Assessing Sensor-Derived Features From a Wrist-Worn Wearable Device as Indicators of Upper Extremity Function in
Tessa C Johnson1, Cole Hagen1, Donna L Coffman2
1Department of Health and Rehabilitation Sciences, College of Public Health, Temple University, Philadelphia, PA.
Archives of Physical Medicine and Rehabilitation
|March 15, 2025
Summary
Sensor-derived features effectively measure upper extremity function in cervical spinal cord injury (cSCI) patients. These metrics show promise for monitoring recovery and guiding clinical adoption.
Area of Science:
- Biomedical Engineering
- Rehabilitation Science
- Neuroscience
Background:
- Cervical spinal cord injury (cSCI) significantly impairs upper extremity function.
- Objective assessment of functional recovery in cSCI is crucial for effective rehabilitation.
- Wearable sensor technology offers a potential avenue for objective functional assessment.
Purpose of the Study:
- To evaluate the correlation between sensor-derived metrics and upper extremity function in individuals with acute and chronic cSCI.
- To determine the reproducibility of these sensor-derived features in individuals with chronic cSCI.
Main Methods:
- A prospective, longitudinal study involving 40 adults with cSCI.
- Participants wore a wrist-worn inertial measurement unit to collect accelerometer and gyroscope data.
- Upper extremity function was assessed using the Capabilities of Upper Extremity Test (CUE-T) at two time points, four weeks apart.
Main Results:
- Multiple sensor-derived features demonstrated strong correlations with both hand and arm function scores in individuals with cSCI.
- In the chronic subgroup, CUE-T scores exhibited excellent reproducibility (ICC, 0.94-0.99).
- Sixteen sensor-derived features showed moderate to good reproducibility (ICC, 0.50-0.77) in the chronic subgroup.
Conclusions:
- Sensor-derived features are viable indicators of upper extremity function in cSCI.
- These features support the monitoring of recovery and functional outcomes in cSCI patients.
- Further validation is recommended for digital biomarker development and clinical integration.
