LITERATURE REVIEW: DESIGN OF INTERNET OF HEALTH THINGS (IOHT) MODEL FOR FALL RISK DETECTION IN ELDERLY AT HOME

Abstract

Background: Fall is a serious health issue among the elderly population with various contributing internal, environmental, and balancing activity-related risk factors. Internet of Health Things (IOHT) has a great potential to improve real-time elderly health monitoring and enable early detection of falls through risk-based intervention. Purpose: To know design an IOHT-based fall risk detection model for the elderly at home utilizing appropriate sensors and machine learning algorithms. Methods: A literature review was conducted to explore recent fall detection studies using motion, physiological, and environmental sensors in an IoT/IOHT-based system. Key findings were extracted and categorized based on sensor types and fall detection approaches. Results: Several motion sensors (accelerometer, gyroscope), physiological sensors (plantar pressure, inertial sensors), and environmental sensors (ultrasonic, sound) have been applied individually or in combination for falls risk prediction and detection among the elderly. Deep learning-based models have shown promising performance in identifying fall risks using multi-parameter sensor data. Conclusions: An IOHT model integrating various sensors shows potential for comprehensive fall risk monitoring and early intervention for the elderly at home. However, further developments in hardware, algorithms, clinical validation, and privacy/security are still needed to maximize the benefits of IOHT-enabled elderly healthcare.