Automated Cognitive Health Assessment Using Smart Home Monitoring of Complex Tasks.
Automated Cognitive Health Assessment Using Smart Home Monitoring of Complex Tasks.
复制标题
DOI：
10.1109/tsmc.2013.2252338
复制
发表时间：
2013-11
期刊：
影响因子：
--
通讯作者：
Schmitter-Edgecombe M
中科院分区：
文献类型：
--
作者：
Dawadi PN;Cook DJ;Schmitter-Edgecombe M
智能系统可以提供的许多服务之一是对居民福祉的自动评估。我们假设，个人的功能健康，或个人在没有帮助的情况下独立进行活动的能力，可以通过使用智能家居技术跟踪他们的活动来估计。在本文中，我们介绍了一种基于机器学习的方法来评估智能家居中的活动质量。为了验证我们的方法，我们量化了179名志愿者参与者的活动质量，他们在我们的智能家居公寓中进行了一系列复杂的交织活动。我们观察到一个统计学上显着的相关性（r=0.79）之间的自动评估任务质量和直接观察分数。使用机器学习技术根据任务质量预测参与者的认知健康，AUC值为0.64。我们认为，这种能力是了解个人在其家庭环境中的日常功能健康的重要一步。
One of the many services that intelligent systems can provide is the automated assessment of resident well-being. We hypothesize that the functional health of individuals, or ability of individuals to perform activities independently without assistance, can be estimated by tracking their activities using smart home technologies. In this paper, we introduce a machine learning-based method for assessing activity quality in smart homes. To validate our approach we quantify activity quality for 179 volunteer participants who performed a complex, interweaved set of activities in our smart home apartment. We observed a statistically significant correlation (r=0.79) between automated assessment of task quality and direct observation scores. Using machine learning techniques to predict the cognitive health of the participants based on task quality is accomplished with an AUC value of 0.64. We believe that this capability is an important step in understanding everyday functional health of individuals in their home environments.