Context-Aware, Accurate, and Real Time Fall Detection System for Elderly People

Abstract

With all these smart devices around us, and the spread of the Internet and emerging technologies, our life became easier and busier. The un-controlled excessive use of the mobile applications and social media made many of the people around the world feel they are alone, and socializing with electronic devices rather than human beings. This problem is more sever when it comes to the elderly people who are already in excessive need for healthcare and socializing. In this research, we present a Context-aware and private real-time fall detection system (FDS) for elderly people. The proposed system consists of sensor pad placed under a carpet; the electronics reads walking activity to provide an automated health monitoring and alert system. We extended the functionalities of the smart carpet to improve its ability to detect falls, alert health care personnel, by providing a real-time, long-term access, and monitoring event system. The sensed data is stored in the cloud, and analyzed using data mining tools to enhance the system accuracy in detecting falls and monitor elderly people activities. The user can obtain these results and make decisions by accessing the cloud through his/her mobile devices and in real time manner. Results showed that our system detects falls using computational intelligence techniques with 96.2% accuracy, 95% sensitivity and 85% Specificity.