JMIR Publications

ABSTRACT

Background:

Racial/ethnic minority older adults with low socioeconomic status are at an elevated risk of developing dementia, but resources for assessing functional decline and early detection are limited. Cognitive impairment affects the ability of the person to perform and complete daily activities and mobility behaviors. To overcome challenges of traditional assessment methods, smart home technologies (SmHT) have emerged as a tool to provide objective, high-frequency, and remote monitoring. However, these technologies are primarily based on motion detection sensors that do not have the capability to label specific activity types. Furthermore, this group of older adults often has inadequate access to activity sensing technology due to limited social and financial capital and lack of technology experience. There is a need to develop new sensing technology that can characterize and quantify different patterns of in-home daily activities while also being discreet, affordable, and requiring minimal user engagement.

Objective:

The goal of this study is to develop a novel Channel State information (CSI)-based device-free, low-cost WiFi sensing system and associated machine learning (ML) algorithms for localizing and recognizing different patterns of in-home activities and mobility in older adult residents of low-income housing with and without mild cognitive impairment (MCI).

Methods:

This feasibility study will be conducted over a year in collaboration with an interprofessional wellness care group, Richmond Health and Wellness Program, that serves the healthy aging needs among older adult residents of low-income housing. We will collect CSI data from several activity scenarios (e.g., sitting, walking, preparing meals, etc.) by using the proposed WiFi sensing system continuously over a week in homes of low-income housing residents. These activities will be videotaped for generating ground truth annotations for testing the accuracy of the ML algorithms derived from the proposed system. We will examine the association of our WiFi sensing-based activity and mobility metrics with self-reported physical function and psychosocial measures and compare the levels of WiFi sensing-based measures between MCI and Non-MCI groups. Using qualitative individual interviews, we will explore the acceptability of the WiFi sensing system and implementation barriers in the low-income housing setting.

Results:

Development of the WiFi sensing system began in November 2022, with participant recruitment starting in summer 2023. We anticipate that preliminary results will be available in summer 2024.

Conclusions:

This study can make a valuable contribution to SmHT science and ML capabilities for early detection of cognitive decline particularly among socially vulnerable older adults, because sensing devices are not readily available to this population due to cost and information barriers. Our passive sensing device has the potential to identify individuals at risk for cognitive decline by assessing the level of physical function by tracking their in-home activities and mobility behaviors. Clinical Trial: Not applicable