Prof. Anastasios Delopoulos (male) graduated from the Department of Electrical Engineering of the National Technical University of Athens (NTUA) in 1987, received the M.Sc. from the University of Virginia in 1990 and the PhD degree from NTUA in 1993. He is a Professor of Multimedia Systems in the Electrical and Computer Engineering Department of the Aristotle University of Thessaloniki. The research interests of him and his group, The Multimedia Understanding Group (mug.ee.auth.gr), lie in the areas of multimedia data understanding, learning and computer vision. He is author of more than 100 journal and conference scientific papers. He has participated in 23 European and National R&D projects related to application of signal, image, video and information processing to entertainment, culture, education and health sectors. Dr Delopoulos was the coordinator of SPLENDID - Personalised Guide for Eating and Activity Behaviour for the Prevention of Obesity and Eating Disorders (FP7 – 610746) and he is currently the coordinator of BigO - Big data against childhood Obesity (H2020 - 727688). Dr. Delopoulos is a member of the Technical Chamber of Greece and of the IEEE.
The subject of his speech...
Monitoring Human Behaviour and Measuring the Environment: An application on tackling childhood obesity
The way we eat and what we eat, the way we move and the way we sleep significantly impact the risk of becoming obese. These three aspects of behaviour decompose into a long list of personal behavioural elements including our food choices, eating place preferences, transportation choices, sleeping periods and duration etc. In the first part of the presentation we will examine technological solutions that have been developed to objectively monitor a matrix of obesogenic behavioural elements using signals captured by very simple wearable devices (accelerometers, gyroscopes, GPS) embedded in smart phones and smart watches. Similarly, we examine novel methodologies for measuring the conditions of the local urban and socioeconomic environment. The second part of the presentation is devoted to the aggregation procedure that converts these highly personal time series of personal data into depersonalized spatial distributions. The proposed approach supports k-anonymity and offers controlled spatial granularity. Multiresolution visualizations methods of the resulting distributions are also presented. Most of the measured behavioural elements are highly correlated in a causal way with the conditions of our local urban, social and economic environment. The obtained correlations are thus unique sources of evidence for restructuring and improving the local environment.