Related Experiment Videos
Detecting bouts of physical activity in a field setting
L C Mâsse1, J E Fulton, K L Watson
1Department of Behavioral Sciences at the University of Texas-Houston, USA. LMASSE@utsph.uth.sph.tmc.edu
Research Quarterly for Exercise and Sport
|October 16, 1999
Summary
The TRITRAC and CSA accelerometers showed strong agreement in measuring physical activity patterns. Discrepancies in detecting moderate-intensity activity may stem from differing classification thresholds.
Area of Science:
- Physical activity monitoring
- Biomedical engineering
- Wearable technology
Background:
- Objective physical activity assessment is crucial for health research.
- Accelerometers are widely used to measure physical activity.
- Comparing different accelerometer devices is essential for data consistency.
Purpose of the Study:
- To evaluate the TRITRAC and CSA accelerometers.
- Assess inter-device agreement for physical activity patterns.
- Compare detection of moderate-intensity activity and walking patterns.
Main Methods:
- Thirty-one women participated in the study.
- Wore both TRITRAC and CSA accelerometers for three days.
- Agreement assessed using generalizability coefficients and pattern comparison.
Main Results:
- High inter-device agreement observed (generalizability coefficients .88–.99).
- 71.3% agreement in pattern length; CSA patterns were longer.
- Comparable agreement in detecting brisk walking patterns (69.4%).
Conclusions:
- TRITRAC and CSA demonstrate good agreement for physical activity monitoring.
- Differences in moderate-intensity pattern detection may relate to device thresholds.
- Further research needed to understand discrepancies in accelerometer data.