Assessing the Reliability and Validity of a Balance Mat for Measuring Postural Stability: A Combined Robot-Human Approach

TL;DR

This study evaluates a portable Balance Mat device's reliability and validity for postural sway assessment, comparing it to force plates through robot and human experiments, and demonstrates its potential as a practical alternative.

Contribution

It introduces a low-cost, portable Balance Mat using optical fibre technology and validates its effectiveness against force plates with calibration for improved accuracy.

Findings

Balance Mat shows good to excellent reliability in robot tests.

Moderate to strong correlations with force plates in human tests.

Calibration improves agreement between Balance Mat and force plates.

Abstract

Postural sway assessment is important for detecting balance problems and identifying people at risk of falls. Force plates (FP) are considered the gold standard postural sway assessment method in laboratory conditions, but their lack of portability and requirement of high-level expertise limit their widespread usage. This study evaluates the reliability and validity of a novel Balance Mat (BM) device, a low-cost portable alternative that uses optical fibre technology. The research includes two studies: a robot study and a human study. In the robot study, a UR10 robotic arm was used to obtain controlled sway patterns to assess the reliability and sensitivity of the BM. In the human study, 51 healthy young participants performed balance tasks on the BM in combination with an FP to evaluate the BM's validity. Sway metrics such as sway mean, sway absolute mean, sway root mean square (RMS), sway path, sway range, and sway velocity were calculated from both BM and FP and compared. Reliability was evaluated using the intra-class correlation coefficient (ICC), where values greater than 0.9 were considered excellent and values between 0.75 and 0.9 were considered good. Results from the robot study demonstrated good to excellent ICC values in both single and double-leg stances. The human study showed moderate to strong correlations for sway path and range. Using Bland-Altman plots for agreement analysis revealed proportional bias between the BM and the FP where the BM overestimated sway metrics compared to the FP. Calibration was used to improve the agreement between the devices. The device demonstrated consistent sway measurement across varied stance conditions, establishing both reliability and validity following appropriate calibration.