GaitKeeper: Revolutionising Standardised Gait Speed Measurement with AI-Enabled Mobile Technology

Abstract ID
3051
Authors' names
N. Davey 1,2, G. Harte 1,5, A. Boran 3,4, P. Mc Elwaine 1, 2, S P Kennelly 1,2,4
Author's provenances
1. Institute of Memory and Cognition, Tallaght University Hospital, Dublin, Ireland. 2. Department of Medical Gerontology, School of Medicine, Trinity College Dublin, Ireland. 3. Insight Centre, Dublin City University, Ireland. 4. Digital Gait Labs, Dubli
Abstract category
Abstract sub-category

Abstract

Introduction
Gait speed, often referred to as the 'sixth vital sign,' is an important health indicator in older adults, predicting morbidity and functional status. This study evaluated GaitKeeper, a novel artificial intelligence (AI)-enabled mobile technology that integrates augmented reality (AR). GaitKeeper is designed to standardise the measurement of gait speed and address inconsistencies commonly encountered in traditional clinical settings due to varied assessment techniques.

Methods
This study was conducted in two phases to validate GaitKeeper against Vicon and GaitRite, two established gait analysis systems. Phase One involved thirty-five healthy volunteers from a university setting, comparing gait speed, stride length and step length between GaitKeeper and Vicon. Phase Two tested GaitKeeper in a clinical environment with thirty participants diagnosed with mild cognitive impairment to assess the comparative accuracy and reliability of GaitKeeper and GaitRite in recording gait dynamics.

Results
In Phase One, GaitKeeper demonstrated high external consistency with the Vicon system, showing less than 2% variance in measurements of gait speed and stride length. Spearman correlation coefficients were 0.947 for gait speed and 0.989 for stride length, both statistically significant (p<0.0001). Phase Two established GaitKeeper's reliability in clinical assessments, exhibiting a strong Pearson correlation coefficient of 0.71 for stride length with GaitRite—also highly significant (p<0.0001). The Spearman correlation coefficient for gait speed was 0.918 (p=0.000) indicating a high degree of consistency between the two systems.

Conclusion
GaitKeeper has been validated as a reliable and precise tool, providing standardised measurements of gait speed in a timely manner. Additionally, GaitKeeper supports longitudinal monitoring, crucial for managing chronic conditions and rehabilitation programmes. Its versatility allows for deployment in a variety of settings, from traditional hospital environments to home-based rehabilitation where routine gait speed assessments can be challenging. This adaptability positions GaitKeeper to revolutionise gait analysis across diverse healthcare contexts.