|
A Real-Time Mobility-Related Activity Tracking System for Mobility and Fall Risk Assessment in Elderly People |
|---|---|
| รหัสดีโอไอ | |
| Creator | กิตติศักดิ์ บัวบก |
| Title | A Real-Time Mobility-Related Activity Tracking System for Mobility and Fall Risk Assessment in Elderly People |
| Contributor | สุภัสสรา จีนบุญมี, วัศวี แสนศรีมหาชัย, มานะชัย โต๊ะชูดี |
| Publisher | Faculty of Information Science and Technology, Mahanakorn University of Technology |
| Publication Year | 2559 |
| Journal Title | Journal of Information Science and Technology |
| Journal Vol. | 6 |
| Journal No. | 1 |
| Page no. | 16-24 |
| Keyword | Mobility-related activity tracking, Elderly, Fall risk assessment, Data stream mining |
| URL Website | https://tci-thaijo.org/index.php/JIST |
| Website title | Journal of Information Science and Technology |
| ISSN | 2651-1053 |
| Abstract | Over the last decade, population structure of Thailand has changed dramatically where the proportion of elderly people (persons aged 60 or over) in the population increases rapidly and continuously. It is expected that Thailand will become a super aged society by 2040 or about 20 years from now. The elderly are usually faced with many problems resulting from the deterioration of health with increasing age. One of the major problems in the elderly is falls balance and gait disorders in the elderly. Falls have significant effects on both physiological and psychological condition of elderly people. They consequently lead to fracture, serious injuries, disability or eventually death. To support medical staffs and caregivers who provide care to elderly people, we propose an innovative software system that can analyze physical activities and assess the risk of falls in elderly remotely. Our system utilizes acceleration force data and gyroscope data derived from a mobile phone attached to the body of the elderly in order to classify mobility-related activities of the elderly in real-time. It can store a significant amount of mobility-related activity information and predict the risk of falls in elderly people. The experimental results demonstrate that our proposed system can work efficiently in real environment with a high activity recognition rate - about 93.90% for overall accuracy. |
