Reliability of a wireless instrumented insole (WalkinSense system) for measuring spatiotemporal and kinematic gait variables

TL;DR

This study shows that a wireless insole system reliably measures gait variables in healthy adults under different walking and running conditions.

Contribution

The study evaluates the test-retest reliability of the WalkinSense system across various speeds and slopes in healthy individuals.

Findings

Spatiotemporal gait parameters showed good to excellent reliability with ICC values between 0.76–0.95.

Foot ground angles had lower reliability (ICC: 0.71–0.86) compared to spatiotemporal variables.

SEM% and MDC% were generally below 10%, indicating low measurement error and high sensitivity.

Abstract

Reliable gait analysis is essential for clinical assessment and research. Wearable technologies such as the WalkinSense system (WSS), a wireless instrumented insole system equipped with an inertial measurement unit, enable the measurement of spatiotemporal and kinematic gait parameters in real‐world settings. The accuracy of the WSS has been previously validated. This study aims to evaluate the test–retest reliability of the WSS in healthy adults walking and running under various speed and slope conditions.

Forty‐nine healthy adults completed two sessions one week apart, walking and running at 3, 4.5 (−3°, −6°, +3°, +6°), 6 and 9 km/h on a treadmill. Spatiotemporal variables, including stance time, swing time, stride time, stride length, single and double support time, as well as kinematic variables such as foot ground angle at initial contact and toe‐off, were recorded using the WSS. Relative reliability was assessed using intraclass correlation coefficients (ICC2,1), while absolute reliability was evaluated using the standard error of measurement (SEM), minimal detectable change (MDC) and their percentage values (SEM%, MDC%). Bland‐Altman plots were used to detect systematic bias and visualise agreement.

Results demonstrated good to excellent reliability for most spatiotemporal parameters across all conditions, with ICC values ranging from 0.76–0.95, while the foot ground angles exhibited lower reliability (ICC: 0.71–0.86). SEM% and MDC% were generally below 10% for spatiotemporal measures (SEM%: 1.63–6.63; MDC%: 3.44–18.36), reflecting both low measurement error and high sensitivity to detect real changes beyond random variation. Bland‐Altman analyses revealed no relevant heteroscedasticity.

These findings support the WSS as a reliable tool for assessing spatiotemporal variables in healthy adults across diverse walking and running conditions, underscoring its potential for use in both clinical and research environments. However, clinical studies are needed to fully establish its utility in patient assessment.

Level II, diagnostic studies.

Figures