# Validity, Reliability and Interpretability of an IMU-Based System to Measure 3D Lower Limb Kinematics of Patients with Heterogeneous Gait Disorders

## TL;DR

This study shows that IMU systems can reliably and accurately measure 3D lower limb movement in people with various gait disorders, similar to more expensive optoelectronic systems.

## Contribution

The study provides evidence for the clinical validity and interpretability of IMU-based gait analysis in heterogeneous patient populations.

## Key findings

- Mean RMSE values were 7.1° for asymptomatic, 9.8° for cerebral palsy, and 9.3° for other motor disorders.
- Correlation between IMU and optoelectronic systems was good to excellent (mean CC = 0.76).
- GPS showed moderate agreement between systems (CC = 0.42).

## Abstract

Inertial measurement units (IMUs) represent a promising alternative to optoelectronic systems for estimating gait kinematics in less resource-intensive laboratories. However, evidence regarding the clinical interpretability of IMU-based gait analysis in pathological populations remains limited. This study aimed to evaluate the concurrent validity, reliability, and interpretability of an IMU-based method for computing 3D lower limb kinematics in asymptomatic and pathological populations. Fifty-five participants, including asymptomatic individuals (AS, n = 15), patients with cerebral palsy (CP, n = 15), and individuals with various motor disorders (OMD, n = 25), were assessed using a 7-IMU system and an optoelectronic system. Validity was quantified using root mean square error (RMSE), centered RMSE, and Pearson correlation coefficients (CCs) across 11 commonly reported lower limb kinematic outcomes. Reliability was assessed using intraclass correlation coefficients (ICCs), and interpretability was examined by comparing Gait Profile Scores (GPS) derived from both systems. Mean RMSE values were 7.1° (AS), 9.8° (CP), and 9.3° (OMD), with centered RMSE values below 3.2°. The correlation between IMU- and optoelectronic-based kinematics was good to excellent (mean CC = 0.76). Reliability ranged from moderate to excellent, and GPS showed moderate agreement between systems (CC = 0.42). These findings support the clinical validity, reliability, and interpretability of IMU-based gait kinematics across heterogeneous gait disorders.