Home // GLOBAL HEALTH 2020, The Ninth International Conference on Global Health Challenges // View article

Lessons Learned from the Development of Mobile Applications for Fall Detection

Authors:
Ítalo Linhares
Rossana Andrade
Evilasio Costa Junior
Pedro Oliveira
Breno Oliveira
Paulo Aguilar

Keywords: Falls, Lessons, IoT Health, IS for healthcare

Abstract:
Falls are the leading cause of older adults injuries and solutions are needed to address this issue. One way to meet this is by developing applications that use sensors embedded in devices like smartphones and smartwatches. This paper presents our experience in developing such applications and the lessons learned during their development and evolution. First, we developed an application called fAlert to identify a fall using data from a smartphone's accelerometer. However, the usage of this kind of mobile device for detecting falls is not natural, because it needs to be positioned at the level of the user’s chest. Then, we developed a new app called WatchAlert, which runs in smartwatches. In that case, we also created an algorithm that uses two sensors, accelerometer and gyroscope, and later evolved it to use only the accelerometer with better results. Moreover, we use first a fall detection threshold algorithm in this solution. Next, we expanded this strategy to use threshold and machine learning algorithms, which were evaluated considering the accuracy, false negative, and time criteria as well as their features. We believe that this study can support the development of new systems and devices for detecting falls. As future work, it would be interesting to assess the related energy cost of the fall detection approaches studied.

Pages: 18 to 25

Copyright: Copyright (c) IARIA, 2020

Publication date: October 25, 2020

Published in: conference

ISSN: 2308-4553

ISBN: 978-1-61208-817-4

Location: Nice, France

Dates: from October 25, 2020 to October 29, 2020