Towards a social and context-aware multi-sensor fall detection and risk assessment platform

Area of Science:

• Gerontology
• Biomedical Engineering
• Computer Science

Background:

• Falls in elderly individuals can lead to severe autonomy loss.
• Existing fall detection systems lack integration and suffer from inaccuracies (undetected falls/false alarms).

Purpose of the Study:

• To present a social- and context-aware multi-sensor platform for improved elderly fall detection.
• To enhance accuracy, reduce false alarms, and enable timely, appropriate caregiver notification.

Main Methods:

• Developed a cloud-based solution integrating multiple fall detection systems and sensors.
• Utilized an ontology to model patient situations and caregiver information (static and dynamic).
• Implemented automatic, continuous fall risk assessment and caregiver notification logic.

Main Results:

• The integrated platform demonstrated improved accuracy in fall detection.
• Reduced instances of false alarms compared to standalone systems.
• Enabled automated and context-specific caregiver alerts based on patient needs and caregiver availability.

Conclusions:

• The proposed platform offers a flexible and reliable solution for elderly fall detection by integrating diverse sensors.
• Contextual information significantly enhances the accuracy and efficiency of fall detection and response.
• The system facilitates better management of elderly care through intelligent monitoring and notification.