Information Technology Reference
In-Depth Information
Assessing the Kinect's Capabilities
to Perform a Time-Based Clinical Test
for Fall Risk Assessment in Older People
Jaime A. Garcia
1
, Yusuf Pisan
2
, Chek Tien Tan
2
, and Karla Felix Navarro
1
1
mHealth Lab - iNEXT Reasearch Group, University of Technology Sydney, Sydney, Austalia
2
Games Studio, University of Technology Sydney, Sydney, Australia
Jaime.A.GarciaMarin@student.uts.edu.au,
{Yusuf.Pisan,ChekTien.Tan,Karla.FelixNavarro}@uts.edu.au
Abstract.
The Choice Stepping Reaction Time (CSRT) task is time-based
clinical test that has shown to reliably predict falls in older adults. Its current
mode of delivery involves the use of a custom-made dance mat device. This
mat is a measurement tool that can reliably obtain step data to discriminate
between fallers and non-fallers. One of the pitfalls of this test is that the
technology in use still imposes an obstacle on the degree of freedom to be able
to perform adaptive exercises suitable for the elderly. In this paper, we describe
a Kinect-based system that measures stepping performance through the use of a
hybrid version of the CSRT task. This study focuses on assessing this system's
capabilities to reliably measure a time-based clinical test of fall risk. Results
showed a favorable correspondence and agreement between the two systems,
suggesting that this platform could be potentially useful in the clinical practice.
Keywords:
Kinect, Elderly, Fall Risk Assessment, Reaction Time Test,
Stepping Performance.
1
Introduction
Since the release of the Microsoft Kinect, both research communities and industry
have been actively investigating its potential use in the area of aged care and
rehabilitation. This is mainly due to its capability to track real-time full body
movements in 3D, a characteristic that was not available in early consumer game
technologies such as the Nintendo Wii or the PS3. In the area of fall prevention and
safety for the elderly, the Kinect has also gained much interest. Kinect-based
applications range from health and home monitoring systems [1], through unobtrusive
fall detection platforms [2]; coaching, rehabilitation and therapeutic tools[3], fall
prevention training systems [4], to fall risk assessment tools [5]. Kinect-based serious
games, being one of the most popular approaches, have shown a positive acceptance
among seniors [6]. The fun factor inherent in such games and their ability to promote
physical movements are ideal to encourage the elderly to exercise [7]. An example
can be seen in the work done by Kim et al. [8], where the use of a commercial Kinect
