Design and evaluation of a fall detection algorithm on mobile phone platform
Abstract
The increasingly aging population will pose a severe burden to the health services. Falls are a major health risk that diminishes the quality of life among the elderly people and increases the health services cost. Reliable fall detection and notification is essential to improve the post-fall medical outcome which is largely dependent upon the response and rescue time. In this paper, we analyze mobile phones as a platform for developing a fall detection system. The feasibility of such platform is assessed by running an acceleration based fall detection algorithm on the phone. The algorithm was implemented for the Android OS and tested on several HTC models, which included a MEMS accelerometer. Extensive simulations of fall events as well as activities of daily life were conducted on a lab environment to evaluate the system performance. Experimental results of our system, which we still consider as work in progress, are encouraging making us optimistic regarding the feas ibility of a highly reliable phone-based fall detector.
