Submission Deadline: June 30th, 2024 AOE

Notification of Acceptance: July 15th, 2024 AOE

Publication Deadline: July 30th, 2024 AOE

Workshop @ MuC 2024: September 1st, 2024

• Deadline: June 30th, 2024 AOE.
• Format: Position paper of 1 to 2 pages in the ACM two-column template (\documentclass[sigconf, review]{acmart} using the Overleaf template).
• Topics: Position papers should reflect the authors' viewpoint on topics related to AI and Health, propose specific concepts for future applications, or address related research questions. For detailed topic suggestions fitting the workshop, see below.
• Submissions should contain all author information.
• Please submit your position paper through ConfTool https://www.conftool.com/muc2024/  using the submission field "MCI-WS114: AI and Health: Using Digital Twins to Foster Healthy Behavior"
• All submissions will receive reviews from the workshop's organizers and other submitting authors.

At least one author of each accepted submission must attend the workshop either in person or remotely. All participants must register for the workshop and for at least one day of the conference. Six selected submissions will be presented in the form of lightning talks (5 mins each) with subsequent discussion (5 mins each). All other accepted submissions will be presented in the form of posters. Accepted submissions can be included in the workshop proceedings published via the GI Digital Library (https://dl.gi.de/).

For further questions please contact Jonas.

In this workshop we will solicit submissions that cover a broad range of topics related to the overall theme. In particular, we are interested in work that addresses any of the following topics:

• AI-Based Health Recommendations:  Exploring the efficacy and ethical considerations of utilizing artificial intelligence to provide real-time, personalized health recommendations in everyday contexts. This means looking closely at how we can use AI to give people personalized health advice in an ethical and helpful way. It involves issues such as data privacy, user consent, and the seamless integration of these systems into daily life to encourage positive health behavior while mitigating potential harm. Moreover, it is crucial to examine the social, psychological, and ethical implications of deploying such technologies.

• Smart Home Devices for Health:  Investigating the role of smart home devices and digital artifacts in communicating both healthy and unhealthy behaviors to users. We want to dive into smart health technologies at home and in our daily lives to look at how everyday devices can tell us about the healthiness of our behavior and help us make better choices on a daily basis. Digital artifacts may communicate progress toward daily physical activity goals or predict shifts in users' body composition or weight, based on dietary choices and physical activity levels over recent weeks or months.

• Feedback Mechanisms:  Assessing both explicit and implicit techniques for delivering health-related feedback and evaluating their respective impacts on user behavior and well-being. We will examine how different ways of giving health feedback can influence people's individual behavior and overall health. Explicit feedback mechanisms may involve notifications that offer healthy activity recommendations for daily life (e.g., disembarking the train one stop early to complete the journey on foot) or dietary choices (e.g., presenting the consequences of food selections at cafeterias or grocery stores during the decision-making process). Such explicit methods may also offer detailed explanations via explainable AI and forecast the potential long-term outcomes of these choices, both positive and negative. Conversely, implicit feedback incorporates nudges seamlessly into daily routines, without overtly presenting the user with recommendations; for example, by choosing a route that avoids fast-food restaurants known to tempt the user.

• Human-in-the-Loop Systems:  Examining the integration of practitioners and health experts within human-in-the-loop systems to guide users toward healthier behavioral choices. This exploration can involve the display of user data, such as dietary intake and physical activity levels, to health consultants or personal trainers. Utilizing tools such as dashboards, experts could work collaboratively to create personalized plans for the user or patient, identifying opportunities for increased activity in daily routines, or avoiding situations that trigger unhealthy habits. Feedback can subsequently be provided to the AI system to enhance its learning capabilities and situational recognition. Improvements to the system should be based on user data and expert input while ensuring that privacy and user autonomy are maintained.

