Measuring Gait Variables Using Computer Vision to Assess Mobility and Fall Risk in Older Adults With Dementia.

Kimberley-Dale Ng,Babak Taati,Sina Mehdizadeh,Alastair Flint,Avril Mansfield,Andrea Iaboni

doi:10.1109/jtehm.2020.2998326

Abstract

Fall risk is high for older adults with dementia. Gait impairment contributes to increased fall risk, and gait changes are common in people with dementia, although the reliable assessment of gait is challenging in this population. This study aimed to develop an automated approach to performing gait assessments based on gait data that is collected frequently and unobtrusively, and analysed using computer vision methods. Recent developments in computer vision have led to the availability of open source human pose estimation algorithms, which automatically estimate the joint locations of a person in an image. In this study, a pre-existing pose estimation model was applied to 1066 walking videos collected of 31 older adults with dementia as they walked naturally in a corridor on a specialized dementia unit over a two week period. Using the tracked pose information, gait features were extracted from video recordings of gait bouts and their association with clinical mobility assessment scores and future falls data was examined. A significant association was found between extracted gait features and a clinical mobility assessment and the number of future falls, providing concurrent and predictive validation of this approach.

Highlights

Dementia is a neurodegenerative disorder that impairs control of gait and increases the risk of falling [1]
We explore the utility of vision-based human pose estimation, operating on regular image/video data and without the need for special sensors, for longitudinal gait monitoring in older adults with dementia
In this study, we demonstrated that human pose estimation can be used to extract clinically meaningful data from video collected of older adults with dementia walking in environment in which they live

Summary

Introduction

Dementia is a neurodegenerative disorder that impairs control of gait and increases the risk of falling [1]. Numerous technologies exist which can be used to reduce or eliminate the subjectivity of gait assessment The majority of these tools, require expensive and difficult-to-use hardware, e.g. motion capture systems. CLINICAL GAIT ASSESSMENT Over the past two decades, functional assessments have been the most commonly used type of gait assessments for older adults [3] These tools generally test the limitations of the patients’ gait and balance and require a trained clinician to score aspects of a patient’s gait based on observations. Due to the subjective nature of these test, results can vary based on the assessor [4] They are performed infrequently due to lack of resources such as trained staff and specialized equipment [4], [5]. POMA was first published as a functional mobility assessment geared towards institutionalized older adults [6].

Objectives

Methods

Results

Conclusion