Related Experiment Video
Updated: Aug 7, 2025
07:24
Home-Based Monitor for Gait and Activity Analysis
Published on: August 8, 2019
6.8K
Optimal sensor location and direction to accurately classify people with early-stage multiple sclerosis using gait
L Eduardo Cofré Lizama1, Maya G Panisset2, Liuhua Peng3
1Department of Medicine, The University of Melbourne, Parkville, VIC 3050, Australia; School of Allied Health, Human Services and Sport, La Trobe University, Bundoora, VIC 3086, Australia.
Gait & Posture
|March 8, 2023
Summary
The local divergence exponent (LDE) effectively classifies early-stage multiple sclerosis (MS) gait instability. Sternum sensor data combined with gait speed offers a simplified, sensitive method for detecting MS-related gait changes.
Area of Science:
- Biomedical Engineering
- Neurology
- Rehabilitation Science
Background:
- Gait stability assessment in people with multiple sclerosis (pwMS) often uses the local divergence exponent (LDE).
- Previous studies show reduced gait stability in pwMS, but methodologies vary across disability levels.
- Early-stage MS gait impairment detection remains challenging with current clinical tests.
Purpose of the Study:
- To identify optimal sensor locations and movement directions for classifying early-stage MS.
- To evaluate the effectiveness of the local divergence exponent (LDE) in detecting subtle gait changes in pwMS.
- To determine the best combination of LDE measures and covariates for accurate classification.
Main Methods:
- 49 pwMS (EDSS ≤ 2.5) and 24 controls walked overground; 3D acceleration data collected from sternum (STR) and lumbar (LUM) sensors.
- Unidirectional (VT, ML, AP) and 3D LDEs calculated over 150 strides.
- Receiver operating characteristic (ROC) analyses performed using LDEs, with/without gait velocity (VELLAP) as a covariate.
Main Results:
- Four models achieved high classification accuracy (AUC = 0.879) using combinations of VELLAP and various LUM/STR LDEs.
- The best single-sensor model incorporated VELLAP and STR LDEs (3D, ML, AP) (AUC = 0.878).
- Models using VELLAP with a single STR LDE (VT or 3D) also showed strong performance (AUCs = 0.869, 0.858).
Conclusions:
- LDE is a promising tool for assessing gait impairment in early-stage MS, outperforming current insensitive tests.
- Simplified clinical implementation is possible using a single sternum sensor and LDE measure, considering gait speed.
- Further longitudinal studies are needed to establish LDE's predictive power and responsiveness to MS progression.
