Development of an Assistance Robot for Fall Detection and Reporting in Healthcare

Area of Science:

• Gerontology
• Robotics
• Health Informatics

Background:

• Falls are a significant health risk for individuals over 65, leading to severe injuries and increased healthcare costs.
• Existing fall detection systems often lack integration with clinical workflows, hindering timely intervention.
• Robotic assistance offers a novel approach to proactive health monitoring in elderly care settings.

Purpose of the Study:

• To evaluate the tēmi robot's efficacy in detecting falls among elderly residents in simulated care facilities.
• To assess the integration of tēmi robot fall detection data into a simulated clinical workplace system using Fast Healthcare Interoperability Resources (FHIR).
• To identify areas for improvement in the robotic system for real-world deployment.

Main Methods:

• Development of a tēmi robot prototype equipped with the YOLOv8 image recognition model for automated fall detection.
• Implementation of a simulated clinical workplace system to receive and process incident data.
• Utilizing Fast Healthcare Interoperability Resources (FHIR) standards for seamless data exchange between the robot and the clinical system.
• Conducting initial tests to validate the fall detection and data transmission capabilities.

Main Results:

• The tēmi robot prototype successfully detected fallen individuals during simulated patrols.
• Incident data was effectively transmitted to the simulated clinical system via FHIR.
• Initial test results demonstrated the feasibility of the robotic fall detection system.
• The study identified a need for enhanced image recognition accuracy for robust real-world application.

Conclusions:

• The tēmi robot shows potential as a tool for fall detection in elderly care facilities.
• Integration with clinical systems via FHIR is achievable, facilitating efficient data management.
• Further development is necessary to improve image recognition accuracy for reliable deployment.
• Robotic fall detection systems can enhance patient safety and support clinical workflows.