• MR/VR in Health Interventions:  Assessing the potential of Mixed Reality (MR) and Virtual Reality (VR) applications in delivering effective health interventions and recommendations. Looking toward the distant future, we envision employing MR and VR in daily life to illustrate the consequences of various habits. MR could, for example, be applied to communicate certain details about potential decisions in-situ. VR on the other hand could be used to simulate a certain result of a long-term intervention.

• Motivation for Sports and Exertion:  Investigating how exergames and AI algorithms that adjust exercise difficulty can effectively motivate individuals for sports and physical exertion. This includes the integration of gamification and AI components to address different player types. A possible application is the dynamic adjustment of exergame difficulty based on user performance to maintain a targeted heart rate for effective training. Further, an AI health assistant can aim to identify opportune moments in everyday life to prompt users to engage in physical activities in the first place.

• Dietary Behavior and Digital Twins:  Evaluating methodologies for tracking nutritional intake as a basis for digital twins and examining how AI tools can foster and communicate healthier food choices. By improving tracking possibilities with AI the data about the user will improve allowing the creation of better user models and improved recommendations. Further, novel approaches to model humans including their health-related behavior is a crucial challenge.

• AI-Driven Time Management and Health Routines:  Investigating the impact of AI-driven suggestions on daily life routines on overall health and well-being. This includes also research tools, methods, and guidelines to evaluate such AI-driven applications. Given the specific context and the goal of long-term improvement, researchers are in need of creating new tools and methods. Additionally, guidelines and lessons learned from evaluations provide a valuable source for future research endeavors.

• Explainability of AI-driven recommendations:  Providing means to explain recommendations in health-related scenarios is important to create trust in the recommendations. Users should be able to understand and question the reasoning behind AI-driven recommendations so that a certain level of explainability (e.g., why a specific recommendation is provided) is available and the AI system is not a black box.

Jonas Keppel is a PhD student and research assistant in the Human-Computer Interaction group at the University of Duisburg-Essen. He is currently engaged in a BMBF-funded project about enhanced health intelligence for personal behavioral strategies in everyday life (Eghi). His research interest focuses on nudging users toward a more active lifestyle using AI-based health recommendations, digital artifacts, and exergames.

Dijana Ivezić is a research associate at University of Freiburg as a part of Intelligent Embedded Systems Lab. She is currently engaged in human behaviour modelling and data driven recommender systems.

Uwe Gruenefeld is a Postdoc Researcher in Human-Computer Interaction at the University of Duisburg-Essen. He is fascinated by all flavors of Human-Computer Interaction (HCI), but Mixed Reality devices and technologies are particularly interesting to him. In his fellowship-funded dissertation at the University of Oldenburg, he focused on visualizing objects outside the human field of view. Today, his research focuses on the intersection of Mixed Reality and Artificial Intelligence, including topics such as robotics, haptics, and security & privacy.

Paul Lukowicz is the Scientific Director and Head of the "Embedded Intelligence" research division at the German Research Center for Artificial Intelligence (DFKI) in Kaiserslautern. He  founded the Wearable Computer Group at the ETH Zurich between 1999 and 2004 and served as a full professor of Computer Science at the Private University for Health Sciences, Medical Informatics, and Technology (UMIT) in Hall in Tirol. Since 2014, he manages the DFKI SmartCity Living Lab. His research focuses on cyber-physical systems, pervasive computing, and social-interactive systems.

Oliver Amft is a full professor at the University of Freiburg, specializing in Intelligent Embedded Systems, and serves as the Institute Director at the Hahn-Schickard-Institute. He holds a PhD from ETH Zurich and has a diverse background that includes R&D project management and academic roles at multiple universities. With expertise in context recognition, wearable sensor technology, and biomedical system design, Amft has co-authored over 200 publications.

Stefan Schneegass is a full professor of Human-Computer Interaction at the University of Duisburg-Essen. He is interested in researching the crossroads of Human-Computer Interaction and Ubiquitous Computing. He organized several workshops at conferences such as CHI and Ubicomp. He also served as subcommittee chair for CHI 2023 and 2024.