Assessing elderly persons' fall risk using spectral analysis on accelerometric data - a clinical evaluation study

Abstract

Falls are among the leading causes for morbidity, mortality and lasting functional disability in the elderly population. Several studies have shown the applicability of accelerometry to detect persons with a high fall risk. Most of these studies have been conducted under laboratory settings and without clear definition of 'fall risk' reference measures. The aim of our work is to provide a simple unsupervised method to assess the fall risk of elderly persons as measured by reference clinical fall risk assessment scores. Our method uses parameters computed by spectral analysis on triaxial accelerometer data recorded in a clinical setting, and is evaluated using simple logistic regression classifier models with reference to three clinical reference scores. The overall prediction accuracy of the models ranges from 65.5-89.1%, with sensitivity and specificity between 78.5-99% and 15.4-60.4%, respectively. Our results show that our simple method can be used to detect persons with a high fall risk with a fair to good predictive accuracy when tested against common clinical reference scores. Our parameters are independent of specific test procedures and therefore are suited for use in an unsupervised setting. Our future research will include the evaluation of our method in a large prospective study.

Phone Based Fall Risk Prediction

“…[55], [56], [57], [58] and [59]) or using machine learning techniques, with the ground truth provided by clinically accepted scales (cf. [47], [32], [48], [6] and as suggested in [50]). …”

“…time to complete the task or the number of steps required) is easy to assess. Other qualitative measures are also evaluated by doctor, based on the observation of the task [7,31,32]. A popular test of this kind is the Timed Up and Go Test (TUGT), which is highly recommended by [19].…”

“…cf. [32], [47] and [7]), directed routines with a series of movements and assessment tasks (e.g. cf.…”

The GAL monitoring concept for distributed aal platforms

“…• A fall prevention through a gait analysis [2] by the above sensors and an additional laser scanner.…”

“…Many different supporting techniques rely on data gathered from many different sensors for various purposes. E.g., while most fall detection algorithms [1] only base on data from the past few seconds, a fall risk assessment [2] may need data from a longer period of time, and when performing Activity of Daily Living (ADL) monitoring [3] nearly everything has to be stored for further processing and analyses.…”