Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

11-2018

Abstract

Along with the upward trend in population ageing is the increasing proportion of the elderly population living alone in the community. This group is especially vulnerable as the onset of various physical, social and mental health issues may be more likely and may go undetected. However, smart homes enabled with elderly monitoring and care systems (EMCS) can now be used to alert caregivers of anomalies in the daily living patterns of the elderly. In this study, we focus on the sleep quality as the key living pattern, as it has been shown that poor sleep quality can lead to health issues. To ensure data collection while preserving their living patterns, we have deployed the EMCS comprising passive and unobtrusive sensors in more than 90 homes of elderly living alone in Singapore. We have built a binary classification model based on Random Forests using the sensor data collected and the subjective PSQI scores obtained from surveys as the ground truth. From the latter, the elderly’s sleep qualities for each survey were divided into 2 groups, representing good and poor sleep qualities. Our model, based on data from 39 participants, achieved 84% classification accuracy, and holds promise for improved accuracy with additional data points.

Keywords

Health and social care, Sleep Quality, Smart Home, IoT

Discipline

Health Information Technology | Software Engineering

Publication

Goodtechs '18: Proceedings of the 4th EAI International Conference on Smart Objects and Technologies for Social Good, Bologna, Italy, November 28 - 30

First Page

94

Last Page

99

ISBN

9781450365819

Identifier

10.1145/3284869.3284894

Publisher

ACM

City or Country

New York

Additional URL

https://doi.org/10.1145/3284869.3284894

Download

Share

COinS