Home Health System Deployment Experience for Geriatric Care Remote Monitoring CCS: Human-centered computing Empirical studies in HCI CCS: Applied computing Health care information systems CCS: Security and privacy Privacy protections

Dong Yoon Lee, Alyssa Weakley, Hui Wei, Daniel Cardona, Shijia Pan · 2026

Read Home Health System Deployment Experience for Geriatric Care Remote Monitoring CCS: Human-centered computing Empirical studies in HCI CCS: Applied computing Health care information systems CCS: Security and privacy Privacy protections by Dong Yoon Lee, Alyssa Weakley, Hui Wei, Daniel Cardona, Shijia Pan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-17608.txt.

Introduction

Dong Yoon Lee and four co-authors report field experience with a home health system for older adults. Adult children often care for aging parents from afar. They want remote tools that are plug-and-play. Privacy must hold while activity is tracked all day. The system aims to give real-time, clear, actionable updates.

Deployment experience shows where lab designs meet real homes. Sensors, networks, and family habits all test the plan. Lessons from live use guide the next design round. Aging in place is a growing need worldwide. Tools that families trust can delay costly care moves. The report grounds design claims in homes, not demos.

Catalog record

CREATORDong Yoon Lee, Alyssa Weakley, Hui Wei, Daniel Cardona, Shijia Pan
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-17608

