Automated In-Home Fall Risk Assessment and Detection Sensor System for Elders

Area of Science:

• Gerontology
• Biomedical Engineering
• Health Informatics

Background:

• Falls pose a significant threat to the elderly, causing injuries, disabilities, and mortality.
• Unobtrusive, continuous monitoring systems are needed for effective fall prevention and care in older adults.

Purpose of the Study:

• To develop and evaluate an in-home sensor system for continuous fall risk monitoring and fall detection in older adults.
• To assess the system's potential to revolutionize fall prevention and enhance elder care.

Main Methods:

• Developed an in-home system using pulse-Doppler radar, Microsoft Kinect, and web cameras.
• Collected data over 2 years in 19 senior living apartments, including simulated falls and standardized fall risk assessments (FRAs).
• Utilized gait parameters measured on a GAITRite mat as ground truth for algorithm development and comparison.

Main Results:

• High correlation (p < .01) between FRAs and Kinect-derived gait velocity and stride length.
• Significant correlation (p < .05 to p < .01) between radar velocity and FRAs.
• Real-time fall alerts are successfully transmitted to clinicians, enabling faster emergency responses.

Conclusions:

• The developed in-home fall risk assessment and detection system shows promise for enabling older adults to maintain independence.
• The system has the potential to improve functional ability and support longer independent living at home.