Investigating an Intelligent System to Monitor \& Explain Abnormal Activity Patterns of Older Adults

TL;DR

This paper tackles the low adoption of older adult care technologies by foregrounding human-centered design and explainable interaction. It combines a caregiver-focused focus group, a high-fidelity prototype with wireless motion-sensor monitoring and an interpretable dialogue system, and a qualitative study with professional caregivers and older adults to assess usefulness, trust, and control. Key findings show potential for faster, personalized care and reduced caregiver burden, while highlighting challenges in system performance, interaction modes, and policy barriers. The work offers design guidance across explainability, benchmark data, multimodal sensing, omnipresent interactions, and intuitive tutorials to enhance the practicality and equity of eldercare technologies.

Abstract

Despite the growing potential of older adult care technologies, the adoption of these technologies remains challenging. In this work, we conducted a focus-group session with family caregivers to scope designs of the older adult care technology. We then developed a high-fidelity prototype and conducted its qualitative study with professional caregivers and older adults to understand their perspectives on the system functionalities. This system monitors abnormal activity patterns of older adults using wireless motion sensors and machine learning models and supports interactive dialogue responses to explain abnormal activity patterns of older adults to caregivers and allow older adults proactively sharing their status with caregivers for an adequate intervention. Both older adults and professional caregivers appreciated that our system can provide a faster, personalized service while proactively controlling what information is to be shared through interactive dialogue responses. We further discuss other considerations to realize older adult technology in practice.

Figures (3)

• Figure 1: High-level flow diagram of an intelligent system for older adult care
• Figure 2: Interface of the System: (a) Notification about an abnormal event, (b) Dialogue responses to explain an abnormal event for caregivers, and (c) Dialogue responses for older adults, in which they will receive a follow-up question by the caregiver and determine what information to be shared.
• Figure 3: State machine to generate dialogue responses: When a caregiver or an older adult initiates the interface, the system will generate the greeting message at the "Init" state. If a user requests to explain a detected activity or abnormal event, contextual information (e.g. time of a recognized activity, contextual features that contribute to an abnormal event) at the "Explain" state. In addition, when a caregiver requests to elicit missing information, the system will be at the "Store Request" state and store a corresponding request. When an older adult either takes a rest or is being idle, the system will prompt a question/request to elicit additional information at the "Prompt to Confirm" state.