Real-Time Performance, Detection Accuracy, and Response Latency of IoT-Based Fall Detection Systems for Elderly Healthcare: A Systematic Review

This systematic review aims to comprehensively identify, appraise, and synthesize the existing evidence on the real-time performance, detection accuracy, and response latency of Internet of Things (IoT)-based fall detection systems designed for elderly healthcare settings. Falls represent one of the leading causes of morbidity and mortality among older adults, and IoT-based fall detection systems have emerged as a promising technological solution to enable rapid alert and response. However, the real-time performance characteristics of these systems — including detection accuracy (sensitivity, specificity, F1-score), response latency (time-to-alert), and computational efficiency — have not been systematically evaluated across the literature. This review will search four major databases (PubMed/MEDLINE, IEEE Xplore, Scopus, and Web of Science) to identify relevant studies published from 2014 to 2026. The review will follow the PRISMA 2020 guidelines and will categorize findings by sensor type (wearable, ambient, hybrid), system architecture (edge computing, cloud computing, fog computing), detection algorithm (threshold-based, machine learning, deep learning), and deployment setting (home care, nursing home, hospital, community). The protocol has been registered on OSF prior to the screening stage to ensure transparency and reduce the risk of reporting bias.