A Kinect and Inertial Sensor-Based System for the Self-Assessment of Fall Risk: A Home-Based Study in Older People. (3rd July 2016)

• Record Type: Journal Article
• Authors: Ejupi, Andreas; Gschwind, Yves J.; Valenzuela, Trinidad; Lord, Stephen R.; Delbaere, Kim
• Is Part Of: Human-computer interaction. Volume 31:Isuse 3/4(2016)
• Page Start: 261
• Page End: 293
• Publication Date: 2016-07-03
• Subjects: System design -- Periodicals; Computers -- Psychological aspects -- Periodicals; Human-machine systems -- Periodicals; 004.019
• DOI: 10.1080/07370024.2015.1085309
• Languages: English
• ISSNs: 0737-0024

Abstract

Abstract : Falls remain an important problem in older people. There is strong evidence that falls can be prevented with appropriately designed intervention programs. To start a targeted fall prevention program, a first step is to identify those at high risk of falls. Sensor-based tests hold great promise for more frequent and accurate assessment of fall risk in clinical and home settings. The aims of this study were to (a) empirically examine the feasibility of the iStoppFalls (Information and communications technology–based System to Predict & Prevent Falls) assessment, a Kinect and inertial sensor-based test for regular and unsupervised fall risk assessments at home, (b) investigate the experience of older adults with this home-based self-assessment, and (c) make recommendations for future assessments. The iStoppFalls assessment system was installed into the homes of 62 community-living older people in Australia, Germany, and Spain for the duration of 4 months. Participants were asked to perform at least 1 assessment each month. The system use and the user experience were evaluated. To our knowledge, these are the first results on the long-term use of an unsupervised directed routine fall risk assessment system at private homes. In total, 241 assessments were independently performed by the participants. Most participants felt positive about their experience and could see themselves continuing with the assessment on a regular basis. Through the analysis the user motivation, the design and selection of appropriate tests, the user feedback, the reliability and usability of the applied technology, the frequency and duration of the assessment and the safety and support aspects were identified as important characteristics of a home-based self-assessment. The findings demonstrate the feasibility of a sensor-based self-assessment for fall risk but also highlight that further work is necessary. Future research should consider the necessary design requirements identified by this study.