Citation

Lu W, Wang C, Stevens MC, Redmond SJ, Lovell NH, Wei Lu, Changhong Wang, Stevens MC, Redmond SJ, Lovell NH, Wang C, Lovell NH, Lu W, Redmond SJ, Stevens MC. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2016; 2016: 2010-2013.

Copyright

(Copyright © 2016, IEEE (Institute of Electrical and Electronics Engineers))

DOI

10.1109/EMBC.2016.7591120

PMID

28226913

Abstract

The use of a barometric pressure sensor in a wearable fall detector has been shown to improve the detection accuracy by determining the altitude change associated with the fall event. However, the barometer is a high-power-consuming sensor. This paper proposes a fall detection approach using a hermetically sealed and waterproof enclosure incorporating a small window covered by a semi-permeable membrane (SPM) to delay the equilibrium of internal and external pressures. This feature can be utilized to limit the time the barometer is powered but still capturing critical pressure information to discriminate fall and non-fall events. The proposed fall detection system is evaluated with an existing data set of simulated fall and activities of daily living in which the barometric pressure data are delayed using a mathematical model of the enclosure and SPM assembly. Also, a benchtop test is performed to estimate the power and battery life. The proposed fall detection system achieves 94.0% sensitivity and 90.0% specificity with an estimated battery life of 995.7 days.

Language: en