Key points are not available for this paper at this time.
Social isolation has been identified as a major risk in elderly people living alone because of their association with cognitive decline, depression, and other mental health-related issues. Ambient Assisted Living (AAL) is identified as a key technology to facilitate independent living and maintain social connnectedness between elderly, their families, and caregivers. AAL combines Internet of Things, Smart Homes, and machine learning to produce a smart solution that encourages independent, safe, and socially active life for elderly people within their own home. In this article, we propose, develop, implement, and validate a novel Internet-of-Things-based solution that uses passive (i.e., non-obstructive methods) sensing for real-time monitoring of elderly in their homes. The significance of the proposed solution is in the use of machine learning and statistical models to automatically build a personalised model by learning the normal behavioural pattern for the person from deployed sensors in the house. It then uses this model to detect significant changes in the behavioural pattern, should they occur, that could be a consequence of possible health deterioration. We evaluate the performance of the proposed solution via real-world in-home trials installed in six elderly people’s home for a period from 1.5 to 4 months. A discussion and analysis of the in-home trial outcomes and feedback from elderly who participated in the trials conclude the article.
Forkan et al. (Mon,) studied this question.