Related Experiment Video
Updated: Jun 27, 2026
08:45
Assessment of Physical Activity Intensity with Accelerometers and Oxygen Consumption
Published on: June 20, 2025
Accelerometry to Estimate Energy Expenditure during Activity: Best Practice with Data Loggers
L G Halsey1, J A Green, R P Wilson
1Roehampton University, London, United Kingdom. l.halsey@roehampton.ac.uk
Physiological and Biochemical Zoology : PBZ
|November 21, 2008
Summary
Researchers found that while multi-dimensional acceleration best predicts oxygen consumption, simpler single-axis or overall dynamic body acceleration (ODBA) also serve as effective proxies. Reducing sampling frequency and axes measured can extend data logger deployment, optimizing energy expenditure measurements.
Area of Science:
- Animal physiology
- Biomechanics
- Bio-logging technology
Background:
- Acceleration measurement serves as a proxy for energy expenditure during movement.
- Overall dynamic body acceleration (ODBA) combines multi-dimensional acceleration but has limitations.
- High-frequency accelerometry recording is limited by data logger memory and battery life.
Purpose of the Study:
- To identify optimal acceleration-based proxies for oxygen consumption (Vo2) in bantam chickens.
- To evaluate the impact of different sampling frequencies on the accuracy of these proxies.
- To determine methods for maximizing data logger deployment periods.
Main Methods:
- Bantam chickens were equipped with accelerometer loggers and walked at varying speeds within a respirometer.
- Various acceleration-based models were tested against measured oxygen consumption (Vo2).
- The effect of reducing sampling frequency on calibration accuracy was assessed.
Main Results:
- A multiple regression model using individual dimensions of dynamic acceleration was the best predictor of Vo2.
- Overall dynamic body acceleration (ODBA) and single-dimension acceleration measures also showed relatively high predictive accuracy (R²).
- Vo2 calibration against ODBA remained consistent until sampling dropped below 600 samples per ODBA data point.
Conclusions:
- Optimizing data logger storage for acceleration-based Vo2 proxy measurements can be achieved by reducing the number of axes measured and decreasing sampling frequency.
- Simpler, single-variable acceleration metrics offer practical alternatives for energy expenditure estimation.
- Careful consideration of sampling parameters is crucial for balancing data accuracy and logger efficiency.
