IoT-Based System for Real-Time Fall Risk Assessment and Health Monitoring
DOI:
https://doi.org/10.37256/jeee.3220245472Keywords:
real-time fall risk assessment, health monitoring system, IoT devices, random forest algorithm, sensor data analysisAbstract
Real-time fall risk assessment and continuous health monitoring are critical components in enhancing elderly care and preventing fall-related injuries. This study presents an IoT-based system designed to provide real-time fall risk assessment and monitor health parameters using wearable sensors. The system integrates the MPU6050 sensor with IoT technology for efficient data collection and analysis. A Random Forest algorithm is employed to process the complex health data, offering precise and reliable fall detection models. The algorithm demonstrates high sensitivity, precision, and accuracy, making it well-suited for processing sensor data to detect falls. The study identifies the waist as the optimal sensor placement, achieving up to 97.9% accuracy, 95.0% precision, and 95.4% sensitivity in detecting mild falls while standing. The wrist sensor performs well in detecting sudden falls, while the leg sensor shows lower accuracy, highlighting challenges in identifying specific fall types. Model validation with Support Vector Machine (SVM) and Random Forest (RF) reveals that the RF model outperforms the SVM, confirming its superiority for fall detection tasks. The system's adaptability and potential for personalized risk assessment promise significant improvements in fall prevention strategies. These findings highlight potential applications that go beyond elderly care, involving at-risk individuals in future research, including those with neurological disorders, sports injuries, or disabilities.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Sona K S, et al.
This work is licensed under a Creative Commons Attribution 4.0 International License.