Performance, Challenges, and Limitations in Multimodal Fall Detection Systems: A Review
Vasileios-Rafail Xefteris, Athina Tsanousa, Γεώργιος Μεδίτσκος, Stefanos Vrochidis, Ioannis Kompatsiaris
Abstract
Fall events among older adults are a serious concern, having an impact on their health and well-being. The development of the Internet of Things (IoT) over the last years has led to the emergence of systems able to track abnormal body movements and falls, thus facilitating fall detection and in some cases prevention. Fusing information from multiple unrelated sources is one of the recent trends in healthcare systems. This work aims to provide a survey of recent methods and trends of multisensor data fusion in fall detection systems and discuss their performance, challenges, and limitations. The paper highlights the benefits of developing multimodal systems for fall detection compared to single-sensor approaches, categorizes the different methods applied to this field, and discusses issues and trends for future work.