Measuring postural stability with an inertial sensor: validity and sensitivity

Abstract

Introduction/purpose

To examine the concurrent validity, and sensitivity, of an inertial sensor for use in the assessment of postural sway.

Methods

This was a laboratory-based, repeated-measures design with ten healthy participants. Concurrent validity was tested between an inertial sensor, forceplate, and rigid-body kinematics across three commonly used balance tests. Further, the inertial sensor measures were compared across eight commonly used tests of balance. Variables manipulated include stance position, surface condition, and eyes-open versus eyes-closed.

Results

The inertial sensor was correlated to both the forceplate-derived measures (r=0.793) and rigid-body kinematics (r=0.887). Significant differences between the balance tests were observed when tested with the inertial sensor. In general, there was a three-way interactions between the three balance factors (surface, stance, and vision) leading to pairwise comparisons between each balance test. The root-mean-square showed an increase across tasks of greater difficulty ranging from an average of 0.0368 with two legs, eyes-open to 0.911 when tested during tandem stance, eyes-closed tested on a foam pad.

Conclusion

The new inertial sensor shows promise for use in the assessment of postural sway. Additionally, the inertial sensor appears sensitive to differences in balance tasks of varying degrees of difficulty when tested in a healthy sample of young adults. This inertial sensor may provide new opportunities for further research in the assessment of balance changes in the mild traumatic brain injury population.

Keywords:

• balance
• concussion
• forceplate
• posturography
• accelerometer

Acknowledgments

Research reported in this publication was solely the responsibility of the authors. The inertial sensors used in the study were provided by Motion Intelligence, Inc. (Ithaca, NY), and Upstate Medical University and Motion Intelligence, Inc. provided financial support of the clinical research.

Disclosure

The authors Christopher Neville and Brian Rieger have equity interest in Motion Intelligence, Inc. The authors report no other conflicts of interest in this work.