Measuring and analyzing HR(V) for optimizing health and resilience

Mobile technology and wearables assist us in tracking our well-being and fitness levels, though generated insights are seldomly actionable. Moreover, they are restricted to the quality of today’s measuring techniques. 

The aim of this project is twofold. First, the project aims at exploring new measuring techniques to collect highly accurate heart rate (HR) and heart rate variability (HRV) data. Speckleplethysmography (SPG), for example, shows important advantages over photoplethysmography (PPG), which is currently the most widely implemented technology in wearable health monitoring. Second, the project aims at working toward a next generation of fitness insights using artificial intelligence (AI) and smart sensor fusion. This means the evolution from descriptive statistics to predictive and, even further, prescriptive models. Prescriptive analytics goes beyond just predicting and forecasting the future state of a system. Instead, it focusses on recommending actions and interventions to change the status quo and to reach a desired outcome. 

This research is also related to the wider field of causal inference, addressing the challenge of uncovering cause-and-effect relationships in data. Identifying causal relationships and quantifying their strength from observational (and sometimes interventional) data are key problems in disciplines dealing with complex systems such as the human body. 

In this project you will investigate and bring together various techniques to accurately measure HR(V). You will also develop and implement machine learning models that can provide users with concrete actions to undertake given a desired fitness level in the future. Possible domains of application are workload, nutrition, sports performances, ... and many more. You will work together with data scientists, hardware engineers and health & sports sciences domain experts.