in 1998 from Babes-Bolyai University Cluj-Napoca, Romania, with a thesis on supersymmetry and supergravity. In 2003 she received her PhD title in theoretical physics from Vrije Universiteit Amsterdam. In the same year she joined Philips Research and worked on various applications areas, such as picture quality, image analysis and pattern recognition in medical applications, and  developing new sensing technologies for patient monitoring, combining data analysis with methodology from theoretical physics. Her current interest is particularly in technology that enables a human centric interaction with medical data for healthcare professionals.

Abstract: The digitization of the healthcare system and particularly an artificially-intelligent health Made in Europe remains a difficult undertaking. Grounded in science and highly regulated by the government, health care is a very challenging sector with many stakeholders and hard to navigate. Young and innovative startups are the right mechanisms to bring new technology to the market, but they are confronted with many challenges that make it hard to exploit the big potential of AI. Zana, founded and led by Dr. Hoxha,  is one of those startups in the German KI Landscape working to drive digitalization forward in Health.

The talk presents the innovative Conversational AI technology of Zana for building chatbot and voice assistant systems that can communicate in natural language and accompany patients on their healing journey. A glimpse is offered to the commercial solutions and the various projects where the technology is deployed. The talk further addresses from the personal experience as founder the challenges of bringing innovation to a risk-averse market, mechanisms that help to overcome such difficulties, and the role Universities play to transform the high quality of European research into a competitive, commercializing approach.

Bio: Dr.-Ing. Julia Hoxha is founder and CEO of Zana Technologies GmbH, an AI healthtech startup offering innovative Conversational AI technology for voice and chatbot applications in healthcare. Before founding Zana, she completed her PhD at KIT and performed PostDoc work as scientific researcher at Columbia university the field of AI and medical informatics. She is now head of Working Group “Health” at the German AI Association.