Published in: Journal of Computer Science and Engineering in Innovations and Research (ISSN No: 3049-1762 online)
Publication Date: May 15, 2025
Aging elevates the risk for falls, and consequently, resultant serious health concerns like fractures and loss of mobility. Fall prevention and detection systems reduce these risks by employing the latest technology in detecting falls ahead of time and notifying caregivers. This project implements a sensor-driven, machine learning-based system for improving elderly safety.
The system integrates threshold-based and machine learning-based detection for enhanced accuracy. When it detects a fall, an IoT-enabled alert instantly informs caregivers. It also evaluates movement patterns to issue early warnings, avoiding falls. The model is validated on real-world and simulated datasets, with high accuracy, sensitivity, and specificity. In comparison to current solutions, our system is economical, transportable, and energy-efficient, and thus well-suited for home and healthcare applications. Machine learning enhances flexibility with fewer false alarms. This work contributes to ambient assisted living with future improvements being in the form of complex sensor fusion and deep learning for improving predictions.
Fall Detection, Elderly Care, Machine Learning, Raspberry Pi, IoT, Real-Time Monitoring, Fall Prevention