{"title":"A light-weight system for detecting indoor wandering of demented elders living alone","authors":"Weichao Zhao, Xingshe Zhou, Hongbo Ni, Qiang Lin","doi":"10.1109/ICOT.2014.6954695","DOIUrl":null,"url":null,"abstract":"Wandering can be seen as one of the most frequent and dangerous behaviors of the elderly with dementia, bringing negative effects on their physical and mental health and heavy caring burden on stakeholders. In order to detect the indoor wandering of demented elders living alone in real time, this paper proposes a light-weight wandering detection system named LWDS. Specifically, the LWDS system first collects the location information with timestamp using a pair of pyroelectric infrared sensors mounted on the doors of each room, and then classifies the ongoing trajectory consisting of a series of locations into different wandering locomotion patterns according to an established wandering model. After a trajectory has been detected as containing wandering behavior, the developed system will send an alert to their carers for necessary treatment. The non-intrusive and easy-to-deploy properties ensure the proposed LWDS system can find wandering behaviors in a cost-effective and energy-efficient manner around the clock. An experimental evaluation conducted on a group of real data collected from two elders showed that the LWDS is effective in detecting indoor wandering when they moved across different rooms and obtained a detection ratio of more than 90.03% for the wandering patterns.","PeriodicalId":343641,"journal":{"name":"2014 International Conference on Orange Technologies","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 International Conference on Orange Technologies","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICOT.2014.6954695","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
Wandering can be seen as one of the most frequent and dangerous behaviors of the elderly with dementia, bringing negative effects on their physical and mental health and heavy caring burden on stakeholders. In order to detect the indoor wandering of demented elders living alone in real time, this paper proposes a light-weight wandering detection system named LWDS. Specifically, the LWDS system first collects the location information with timestamp using a pair of pyroelectric infrared sensors mounted on the doors of each room, and then classifies the ongoing trajectory consisting of a series of locations into different wandering locomotion patterns according to an established wandering model. After a trajectory has been detected as containing wandering behavior, the developed system will send an alert to their carers for necessary treatment. The non-intrusive and easy-to-deploy properties ensure the proposed LWDS system can find wandering behaviors in a cost-effective and energy-efficient manner around the clock. An experimental evaluation conducted on a group of real data collected from two elders showed that the LWDS is effective in detecting indoor wandering when they moved across different rooms and obtained a detection ratio of more than 90.03% for the wandering patterns.