Related Experiment Video
Updated: May 17, 2026
08:45
Assessment of Physical Activity Intensity with Accelerometers and Oxygen Consumption
Published on: June 20, 2025
Energy expenditure estimation during normal ambulation using triaxial accelerometry and barometric pressure.
Jingjing Wang1, Stephen J Redmond, Matteo Voleno
1Graduate School of Biomedical Engineering, University of New South Wales, Sydney, NSW, 2052, Australia.
Physiological Measurement
|November 1, 2012
Summary
This study improves energy expenditure estimation using accelerometry and barometric pressure sensors. Adding barometric data significantly enhances activity classification and energy expenditure accuracy, especially during stair climbing.
Area of Science:
- Biomedical Engineering
- Sports Science
- Wearable Technology
Background:
- Accurate energy expenditure (EE) estimation is crucial for physical activity assessment.
- Current wearable sensors struggle to precisely measure EE during activities involving significant altitude changes, like stair climbing.
- Existing practical methods often provide inaccurate EE estimates in free-living conditions.
Purpose of the Study:
- To evaluate the effectiveness of augmenting a triaxial accelerometry sensor with a barometric pressure sensor for improved EE estimation.
- To enhance EE accuracy, particularly during stair ascent and descent, in unsupervised environments.
- To investigate the impact of activity classification on EE estimation models.
Main Methods:
- A waist-worn triaxial accelerometry sensor was augmented with a barometric pressure sensor.
- Features from both sensors were extracted to train a state-space model for EE estimation.
- An activity classification algorithm was developed and its output was used as a model input.
- A protocol involving various activities (lying, sitting, walking, stair climbing) was performed by 13 healthy volunteers.
- Instantaneous oxygen uptake (VO2) was measured using indirect calorimetry for validation.
Main Results:
- Activity classification accuracy improved from 81.65% to 90.91% with barometric pressure data.
- Accuracy for classifying walking activities alone increased from 70.23% to 98.54%.
- The combined accelerometry and barometry model, including activity classification, yielded a VO2 estimation bias of -0.00095 and precision of 3.54 ml min(-1) kg(-1).
- The accelerometry-only model showed worse performance with a bias of -0.09 and precision of 5.99 ml min(-1) kg(-1).
Conclusions:
- Augmenting accelerometry with barometric pressure significantly improves the accuracy of energy expenditure estimation, especially during stair climbing.
- The proposed method offers a more reliable approach for assessing physical activity in real-world, unsupervised settings.
- Integrating activity classification further refines EE estimation, reducing errors associated with changes in altitude.
