review · IEEE Internet of Things Journal
Fall detection systems are vital tools for identifying falls quickly and securing rapid assistance, thereby limiting severe injuries and medical complications. Linking these systems with the Internet of Things, especially the Internet of Medical Things, provides new opportunities to support personal safety and healthcare monitoring. This review examines wearable sensor technologies used for fall detection, focusing on the protection of older adults and vulnerable individuals who are susceptible to falling. It organises existing detection methods according to their computational approaches, specifically covering threshold-based algorithms, conventional machine learning techniques, and deep learning architectures. Furthermore, the review outlines and assesses the standard reference datasets currently available to benchmark and evaluate the performance of these different detection methods, offering a structured foundation to assist ongoing research and technical advancement in the sector.
Falls present major health hazards, particularly for older adults and individuals with specific medical conditions. Deploying connected wearable sensors allows for immediate detection and prompt medical response, reducing the risk of lasting injury. Understanding the computational methods and available testing data helps guide the creation of more reliable personal safety devices for vulnerable populations.
The reviewed technologies support the development of connected health products and wearable medical alerts tailored for elderly care and patient safety. Potential users include care homes, healthcare providers, and individual consumers. Because this work is a review categorising existing algorithms and datasets rather than presenting a finished product, the underlying solutions vary widely in maturity, ranging from early-stage algorithmic research to applied healthcare tools.
AI-generated from the published abstract. Always read the original work before citing.
Fall detection (FD) systems are crucial for identifying falls and ensuring timely assistance, thus reducing the risk of serious injuries. With the development of society and increasing attention to health issues, researchers have conducted extensive studies on falls to reduce the severe sequelae of falls. Integrating FD systems with the Internet of Things (IoT), particularly the Internet of Medical Things (IoMT), has significantly advanced healthcare and personal safety. This dynamic relationship between FD technology and IoT has opened up new vistas for monitoring and assisting individuals, particularly the elderly and those with health conditions that make them prone to falls. This article presents a review of wearable sensor-based FD techniques. We classify the detection methods into their categories from an algorithmic perspective: threshold-based, conventional machine learning-based, and deep learning-based methods. In addition, we identify and summarize the available data sets that can be used to evaluate the performance of the introduced methods. This review aims to provide researchers with a better comprehension of the FD problem, intending to foster further advancements in the field.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/jiot.2024.3421336
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
