Fall detection of elderly through floor vibrations and sound

Area of Science:

• Gerontology
• Biomedical Engineering
• Signal Processing

Background:

• Falls are a significant health concern for adults aged 65 and older, with approximately one in three experiencing a fall annually.
• Serious injuries result from nearly 30% of falls, highlighting the critical need for timely medical intervention.
• Effective fall detection systems are crucial for improving patient outcomes by reducing response and rescue times.

Purpose of the Study:

• To develop a unique, inexpensive, and non-wearable fall detection system for the elderly.
• To accurately discriminate between human falls and other events using sensor data.
• To provide a reliable solution for immediate fall detection to facilitate prompt medical support.

Main Methods:

• The system utilizes floor vibration and acoustic sensing to capture environmental data.
• A pattern recognition algorithm is employed to analyze sensor data and identify fall events.
• The system is designed to be non-intrusive, requiring no wearable devices for the user.

Main Results:

• The developed system demonstrates high accuracy in detecting human falls.
• Achieved a sensitivity of 95%, indicating effective detection of actual fall events.
• Achieved a specificity of 95%, indicating a low rate of false alarms for non-fall events.

Conclusions:

• The proposed floor vibration and acoustic sensing system offers a promising solution for elderly fall detection.
• The system's non-wearable nature and high accuracy make it a practical and effective tool for home monitoring.
• Immediate and reliable fall detection can significantly improve medical response and patient outcomes.