|
Title: Machine Learning for the Health Domain: Challenges and Solutions
Mark Hoogendoorn (VU University Amsterdam, Netherlands)
Date and Time: 9:10-10:00, October 23, 2020
Zoom Conference Room One
Abstract:
Over the last decade, machine learning techniques have attracted a lot of attention. While impressive applications are seen in a variety of domains, the health domain lags behind. Examples where machine learning driven tools really make it to the doctor or patient are rare, while ample data is often available. Different causes can be identified that can explain this lack of success. In order to take away these barriers, improved machine learning algorithms are needed that fit the challenges in the health domain better. In this talk I will discuss the machine learning challenges in the health domain in more detail, and will show a variety of techniques that we have developed over the years to tackle some of those challenges. These include techniques for prediction of health states as well as personalization of treatments. The example projects are mainly based on structured data. I will end the talk with an overview of several (in my opinion) promising directions for future developments.
Bio:
Mark Hoogendoorn is a Full Professor of Artificial Intelligence at the Vrije Universiteit Amsterdam (VU) where he chairs the Quantitative Data Analytics group. He obtained his PhD degree at the VU in 2007, was a Postdoctoral Researcher at the University of Minnesota and started as an Assistant Professor at the VU right thereafter. In 2015 he was a Visiting Scientist at the Massachusetts Institute for Technology within the Clinical Decision Making Group. In his research, he focuses on predictive modeling and personalization using machine learning techniques, predominantly applied in the health domain. He participated in, and has led, a large number of (both EU- and nationally funded) research projects related to AI and health and has been the joint coordinator of the FP7 ICT4Depression project. He is currently a board member of the International Society of Research on Internet Interventions (ISRII), and chairs the special interest group on Data Sharing and Standard in the same organization. Furthermore, he is part of the board of Amsterdam Medical Data Science and the VU Campus Center for AI & Health and is an Associate Editor of the Internet Interventions journal.
HIS2020 Workshop Keynote
Title: Introduction into conversational agents for mental health
Willem-Paul Brinkman (Technical University of Delft, Netherlands)
Date and Time: 9:10-10:00, October 23, 2020
Zoom Conference Room One
Abstract:
The WHO expects mental disorders to represent 15% of the global burden of disease by 2020. Advancement in technology creates an opportunity to address the need for mental health care. In this talk, I will specifically look at how we can use conversational agents in this context. I will discuss how we can use them to support individuals during an intervention or help them with monitoring but also to collect self-reported data. During the presentation, I will discuss examples from my current and past research. They include (embodied) conversational agents for social anxiety and PTSD treatment; offering support for insomnia therapy; and data collection to assess core believes.
Bio:
Willem-Paul Brinkman (PhD) is an associate professor at Delft University of Technology, The Netherlands. His primary research interests are human-computer interaction, behaviour change support systems, specifically eHealth systems including virtual reality therapy systems, and conversational agents. He is fascinated by eHealth systems that include conversational agents that offer psychological support. His ultimate objective is to establish an autonomous eHealth system with a digital psychologist that can assist individuals in achieving a broad set of behaviour change goals ranging from overcoming mental illness to lifestyle modification for coping with a chronic disease. He is, therefore, determined to build these systems and establish an empirically grounded understanding of them. For this, he works on several research grants that focus on these type of eHealth systems. They include systems for the treatment of patients with social anxiety, posttraumatic stress disorder patients, insomnia, and depression.
|