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Home Health System Deployment Experience for Geriatric Care Remote Monitoring CCS: Human-centered computing Empirical studies in HCI CCS: Applied computing Health care information systems CCS: Security and privacy Privacy protections Dong Yoon Lee, Alyssa Weakley, Hui Wei, Daniel Cardona, Shijia Pan Abstract . To support aging-in-place, adult children often provide care to their aging parents from a distance. These informal caregivers desire plug-and-play remote care solutions for privacy-preserving continuous monitoring that enabling real-time activity monitoring and intuitive, actionable information. This short paper presents insights from three iterations of deployment experience for remote monitoring system and the iterative improvement in hardware, modeling, and user interface guided by the Geriatric 4Ms framework (matters most, mentation, mobility, and medication). An LLM-assisted solution is developed to balance user experience (privacy-preserving, plug-and-play) and system performance. 1. Introduction The United States faces a critical caregiver shortage, a challenge exacerbated by the rapidly growing older population. In 2020, 55.8 million individuals (17% of the population) were aged 65 or older ( 6 ) , a number projected to rise to 85.7 million (23%) by 2050 ( 35 ) . Aging is associated with a rise in health conditions requiring medical management and a decline in cognitive and physical conditions which threatens a person to live independently. To assist older adults age-in-place, friends and family, mainly adult children, often take on caregiving roles from a distance. Family members who take on informal caregiving roles contribute an estimated $600 billion annually in unpaid support ( 29 ) . These informal caregivers desire digital tools to assist them in remote care delivery, monitoring, and actionable insights. Technology advancements in the Internet of Things (IoTs) and artificial intelligence (AI) have the potential to improve informal caregiving as well as older adult quality-of-life by enabling real-time activity monitoring ( 40 ) , just-in-time support ( 26 ) , and actionable information ( 21 ) for caregiver and clinical support teams. For example, prior work has demonstrated the feasibility of capturing and recognizing activities under the lab environment with healthy young adults using sensors such as mmWave ( 33 ) , vibration ( 14 ) , PIR ( 15 ) , and wearables ( 19 ) . Corresponding machine learning algorithms have been developed to predict target information for different people in varying environments, mitigating the impact of domain variances ( 9 ; 32 ) . However, challenges in real-world deployments with older adults and their caregivers remain largely unaddressed. For example, limited domain knowledge about sensor systems and their configurations across diverse deployment environments may hinder families and caregivers from effectively using home-based monitoring systems. Suboptimal deployment and configuration can result in insufficient or low-quality data, reducing the model’s ability to profile users accurately. Declining memory and other aspects of cognition may impact the usability and feasibility of adopting technology. Without bridging the gaps among hardware, modeling, and end users, these challenges will continue to impede the practical adoption of remote monitoring technologies for geriatric care at home. To fill in these gaps, we present a plug-and-play long-term continuous monitoring solution with ambient vibration-based sensing system. We adapt multiple aspects of the system based on the Geriatrics 4Ms Framework – (what) matters most, mentation (cognition and mood), mobility, medication – to systematically reflect the needs of older adults and their caregivers. We identify and tackle challenges in hardware (Section 2.1 ), (data) modeling and inference (Section 2.2 ), and user interface (Section 2.3 ), to enable (1) plug-and-play deployment, (2) robust on-edge inference, and (3)user-friendly experience for older adults with cognitive decline and their family caregivers. First, we develop product-grade sensing device hardware and customize network protocol and configuration based on the individual home deployment to enable plug-and-play deployment for end users. We conduct semi-controlled experiment in a physical simulation suite, which is equipped with surveillance video and microphone for ground truth collection. In this deployment, we generate a dataset with multiple participants and second-level ground truth label. The unexpected network issues, including congestion, limiting access to the Internet, and unknown sensing device status, are addressed with updated network design before the in-home uncontrolled data collection. Second, we establish data collection protocol to ensure sensing quality and efficient deployment that are contextual relevant. To ensure data security and privacy, we develop robust activity recognition models that can be deployed on edge devices. A temporal convolutional network (TCN) model for vibration signals is fine-tuned from a pre-trained model originally trained on online acoustic data. Finally, we develop an LLM-assisted interactive deployment recommendation system to facilitate intuitive deployment and usage for non-expert end users. LLM’s broad world knowledge and strong natural language understanding capabilities allows effective reasoning for deployment configuration recommendation. However, technical gaps remain in enabling LLMs to comprehend the spatial aspects of IoT deployments and the implicit requirements of users (e.g., cognitive decline), which are areas that prior work has largely overlooked. In particular, current approaches lack methods for generating consistent and physically feasible recommendations that balance system performance with user experience . We tackle this by designing schema templates to supply the LLM with a structured representation of the sensing system, environment, and user preferences. Then we use instruction prompting to turn the user preferences into a recommendation for the sensor configuration. In summary, the contributions of our paper are as follows: (1) We present a home health monitoring system that is customized for older adults and their caregivers based on Geriatric 4Ms framework. (2) We address the hardware challenges for plug-and-play real-world deployments, including a custom fault-tolerant protocol and network design. (3) We establish the deployment protocol to ensure sensing quality and data relevancy, and develop robust learning models for activity recognition that can be deployed on edge devices. (4) We design an LLM-powered recommendation system, outputting a user-tailored deployment protocol that balances system performance and user experience. (5) We conduct multiple rounds of system evaluation and user study with real-world deployments to demonstrate our system’s effectiveness. 2. Geriatrics 4Ms Framework Guided Home Health System Design The Geriatrics 4Ms framework – (what) matters most, mentation, mobility, and medications ( 7 ; 25 ) – was developed as a guideline for the design of care or services for older adults. We leverage the 4Ms framework to adjust the system and deployment designs. 2.1. Plug-and-Play Hardware and Network Design The goal of hardware design is stable and scalable data collection in home environments. This section introduces the sensing device and network designs to achieve this goal and how the 4Ms framework guides our design. Figure 1 . Plug-and-play system for home health monitoring. (a) depicts the plug-and-play sensing device connected to a vibration sensor. (b) and (c) show the system working in a real-world deployment. (d) presents the system design for LLM-assisted deployment, where the blue arrows indicate sensing data and the orange arrows show deployment data. (a) Deployment 1: Simulation Suite. (b) Deployment 2: Participant #1. (c) Deployment 3: Participant #2. Figure 2 . Three iterations of deployments. Brief text-only description for accessibility (required by ACM). 2.1.1. Plug-and-Play Device Design The key challenge for the sensing device design is to be as discreet as possible due to the following reasons. First, many older adults do not want their visitors to notice

Read the complete text