Small body-fixed movement sensors allow for detection of the type of physical activity but often do not meet the feasibility requirements for epidemiological research. This study was established to develop an initial algorithm for physical activity type classification based on a single sensor location and to develop a multi-sensor system to serve as a gold standard for physical activity classification in future studies. 27 healthy adults between 18 and 55 years were recruited. Self-paced daily life activities organised in nine daily life simulations were investigated. The participants wore 13 sensors (of which 10 were movement sensors, 2 were heart rate sensors, and 1 was a combined acceleration-temperature-galvanic sensor). Additionally, the participant was asked to continue wearing 8 sensors, until the next morning and to note down the travelling mode and time, sleeping time, and sensor wearing time.
Information for researchers
This study is completed. It was led by the MRC Epidemiology Unit with responsibility for data, and funded by the Medical Research Council (MRC).
Publications
PAAS publications on the MRC Epidemiology Unit publications database.
Key publications:
van Hees VT, Golubic R, Ekelund U, Brage S. Impact of study design on development and evaluation of an activity-type classifier. J Appl Physiol (2013)
Data sharing
We welcome proposals for projects and aim to make data as widely available as possible whilst safeguarding the privacy of our participants, protecting confidential data and maintaining the reputations of our studies and participants. Please see our Data Sharing pages for more information.
