Evaluation of Accelerometer-Based Fall Detection Algorithms on Real-World Falls

Department of Electronics, Computer Science and Systems, University of Bologna, Bologna, Italy.
PLoS ONE (Impact Factor: 3.23). 05/2012; 7(5):e37062. DOI: 10.1371/journal.pone.0037062
Source: PubMed

ABSTRACT Despite extensive preventive efforts, falls continue to be a major source of morbidity and mortality among elderly. Real-time detection of falls and their urgent communication to a telecare center may enable rapid medical assistance, thus increasing the sense of security of the elderly and reducing some of the negative consequences of falls. Many different approaches have been explored to automatically detect a fall using inertial sensors. Although previously published algorithms report high sensitivity (SE) and high specificity (SP), they have usually been tested on simulated falls performed by healthy volunteers. We recently collected acceleration data during a number of real-world falls among a patient population with a high-fall-risk as part of the SensAction-AAL European project. The aim of the present study is to benchmark the performance of thirteen published fall-detection algorithms when they are applied to the database of 29 real-world falls. To the best of our knowledge, this is the first systematic comparison of fall detection algorithms tested on real-world falls. We found that the SP average of the thirteen algorithms, was (mean ± std) 83.0% ± 30.3% (maximum value = 98%). The SE was considerably lower (SE = 57.0% ± 27.3%, maximum value = 82.8%), much lower than the values obtained on simulated falls. The number of false alarms generated by the algorithms during 1-day monitoring of three representative fallers ranged from 3 to 85. The factors that affect the performance of the published algorithms, when they are applied to the real-world falls, are also discussed. These findings indicate the importance of testing fall-detection algorithms in real-life conditions in order to produce more effective automated alarm systems with higher acceptance. Further, the present results support the idea that a large, shared real-world fall database could, potentially, provide an enhanced understanding of the fall process and the information needed to design and evaluate a high-performance fall detector.

Download full-text


Available from: Lorenzo Chiari, Sep 28, 2015
83 Reads
  • Source
    • "It should be noted that the detection rates provided by all these studies are very high. However, many authors on this field have noticed strong difficulties when comparing different acceleration-based studies [6] [18]. This is due to the fact that each study uses its own dataset composed of simulated falls and ADL. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Falls are one of the leading causes of mortality among the older population, being the rapid detection of a fall a key factor to mitigate its main adverse health consequences. In this context, several authors have conducted studies on acceleration-based fall detection using external accelerometers or smartphones. The published detection rates are diverse, sometimes close to a perfect detector. This divergence may be explained by the difficulties in comparing different fall detection studies in a fair play since each study uses its own dataset obtained under different conditions. In this regard, several datasets have been made publicly available recently. This paper presents a comparison, to the best of our knowledge for the first time, of these public fall detection datasets in order to determine whether they have an influence on the declared performances. Using two different detection algorithms, the study shows that the performances of the fall detection techniques are affected, to a greater or lesser extent, by the specific datasets used to validate them. We have also found large differences in the generalization capability of a fall detector depending on the dataset used for training. In fact, the performance decreases dramatically when the algorithms are tested on a dataset different from the one used for training. Other characteristics of the datasets like the number of training samples also have an influence on the performance while algorithms seem less sensitive to the sampling frequency or the acceleration range. Copyright © 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
    Medical Engineering & Physics 07/2015; 37(9). DOI:10.1016/j.medengphy.2015.06.009 · 1.83 Impact Factor
  • Source
    • "From the perspective of threshold, the simplest fall detection method is based on an acceleration threshold alone (Perry et al., 2009). However, a threshold-only approach would misclassify activities and result in false positives (Bagalà et al., 2012), as some normal daily activities also produce large acceleration impact. To improve the fall detection performance, other parameters are combined by fall detection systems while a single accelerometer is worn. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Threshold-based fall detection has been widely adopted in conventional fall detection systems. In this paper, we argue that a fixed threshold is not flexible enough for different people. By exploiting the personalised and adaptive threshold, we propose a novel threshold extraction model, which meets being adaptive to detect a fall, while only taking consideration of data from activity of daily living (ADL). We believe this is a solid step toward improving the performance of the threshold-based fall detection solution. Furthermore, we incorporate the proposed idea into Chameleon. To evaluate the performance of this threshold extraction model, we compared Chameleon with advanced magnitude detection (AMD) and fixed and tracking fall detection (FTFD). The results show Chameleon has an accuracy of 96.83% when detecting falls, which is 1.67% higher than FTFD and 2.67% higher than AMD. Meanwhile, the sensitivity and the specificity of Chameleon are also higher than the other two algorithms. Reference to this paper should be made as follows: Ren, L. and Shi, W. (xxxx) 'Chameleon: personalised and adaptive fall detection of elderly people in home-based environments', Int.
  • Source
    • "This leads to challenges and debates about the sensitivity/specificity of the different algorithms used to detect falls [20] and associated with this minimisation of the number of false negative and false positive detections. Evaluating both sensitivity and specificity is clearly important as, in order for older people to feel secure when using falls detectors, they need to be assured that they are reliable. "
    [Show abstract] [Hide abstract]
    ABSTRACT: . Falls and fear of falling present a major risk to older people as both can affect their quality of life and independence. Mobile assistive technologies (AT) fall detection devices may maximise the potential for older people to live independently for as long as possible within their own homes by facilitating early detection of falls. Aims . To explore the experiences and perceptions of older people and their carers as to the potential of a mobile falls detection AT device. Methods . Nine focus groups with 47 participants including both older people with a range of health conditions and their carers. Interviews were audio recorded, transcribed verbatim, and thematically analysed. Results . Four key themes were identified relating to participants’ experiences and perceptions of falling and the potential impact of a mobile falls detector: cause of falling, falling as everyday vulnerability, the environmental context of falling, and regaining confidence and independence by having a mobile falls detector. Conclusion . The perceived benefits of a mobile falls detector may differ between older people and their carers. The experience of falling has to be taken into account when designing mobile assistive technology devices as these may influence perceptions of such devices and how older people utilise them.
    Current Gerontology and Geriatrics Research 12/2013; 2013(6260):295073. DOI:10.1155/2013/295073
Show more