SALON: Simplified Sensing System for Activity of Daily Living in Ordinary Home

Area of Science:

• Gerontology
• Computer Science
• Engineering

Background:

• Aging populations necessitate monitoring systems for independent living.
• Existing activity of daily living (ADL) sensing systems often lack practicality for home use.
• Key requirements for home-based ADL systems include low cost, robustness, privacy, low maintenance, and ease of use.

Purpose of the Study:

• To propose and evaluate a novel, senior-friendly ADL sensing system.
• To address the limitations of current ADL monitoring systems in real-world home environments.
• To develop a system that is inexpensive, robust, privacy-preserving, maintenance-free, and user-friendly.

Main Methods:

• Development of an ADL sensing system using inexpensive energy harvesting sensors and simple annotation buttons.
• Non-invasive data collection without cameras or microphones.
• Deployment in ten typical elderly homes for a two-month data collection period.
• Data visualization and activity recognition using a long short-term memory (LSTM) model.

Main Results:

• The proposed system successfully collected ADL data from elderly residents in their homes.
• Activity recognition achieved a high average recall rate of 72.3%.
• The system demonstrated effectiveness in a real-world setting, proving its practicality.

Conclusions:

• The developed system is inexpensive and non-invasive, meeting key requirements for home-based elderly monitoring.
• The system accurately collects ADL data and performs activity recognition with high potential for elder care services.
• This technology offers a promising solution for supporting independent and healthy lifestyles for the growing elderly population.