Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Nov 30:2020:6622285.
doi: 10.1155/2020/6622285. eCollection 2020.

On the Heterogeneity of Existing Repositories of Movements Intended for the Evaluation of Fall Detection Systems

Eduardo Casilari  1 , José A Santoyo-Ramón  1 , José M Cano-García  1
Affiliations

On the Heterogeneity of Existing Repositories of Movements Intended for the Evaluation of Fall Detection Systems

Eduardo Casilari et al. J Healthc Eng. .

Abstract

Due to the serious impact of falls on the autonomy and health of older people, the investigation of wearable alerting systems for the automatic detection of falls has gained considerable scientific interest in the field of body telemonitoring with wireless sensors. Because of the difficulties of systematically validating these systems in a real application scenario, Fall Detection Systems (FDSs) are typically evaluated by studying their response to datasets containing inertial sensor measurements captured during the execution of labelled nonfall and fall movements. In this context, during the last decade, numerous publicly accessible databases have been released aiming at offering a common benchmarking tool for the validation of the new proposals on FDSs. This work offers a comparative and updated analysis of these existing repositories. For this purpose, the samples contained in the datasets are characterized by different statistics that model diverse aspects of the mobility of the human body in the time interval where the greatest change in the acceleration module is identified. By using one-way analysis of variance (ANOVA) on the series of these features, the comparison shows the significant differences detected between the datasets, even when comparing activities that require a similar degree of physical effort. This heterogeneity, which may result from the great variability of the sensors, experimental users, and testbeds employed to generate the datasets, is relevant because it casts doubt on the validity of the conclusions of many studies on FDSs, since most of the proposals in the literature are only evaluated using a single database.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
Boxplots of the mean Signal Magnitude Vector (μSMV) for the ADLs (left column) and falls. (a) ADLs. (b) Falls (right column) of all datasets.
Figure 2
Figure 2
Boxplots of the maximum variation of the standard deviation of the Signal Magnitude Vector (σSMV) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 3
Figure 3
Boxplots of the mean rotation angle (µθ) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 4
Figure 4
Boxplots of the mean absolute difference between consecutive samples (μSMVdiff) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 5
Figure 5
Boxplots of the mean module of the not perpendicular acceleration components (µAp) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 6
Figure 6
Boxplots of the maximum variation of the acceleration components (Aωdiff(max)) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 7
Figure 7
Boxplots of the maximum (SMVmax) of the SMV for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 8
Figure 8
Boxplots of the minimum (SMVmin) of the SMV for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 9
Figure 9
Boxplots of the skewness of SMV (γSMV) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 10
Figure 10
Boxplots of the Signal Magnitude Area (SMA) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 11
Figure 11
Boxplots of the sum of the energy (E) estimated in the three axes for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 12
Figure 12
Boxplots of the mean of the autocorrelation function (μR) for the ADLs (left column) and falls (right column) of all datasets. (a) ADLs. (b) Falls.
Figure 13
Figure 13
Multiple comparison test of the means of the following statistical characteristics of the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max).
Figure 14
Figure 14
Multiple comparison test of the means of the following statistical characteristics of the datasets: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR.
Figure 15
Figure 15
Multiple comparison test of the means of the statistical characteristics of the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max). Observation window of 0.5 s.
Figure 16
Figure 16
Multiple comparison test of the means of the statistical characteristics of the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max). Observation window of 2 s.
Figure 17
Figure 17
Multiple comparison test of the means of the statistical characteristics of the basic movements in the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max). Basic movements.
Figure 18
Figure 18
Multiple comparison test of the means of the statistical characteristics of the basic movements in the datasets: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR. Basic movements.
Figure 19
Figure 19
Multiple comparison test of the means of the statistical characteristics of the standard movements in the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max). Standard movements.
Figure 20
Figure 20
Multiple comparison test of the means of the statistical characteristics of the standard movements in the datasets: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR. Standard movements.
Figure 21
Figure 21
Multiple comparison test of the means of the statistical characteristics of the sporting movements in the datasets: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max). Sporting movements.
Figure 22
Figure 22
Multiple comparison test of the means of the statistical characteristics of the sporting movements in the datasets: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR. Sporting movements.
Figure 23
Figure 23
Multiple comparison test of the means of the statistical characteristics of the “near falls” in the two datasets that contain this type of movement: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max).
Figure 24
Figure 24
Multiple comparison test of the means of the statistical characteristics of the “near falls” in the two datasets that contain this type of movement: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR. Near falls.
Figure 25
Figure 25
Multiple comparison test of the means of the statistical characteristics of the movements labelled as “walking” in the seven datasets that contain this type of ADL (sensor located on the waist): (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max).
Figure 26
Figure 26
Multiple comparison test of the means of the statistical characteristics of the movements labelled as “walking” in the seven datasets that contain this type of ADL (sensor located on the waist): (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (e) E. (f) μR.
Figure 27
Figure 27
Multiple comparison test of the means of the statistical characteristics of the datasets for the measurements on the wrist and ADL movements: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max).
Figure 28
Figure 28
Multiple comparison test of the means of the statistical characteristics of the datasets for the measurements on the wrist and ADL movements: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (f) E. (f) μR.
Figure 29
Figure 29
Multiple comparison test of the means of the statistical characteristics of the datasets for the measurements on the wrist and fall movements: (a) μSMV. (b) σSMV. (c) μθ. (d) μSMV(diff). (e) μAp. (f) Aωdiff(max).
Figure 30
Figure 30
Multiple comparison test of the means of the statistical characteristics of the datasets for the measurements on the wrist and ADL movements: (a) SMVmax. (b) SMVmin. (c) γSMV. (d) SMA. (f) E. (f) μR.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. World Health Organization (WHO) Falls (facts sheet, 16 January 2018) 2018.
    1. Lord S. R., Sherrington C., Menz H. B., Close J. C. T. Falls in Older People: Risk Factors and Strategies for Prevention. Cambridge, UK: Cambridge University Press; 2007.
    1. Ku Y.-C., Liu M.-E., Tsai Y.-F., Liu W.-C., Lin S.-L., Tsai S.-J. Associated factors for falls, recurrent falls, and injurious falls in aged men living in taiwan veterans homes. International Journal of Gerontology. 2013;7(2):80–84. doi: 10.1016/j.ijge.2012.07.004. - DOI
    1. Becker C., Schwickert L. Proposal for a multiphase fall model based on real-world fall recordings with body-fixed sensors. Zeitschrift für Gerontologie und Geriatrie. 2012;45(8):707–715. doi: 10.1007/s00391-012-0403-6. - DOI - PubMed
    1. Kangas M., Vikman I., Nyberg L., Korpelainen R., Lindblom J., Jämsä T. Comparison of real-life accidental falls in older people with experimental falls in middle-aged test subjects. Gait Posture. 2012;35(3):500–505. - PubMed

Publication types