Algorithm to detect abnormal states of elderly persons for home monitoring
Kazuyoshi Sawai, Masaki Yoshida · Systems and Computers in Japan · 2007
Abstract With the advance of an elderly‐dominated society, the population of elderly persons who live alone is increasing. To allow health checking of the solitary elderly when at home, it is planned to install sensors in houses for long‐term monitoring. The purpose of this is to detect abnormal states of solitary elderly persons in daily life (such as staying immobile due to bad physical condition or inability to move as a result of falling or sudden change of physical condition). An algorithm was developed to detect abnormal states of the subject (an elderly person) by using data on the operating times of various sensors installed in an actual home. An algorithm based on the simple logical operation of ORing all sensor output signals is considered, following the idea that the sensors will not respond for a long time if an abnormality occurs in the subject. The effect of the setting of the no‐response time of the sensors on the detection rate is investigated. It is concluded that abnormalities can be satisfactorily detected, if a sensor response is absent for 2 hours and 15 minutes. © 2007 Wiley Periodicals, Inc. Syst Comp Jpn, 38(6): 34–42, 2007; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/scj.20411 Copyright © 2004 Wiley Periodicals, Inc.