A Cloud Based Platform for Elderly Sleep Detection and Health Monitoring

• Main Authors: HUA, CHI-LI, 花繼利
• Other Authors: YU, PAO-TA
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/26uj59

碩士 === 國立中正大學 === 雲端計算與物聯網數位學習碩士在職專班 === 106 === Monitoring of elderly people has drawn attention of healthcare and medical professionals. Because of the poor quality of sleep, the elderly fall 2 to 4.5 times higher than the average person during the daytime. For the elderly to fall, we must pay attention to whether the elderly have insufficient sleep or night sleep disorders such similar situation occurring. But the sleep monitoring system, which detects sleep patterns by connecting sensors to the body, can easily lead to sleep discomfort. As a result, many studies have begun to develop non-detectable bed monitoring techniques, so that subjects do not have to rely on professional sleep monitoring equipment to get physiological signals and detect sleep quality. This paper use PIR sensors as base sensing tool for tracking human in surveillance systems. From the experiments, we proved that the PIR sensor-based detection algorithm for out of bed performs very well when the sensor is deployed correctly. When the caregiver is unable to take care of the elderly, the most needed information is whether the old man leaves the bed at night. The elderly need to be watched closely to make sure they are safely back in bed, lest they have fallen. Therefore, this article uses the wear-free device to automatically collect data on sleep activities of the elderly while respecting the privacy of the elderly. The cloud platform can remote monitor the sleep activity status of the elderly, and immediately provide information to the nursing staff's tablet device. The Microservices architecture model is a better choice for complex and sustainable development applications. In this paper, the development and implementation of the cloud platform adopts the method of Microservices architecture development to fully decompose applications to facilitate application of agile development and deployment. By dividing a single application into a set of smaller interconnect services, individual services can be developed faster and easier to understand and maintain. Because standardized behavioral data can be integrated, reused, tracked, and further analyzed, avoiding data silos problem. So this paper uses the latest generation of international learning standards: Experience API (xAPI) to record the daily sleep activities of the elderly. The xAPI is a technical specification for storing and accessing learning experiences, and is particularly adept at capturing the nuances of how people participate in activities. The process of daily sleep for the elderly can be fully stored in the learning record repository to provide further analysis and is highly valuable for other follow-up studies.