Related Experiment Video
Updated: Aug 16, 2025
07:24
A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
Published on: April 21, 2017
12.6K
Stepping towards More Intuitive Physical Activity Metrics with Wrist-Worn Accelerometry: Validity of an Open-Source
Benjamin D Maylor1, Charlotte L Edwardson1,2, Paddy C Dempsey1,2,3,4
1Assessment of Movement Behaviours Group (AMBer), Leicester Lifestyle and Health Research Group, Diabetes Research Centre, University of Leicester, Leicester LE1 7RH, UK.
Sensors (Basel, Switzerland)
|December 23, 2022
Summary
The optimized Verisense algorithm accurately estimates total and moderate-to-vigorous physical activity (MVPA) steps from wrist accelerometers. This open-source algorithm offers reliable data for physical activity monitoring.
Area of Science:
- Physical Activity Measurement
- Wearable Sensor Technology
- Biomedical Engineering
Background:
- Stepping-based targets are common for physical activity goals.
- Wrist-worn accelerometers are increasingly used for surveillance.
- Validated, open-source methods for wrist-based step counting are needed.
Purpose of the Study:
- To validate the Verisense step-count algorithm (original and optimized versions) using wrist accelerometry.
- To compare wrist-derived step counts against thigh-worn activPAL.
- To assess the accuracy of estimating total and moderate-to-vigorous physical activity (MVPA) steps.
Main Methods:
- Concurrent validity study with 713 participants across three datasets.
- Continuous accelerometry wear for >24 hours on the wrist and thigh.
- Comparison of Verisense algorithms (wrist) against activPAL (thigh).
Main Results:
- Both Verisense algorithms overestimated total daily steps compared to activPAL.
- The optimized Verisense algorithm showed improved accuracy for total and MVPA steps.
- MVPA steps were underestimated by both algorithms, with the optimized version being more accurate.
Conclusions:
- The optimized Verisense algorithm demonstrates acceptable accuracy for wrist-derived stepping metrics.
- Algorithm performance beyond total step count is crucial for accurate physical activity assessment.
- The open-source nature of the optimized algorithm facilitates its use in research and public health.
