Wearable sensor assessment of physical and eating behaviours
Description
- Healthy eating behaviour and healthy physical activity behaviour are associated with a lower risk of developing non-communicable diseases and obesity. Unfortunately, the combined measurement of sitting or exercise with simultaneous dietary or eating behaviour is currently still very difficult. In order to offer solutions for future science, WEALTH developed a new methodology. It used an innovative method to process accelerometry data with ecological momentary assessment (EMA), where movement and eating behaviour are measured simultaneously within its context. This approach provides a comprehensive understanding of the context-specific relationship between physical activity and eating behaviour.
WEALTH also developed a web-based infrastructure for processing passive sensor data from commercially available wearables, among others, to establish a system for the estimation of physical activity and eating behaviour and better surveillance of health-related behaviours. For this purpose, physical activity data will be collected both according to a structured protocol and under real environmental conditions from 627 adults in four European study centres in Germany, Ireland, France, and Czech republic. In Bremen, 162 adults participated.
Context-specific behavioural information was recorded using EMA prompts. Machine learning procedures based on the collected raw sensor data were used to train and validate the estimation of classified movement and eating behaviours. Finally, the behaviours were evaluated to demonstrate the feasibility of the combined assessment methods. WEALTH published the methods and procedures in a toolbox: The WEALTH Toolbox – WEALTH Study. Furthermore, the following publications are available to date.
https://doi.org/10.1371/journal.pone.0318772
https://doi.org/10.1123/jmpb.2025-0030
https://doi.org/10.1123/jmpb.2025-0025
https://formative.jmir.org/2025/1/e76167
WEALTH offers an easy-to-use and openly accessible system for measuring physical activity and eating behaviours that will be of high value for future public health surveillance.
The project was funded under the Joint Programming Initiative "A Healthy Diet for a Healthy Life" (JPI HDHL)
Funding period
- Begin: May 2022
End: September 2025
Sponsor
- Federal Ministry of Education and Research
Contact
Application (international)
- Ass. Prof. Alan Donnelly, University of Limerick, Ireland