Related Experiment Video
Updated: Jun 14, 2025
06:28
Biomechanical Changes Related to Low Back Pain: An Innovative Tool for Movement Pattern Assessment and Treatment Evaluation in Rehabilitation
Published on: December 13, 2024
448
Algorithm Validation for Quantifying ActiGraph™ Physical Activity Metrics in Individuals with Chronic Low Back Pain
Jordan F Hoydick1, Marit E Johnson2, Harold A Cook1
1Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA 15213, USA.
Sensors (Basel, Switzerland)
|August 29, 2024
Summary
Researchers successfully replicated ActiGraph™ algorithms in MATLAB, validating its use for physical activity monitoring in chronic low back pain (cLBP) patients and healthy individuals. This method allows comparable accelerometer data analysis, crucial for chronic condition management.
Area of Science:
- Biomedical Engineering
- Wearable Technology
- Rehabilitation Science
Background:
- Physical activity assessment is vital for managing chronic conditions like chronic low back pain (cLBP).
- ActiGraph™ is a common tool for monitoring physical activity using proprietary algorithms on raw acceleration data.
Purpose of the Study:
- To replicate ActiGraph™ algorithms in MATLAB.
- To validate the MATLAB replication against ActiGraph™'s software (ActiLife™).
- To assess validity in both healthy controls and individuals with cLBP.
Main Methods:
- Developed MATLAB code to replicate ActiGraph™'s activity and step count algorithms.
- Processed raw acceleration data from ActiGraph™ GT9X devices worn by 24 participants (12 cLBP, 12 healthy) for up to seven days.
- Compared data processed via MATLAB against ActiLife™ (v6) for accuracy.
Main Results:
- MATLAB code successfully replicated ActiGraph™ algorithms for activity counts, step counts, counts per minute (CPM), and intensity cut points.
- Percent errors between ActiLife™ and MATLAB processing were less than 2% across all participants.
- Validation demonstrated minimal error differences between the two methods for both cLBP and healthy groups.
Conclusions:
- ActiGraph™ algorithms can be accurately replicated in MATLAB.
- The validated MATLAB method provides a reliable alternative for analyzing accelerometer data.
- This approach enables comparable data analysis across different accelerometers, supporting research in physical activity and chronic condition management.
