Frailty assessment based on wavelet analysis during quiet standing balance test

Abstract

Background

A standard phenotype of frailty was independently associated with an increased risk of adverse outcomes including comorbidity, disability and with increased risks of subsequent falls and fractures. Postural control deficit measurement during quiet standing has been often used to assess balance and fall risk in elderly frail population. Real time human motion tracking is an accurate, inexpensive and portable system to obtain kinematic and kinetic measurements. The aim of this study was to examine orientation and acceleration signals from a tri-axial inertial magnetic sensor during quiet standing balance tests using the wavelet transform in a frail, a prefail and a healthy population.

Methods

Fourteen subjects from a frail population (79±4 years), eighteen subjects from a prefrail population (80±3 years) and twenty four subjects from a healthy population (40±3 years) volunteered to participate in this study. All signals were analyzed using time–frequency information based on wavelet decomposition and principal component analysis.

Findings

The absolute sum of the coefficients of the wavelet details corresponding to the high frequencies component of orientation and acceleration signals were associated with frail syndrome.

Interpretation

These parameters could be of great interest in clinical settings and improved rehabilitation therapies and in methods for identifying elderly population with frail syndrome.

Introduction

Frail syndrome has been found to be a risk factor for mortality, comorbidity, disability and hospitalization (Fried et al., 2001). The role of falls as a major source of morbidity and mortality in frail syndrome (Tinetti et al., 1988) has prompted a growing interest in postural control deficit measurements during quiet standing (e.g. postural steadiness; Campbell et al., 1989, Izquierdo et al., 1999, Prieto et al., 1993). Traditional postural steadiness evaluation typically includes separate tests with eyes open and eyes closed performed on a force platform and are usually based in the ability of an individual to maintain the position of the body within specific spatial boundaries without moving the base of support (Prieto et al., 1993, Mathie et al., 2004).

Traditional postural control tests on force platforms require a specialized and dedicated laboratory, not being suitable for ambulatory measurement of human body balance. Inertial/magnetic tracking technology opens new perspectives to evaluate postural sway. This measurement system offers a reliable and low-cost alternative to more sophisticated instrumented approaches that are available for measurement of balance during standing and walking (Moe-Nilssen, 1998a). An inertial/magnetic tracking system uses a combination of accelerometers, rate gyros and magnetic sensors (Zhu and Zhou, 2004, Moe-Nilssen, 1998a, Moe-Nilssen, 1998b, Sabatini, 2005). The signals obtained from a sacrum-mounted accelerometer can be used to distinguish between different balance conditions (e.g. feet together and apart, and eyes open and eyes closed while standing with feet together; Mayagoitia et al., 1999, Kamen et al., 1998), as well as to distinguish between healthy elderly subjects and idiopathic elderly fallers (Cho and Kamen, 1998, Overstall et al., 1977, Campbell et al., 1989). In a previous study, Martínez-Ramírez et al. (2010) showed complementary relationships between acceleration/gyros and force plates to detect dynamic stability deficits in subjects with chronic ankle instability. Moreover, they found that the accelerometer was more sensitive in some tests of the study. However, to the authors' best knowledge, no studies have examined the relationship between trunk orientation and acceleration signal and frail syndrome in older adults.

The displacements of the center of pressure (COP) and the body center of mass (COM) are notable indicators in postural steadiness characterization and are considered to play an important role in the control of standing balance (Lee et al., 2007, Betker et al., 2006, Winter, 1995, Prieto et al., 1992). Time–domain measurements have been classically used in the assessment of postural control in elderly people. Large excursion areas and average distances from the mean COP/COM location in double and single static leg stance are indicative of postural instability (Kamen et al., 1998, Izquierdo et al., 1999, Bohannon et al., 1984, Era and Heikkinen, 1985, Maki et al., 1990). More recently, higher frequencies of postural sway have been related to aging and balance-related pathologies (Kamen et al., 1998, Winter, 1995), as well as to subtle and idiopathic falling disorders (Kamen et al., 1998).

In recent years, the interest in the use of wavelet analysis as a new technique for kinematic signal processing has increased considerably. It provides multiresolution time-localized information on the frequency content of the signal under study by successive decomposition into coarse approximation (e.g. low-frequency features of the signal) and detail information (e.g. high-frequency features of the signal; Rioul and Vetterli, 1991, Mallat, 1989). Previous studies have shown relationships between the spectral pattern of sway measured with force platforms and age-related deterioration of postural steadiness (Izquierdo et al., 1999, Kamen et al., 1998). However, relatively little work has been done on finding to what extent frailty syndrome may also be associated with a decline in the frequency content of postural adjustments for maintaining stability and balance as measured by inertial sensors. Based on these results, the aim of the present study was to use the wavelet transform to examine the orientation and acceleration signals obtained from a tri-axial inertial magnetic sensor suitable for ambulatory measurements during quiet standing tests within a healthy, a prefrail and a frail population.