I am carrying out a pilot study of an intervention to reduce sedentary behaviour in the workplace. I will be testing an under desk machine but am cannot see a reliable and valid measure of this type of physical activity as well as sedentary behaviour.
Hi Gail,
Crowley (2019; DOI: 10.1186/s12966-019-0835-0) recently validated three different monitors worn on the thigh for detecting cycling, but using Acti4 software which is not open source.
If you have some money available, it may be worth contacting Vincent Van Hees to discuss incorporating this into GGIR. Here is a short piece he wrote where he demonstrated using AcitvPAL data to detect PA behaviours - https://medium.com/@vincentvanhees/activity-type-recognition-with-an-accelerometer-worn-on-the-thigh-during-eurovision-song-contest-3d413f55f5fb Obviously this is not validated, but I don't see why it couldn't be developed further in the future. GGIR is open source and fairly simple to use even for someone with no prior experience of using R.
I'm sure there are other methods out there too, but I believe that as long as you have a device on the thigh, you will be able to distinguish between stepping and cycling, assuming that the seat height for the bike is appropriate the user.
Hope that helps in some way,
Kind regards,
Ben
Ben Maylor thank you for your reply and helpful information, however I am looking at stationary cycling and this study is looking at outdoor cycling. From my own investigations a thigh-worn accelerometer does not pick up stationary cycling, I think maybe an ankle worn worn may be better. I'll keep looking!
Hi Gail, please check this validation paper of Fibion device: https://peerj.com/articles/5775/. They used cycling on ergometer, and validated sitting as well.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability